Classifications

• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
  • A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
  • A61B5/14551—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters for measuring blood gases
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
  • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
  • A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
  • A61B5/021—Measuring pressure in heart or blood vessels
  • A61B5/02108—Measuring pressure in heart or blood vessels from analysis of pulse wave characteristics
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
  • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
  • A61B5/02416—Detecting, measuring or recording pulse rate or heart rate using photoplethysmograph signals, e.g. generated by infrared radiation
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
  • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
  • A61B5/02416—Detecting, measuring or recording pulse rate or heart rate using photoplethysmograph signals, e.g. generated by infrared radiation
  • A61B5/02427—Details of sensor
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
  • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
  • A61B5/02438—Detecting, measuring or recording pulse rate or heart rate with portable devices, e.g. worn by the patient
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
  • A61B5/0816—Measuring devices for examining respiratory frequency
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
  • A61B5/316—Modalities, i.e. specific diagnostic methods
  • A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
  • A61B5/332—Portable devices specially adapted therefor
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
  • A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
  • A61B5/7207—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal of noise induced by motion artifacts
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
  • A61B5/7235—Details of waveform analysis

Definitions

• This invention generally relates to vital signs scanning by a portable device with multiple integrated sensors.
• vital sign measurement devices should be user friendly and simple to use so they can be used daily, such as a shaver, a hair dryer or a microwave oven. Measurements with vital sign measurement devices could become part of a daily routine or used sporadically throughout the day.
• An ubiquitous vital signs measuring device may include advantageous features.
• the ubiquitous vital signs measuring device is a single integrated device that makes a plurality of vital sign measurements in a short time period for user friendliness.
• the vital sign measurements are quick and discrete; instead of taking a few minutes of time to set up and use, being noisy and very explicit about the type of vital sign measurements being performed, such is the case with a blood pressure cuff for example.
• the user interface for the measuring device can show useful user-friendly results to a user.
• the vital sign results from the measurement device are locally stored and can be sent to remote servers in a transparent manner, such as with electronic/personal health records for example.
• vital signs measuring devices It is advantageous for vital signs measuring devices to cover a large range of measurements to monitor as many conditions, or signs, as possible. That way, the vital signs measuring devices can be useful to as many users as possible. Some users monitor a set of pathologies with a precise type of measurements prescribed by medical teams, whereas others monitor their bodies out of curiosity, to better know themselves, or to monitor for trends. Some important measurements include: temperature, heart rate, pulse rate, blood oxygenation, blood pressure (systolic and diastolic), and respiration rate.
• Additional measurements such as recording sounds and motions can be interpreted in many ways including respiration rate, and more generally lung and heart functions.
• accelerometers, gyroscopes and GPS have been the foundation of sports and fitness devices as well as activity monitoring.
• Many individual vital sign measurement devices have been developed that focus on measuring one type of vital sign of a patient. There are often four vital signs that are desirable to measure for a patient. Typically, four individual separate vital sign measurement devices are used to measure these four vital signs such as a thermometer, a pulse-oximeter, a blood pressure cuff, and an electrocardiogram (ECG) monitor. The user typically performs these measurements sequentially over several steps. Sequentially performing all of these measurements may from five to seven minutes when using four independent devices.
• the separate vital sign measurement devices are often large in size and thus are not portable.
• the embodiments of the invention concurrently take at least four vital signs measurements in parallel during the same measurement period using one integrated vital signs scanning device.
• a plurality of sensors are integrated into the vital signs scanner, including a temperature sensor (thermometer), an ECG sensor, dual - red and infrared - photoplethysmographic (PPG) sensors, a sound sensor (microphone), and a motion sensor (3D accelerometers) under the control of a centralized signal processor.
• thermosensor thermometer
• ECG sensor ECG sensor
• PPG dual - red and infrared - photoplethysmographic
• microphone microphone
• a motion sensor 3D accelerometers
• the vital signs scanner of the embodiments of the invention is battery operated and has a compact form factor that allows it to be a portable vital signs scanner.
• the vital signs scanner may be one hundred times more compact than the individual measurement devices that it can replace.
• the compact form factor allows it to readily fit within a user's pocket or purse for example.
• the vital signs scanner of the embodiments of the invention is automated, user friendly, and convenient to use.
• the automation in the vital signs scanner of the embodiments of the invention reduces the time to capture vital signs data and improves the usability of a vital signs scanning device.
• the user of the vital signs scanner of the embodiments of the invention can focus on taking quality measurements at the right times. Automated data processing operations of the vital signs scanner are transparent to the user. There is no need to inflate a cuff.
• the vital signs scanner of the embodiments of the invention has a quick start up time avoiding wait times to boot up or go through a warm up phase.
• the vital signs scanner of the embodiments of the invention uses signal processing techniques to take the vital signs measurements during a measuring session so that inconveniences, such as a blood pressure cuff and a noisy pump to inflate it, are avoided.
• a plurality of vital sign sensors in a unified scanner can provide concurrent capture of a plurality of vital signs measurements including, core temperature, heart rate, pulse rate, blood oxygenation (Sp02), respiration rate, and blood pressure.
• a plurality of vital signs measurements including, core temperature, heart rate, pulse rate, blood oxygenation (Sp02), respiration rate, and blood pressure.
• the sensor fusion may be applied to the signals obtained from individual sensors to enhance the data through redundancy and cross-validation.
• Figure 1 illustrates a user holding the device between index finger and thumb.
• Figure 2 A illustrates a perspective view of an exemplary embodiment of the invention.
• Figure 2B illustrates a functional block diagram of the signal processor used within the vital signs scanner 100.
• Figure 3 A provides a flow chart description of an exemplary process of measuring temperature.
• Figure 3B illustrates a thermometer to patient interaction model.
• Figure 3C-3D illustrates an exemplary temperature physiological model.
• Figure 4A is a flow chart of an exemplary process of measuring a user's ECG.
• Figure 4B illustrates details of an ECG to patient interaction model.
• Figures 4C-4E illustrates exemplary heart rate physiological models.
• Figure 5 A is a flow chart of an exemplary process of measuring the user's pulse rate.
• Figure 5B illustrates an exemplary PPG to patient interaction model.
• Figures 5C-5D illustrates exemplary pulse rate physiological models.
• Figure 6A is a flow chart illustrating an exemplary process of measuring blood oxygenation (Sp02).
• Figures 6B-6C illustrates exemplary blood oxygenation physiological model.
• Figure 7 A is a flow chart illustrating an exemplary process of measuring the user's
• FIGS 7B-7C illustrate differing exemplary pulse width transit time (PWTT) physiological models that may be merged together.
• Figure 8A is a flow chart illustrating an exemplary process of measuring blood pressure.
• Figure 8B illustrates an exemplary blood pressure physiological model.
• Figure 9A is a flow chart illustrating an exemplary ECG/PPG based process of measuring respiration rate.
• Figure 9B illustrates an exemplary respiration rate physiological model.
• Figure 10A illustrates an exemplary process of taking sound recordings.
• Figure 11 A is a flow chart illustrating an exemplary microphone-based process of measuring respiration rate.
• Figure 1 IB illustrates an exemplary microphone to patient interaction model.
• Figure 11C illustrates an exemplary microphone physiological model.
• Figure 12 illustrates an exemplary process of taking accelerometer motion recordings.
• Figure 13A is a flow chart illustrating an exemplary accelerometer based process of measuring respiration rate.
• Figure 13B illustrates an exemplary accelerometer to patient interaction model.
• Figure 13C illustrates an exemplary accelerometer physiological model.
• Figure 14 is a block diagram illustrating noise shared across sensors in temple measurements.
• Figure 15 is a flow chart illustrating chest measurement with the noise shared across sensors in chest measurements.
• Figure 16 is a block diagram illustrating sensory fusion across sensors in temple measurements.
• Figure 17 is a flow chart illustrating the sensory fusion process for sensors in chest measurements.
• Figure 18 illustrates a functional block diagram of electronic circuitry within the exemplary portable wireless vital signs scanner.
• the embodiments of the invention include a method, apparatus and system for a portable vital signs scanner with a plurality of integrated sensors.
• the user holds the vital signs scanning device 100 between the index 111A and thumb 11 IB of a left hand making contact with an ECG electrode 122 under the index finger, the thumb, or both.
• the user pushes the vital signs scanner 100 lightly against the left temple or forehead for a period of time. Doing so, an ECG electrode 123 of the side of the vital signs scanner 100 makes contact with the left temple or forehead of the user.
• a thermometer sensor 125 and a PPG sensor 124 near the electrode 123 are also applied over the temple or forehead of the user.
• One or more accelerometers within the vital signs scanner 100 are attached to a printed circuit board (PCB), which is linked to the housing of the scanning device 100 casing.
• the vital signs scanner 100 also includes a microphone. Accordingly, the vital signs scanner 100 can concurrently capture sounds with the microphone and movement with the accelerometer as vital sign signals are being captured by other sensors.
• positions of fingers/thumb are shown in the figures for explanatory purposes. However, positioning of fingers on the vital signs scanner 100 can vary depending on the implementation.
• the index finger 111 A and the thumb finger 11 IB may be positioned on the sides of the scanner with one or both coupling to an ECG electrode to complete the circuit through the body with the second ECG electrode 123.
• the vital signs scanning device 100 is pushed flat against the chest to capture sounds with a microphone and motion of the chest with one or more accelerometers over another period of time.
• FIG. 2A illustrates exemplary locations of the sensors in accordance with one embodiment of the invention.
• the ECG electrode 122 may be positioned under the index finger 111 A, the thumb 11 IB, or both finger and thumb and squeezed by the user.
• the ECG electrode 123 is positioned against the temple/forehead of the user. With the ECG electrodes making contact, the ECG sensor closes an electric circuit with the users body including a circuit path through hand, arm, and heart to the temple.
• thermometer sensor 125 can directly sense the surface temperature at the temple or forehead of the user.
• PPG sensors 124 shown in Figure 2A are also for sensing at the user's temple.
• the primary advantage of the temple location over, for example, a wrist location is that the construction of the vascular tunics in terms of presence/content of tissue elasticity greatly influences "signal fidelity". For example, readings at the temple are much more representative than readings at the wrist, which includes an entirely greater percentage of elastic tissue in the arterial tunics, etc.
• the temple may also be a good location to take PPG measurements because it is in direct connection to the heart via the temporal artery. Unlike the fingers/thumb being used for PPG measurements of known pulse-oximeters, the blood flow at the temple is only marginally modified by body configurations (straight head, turned head, standing, sitting). With such freedom in body configurations, the temple PPG measurements put fewer constraints on a user in an uncontrolled environment.
• the scanning device 100 senses for vital sign parameters with sensors at some of the most reliable sites on the body while minimizing inconvenience to users.
• the vital signs scanner 100 includes a signal processor
• the signal processor 206 controls the plurality of sensors 210 to concurrently capture data during the measurement period.
• the processor 206 is coupled to a storage device 208, such as memory, to read instructions and execute signal processing algorithms to achieve the methods of generating vital sign measurements as described herein.
• a number of algorithmic models may be further stored within the memory that are used by the signal processing algorithms executed by the processor 206. For each type of measurement performed by the vital signs scanning device 100, enhanced measurement models are used that link three types of partial models together.
• a schematic diagram of the vital signs scanner 100 is shown in Figure 18 and described herein.
• a first partial model is a signal processing model that is performed on the signal generated by a sensor.
• the analog signal generated by the sensor is transformed by an analog to digital conversion into a digital signal.
• the digital signal then undergoes an initial signal processing by a signal processor with a signal processing algorithm to typically remove noise.
• Noise in many optoelectronic devices, such as the PPG photodiodes, is removed before it gets to the signal processor and the signal processing algorithms.
• the photon current from the detector is converted to voltage through a transimpedance amp lifer (e.g., Z Amp)
• the Z Amp noise can be removed with a "chop circuit” that chops out the noise and any float in the op amp offset voltage float.
• the signal processing algorithm may then further perform additional transformations or data processing before finally providing the resulting vital signs measurement value to be displayed to users or stored in healthcare records.
• the second partial model is an interaction model that captures the interaction between the user and the sensor.
• This interaction model may entail hysteresis or inertia for some sensors, such as a temperature sensor.
• the interaction model may include the noise affects that a user may introduce, such as light pollution noise for PPG sensors or vibration noise for accelerometers sensing motion.
• the interaction model may include models of the contact between sensor and the user's body. For example, physical properties of the skin like conductivity, elasticity or color may be something to consider when a sensor makes contact with the user's body.
• signal processing algorithms may embed the interaction model, leverage and correct them to provide improved vital signs measurement results.
• a sensor may be used in a plurality of body sites to obtain different vital signs.
• the interaction model may have different interactive conditions with the vital signs scanner for the different positions of the scanner with the same sensor. Accordingly,
• embodiments of the invention may change to a different interaction model in response to the vital sign being captured and the use of the sensor at different body locations.
• a third partial model is a physiological model.
• the physiological model is linked to the measured vital sign.
• Human physiology has been studied for centuries, worldwide.
• Early medical studies by Ambroise Pare (1510-1590) led from belief to evidence and eventually to the science of medicine, modeling vital functions and generating vital signs as a measure of that function.
• Medical or physiological models are often the building blocks of medical terminology. For example, medical students learn to analyze ECG patterns based on the P, Q, R, S, T phases of an ECG signal model for a human.
• Another commonly known physiological range for heart rate of a human is that heart beat rates vary between 25 beats per minute and 250 beats per minute.
• Another commonly known physiological model for temperature is that core temperatures of a human vary between 28 degrees centigrade (C) and 43 degrees centigrade.
• physiological models may be used by the signal processor and its signal processing algorithms to validate the resulting vital sign measurements.
• a mathematically correct signal generated by a sensor and sent to the signal processor for signal processing may be deemed noise for physiological reasons. For example, a sampling of the body temperature at a frequency of 10 Hertz (Hz) may show some variation in the temperature signal up to 5Hz according to Nyquist-Shannon sampling theorem.
• the physiological models of temperature exclude any temperature variation above 0.015 K/s. This is because a human body under normal conditions, is unable to change temperature faster. Accordingly, variations in the temperature signal at higher rates are likely to be noise that can be eliminated.
• the signal processing provided by the vital signs scanner combines these three models together in the process of generating vital signs values.
• the vital signs scanner and signal processor therein uses signal processing methods to extract the signal mathematically relevant within the sampling rate of the device, according to the Nyquist-Shannon sampling theorem.
• the vital signs scanner and signal processor therein uses a sensor interaction model to perform corrections linked to pollutions of readings such as noise generated at the interaction point due to both the human side of the measurement point and the physical characteristics of the sensor.
• the vital signs scanner and signal processor therein further uses the physiological model of the measurement where the human body parameters remain in a family of possible ranges, shapes, dynamic and bandwidth of signal.
• a valid measurement for a vital sign should comply with these partial models.
• a valid measurement for a vital sign should be mathematically correct regarding signal processing.
• the contact interaction for a valid measurement for a vital sign should be validated and corrected during the duration or length of time of the measurement.
• the measured signal for a valid vital sign measurement should be medically relevant.
• the methods and signal processing algorithms used by the vital signs scanner and its signal processor can introduce corrections at each step, for each partial model, resulting in a high accuracy measurement, for each measurement (temperature, heart rate, pulse rate, blood oxygenation, blood pressure, respiration rate), each of which is performed at some of the best sites on the body (temple, forehead, finger/thumb).
• the methods and signal processing algorithms used by the vital signs scanner and its signal processor may also use a sensor fusion model to leverage multiple measurements and improve the results of one or more different types of measurements through data fusion.
• the vital signs scanner 100 concurrently captures signals from its sensors during the same measurement periods. Accordingly, the signal data for one or more of a plurality of sensors may be used to improve the signal data and measurement results for another sensor and vital sign.
• Some signal pollution e.g., signal noises
• an accelerometer measures the motion and vibrations of the vital signs scanning device, deleting motion artifacts.
• the motion and vibrations of the vital sign scanner can be coupled to the signal processor or fed back to other sensors during signal interpretation, to either validate or invalidate the data, or to make data corrections to improve the signal data.
• thermometer device A301 is pressed against the temple of the user to measure the surface temperature above the temporal artery.
• the output signals from the thermometer is coupled into a signal processing module A302 to signal processing.
• the signal processing module A302 such as by an FFT/IFFT for example, initially removes noise in the output signals from the thermometer sensors to obtain what is expected to be a desired data signal.
• the data signal is further processed by the signal processing module A302 taking into account the thermometer-patient interaction model A305.
• a validation signal is generated by the signal processing module A302 when the sensor readings are stable (see Figures 3B and 3C and description thereof) or have sufficiently low noise, and are within expected ranges. With a valid data signal, further processing can occur.
• the resulting valid data signal is coupled into a model merge module A303.
• the model merge module A303 calculates the value of the core temperature T c according to formula
• T c k(T s - T + T a
• T a is the ambient temperature
• T s is the skin temperature, both of which are measured in parallel by two different thermometer elements.
• the value of k for the core temperature equation is an experimental calibration parameter obtained by quadratic approximation, such as described for example in the "Temperature Assessment via the Temporal Artery: Validation of a New Method", by Marybeth Pompei, published by Exergen Corporation, copyright 1999.
• the core temperature T c is then checked for physiological validity against the temperature physiological model A306 (see additional explanation with reference to Figure 3D).
• the overall temperature result A304 is displayed to a user and/or stored into health records.
• thermometer device A301 An example of positioning of the thermometer device A301 is shown in Figure 3B.
• thermometer device A301 is applied against the temple of the user.
• the thermometer device A301 includes an ambient temperature sensor A301a and a surface temperature sensor A301s, both of which concurrently take measurements in parallel together with the other vital signs.
• the surface temperature sensor A301s is receded in the housing
• the surface temperature sensor A301s does not make direct contact with the skin of the user, instead taking its measurement through a copula of air B302.
• This copula of air B302 offers a small thermic inertia that is taken into account by the thermometer-patient interaction model.
• the air copula B302 to avoid the sensor from directly touching skin, complex interaction models can be avoided.
• the sensor avoids being soiled by skin products, sweat etc. that may be on a user's skin.
• FIG. 3C illustrates an exemplary thermometer-patient interaction model showing the thermometer output signal C303 varying over time.
• the output signal C303 Prior to contacting the user temple at time tO at C301, the output signal C303 is constant at tempi . After time tO, the thermometer signal C303 grows to reach an asymptote C302 of temp2.
• the thermometer-patient model provides a threshold time tm C304 after contact that needs to be exceeded before validating the asymptote C302 to be output temporal temperature once the signal has stabilized.
• One mechanism of such an algorithm measures the relative variation of the signal over time. Thus, this algorithm can also function when the thermometer is removed from the temple of the user.
• Figure 3D illustrates an exemplary temperature physiological model for determining core temperature.
• the chart of the model illustrates lines D301,D302,D303,D304 representing ranges within which the temperature signal for the user is not going to exceed.
• Signal D305 represents a raw temperature signal from the thermometer.
• Living humans can function between twenty-eight degrees centigrade (28°C) (e.g., see line D302) and forty three degrees centigrade (43°C) (e.g., see line D301).
• the temperature physiological model illustrates that medical emergencies are declared for temperatures below 32°C (e.g., see line D304) and temperatures above 41°C (e.g., see line D303) during a scan. Otherwise, the vital signs scanner can assume that the user is not undergoing a medical emergency when it is not being used.
• the calculation of ECG and heart rate vital sign values include a noise model and a heart rate physiology model.
• the noise model provides that peaks lower than a rejection threshold R are to discarded.
• the noise model further provides that peaks that have a duration less than 0.02 seconds are to be discarded.
• the heart rate physiology model includes a valid range of heart rates (e.g., 35 beats per minute to 250 beat per minute). Values outside this range are invalid. It further includes an expected signal frequency range with an heart rate variations being limited to a maximum threshold. Signals outside the expected signal frequency range or having larger variation than the maximum threshold are deemed invalid.
• the ECG sensor A401 closes the electrical circuit between the finger/thumb of the user coupled to the electrode A401f and the temple of the user coupled to the electrode A401t.
• the output signal from the ECG sensor feeds into a finite impulse response (FIR) or fast Fourier transform/inverse fast Fourier transform (FFT/IFFT) filter A402 removing noise from the data.
• FIR finite impulse response
• FFT/IFFT fast Fourier transform/inverse fast Fourier transform
• An ECG-patient interaction model A408 (see also description with reference to Figure 4B) is used to validate the measurements at this stage.
• the intensity of filtered ECG signal is then normalized to a normalized signal A403 that is coupled into an LI trend filter A404.
• the LI trend filter A404 produces trend estimates that are piecewise linear, and therefore is well-suited to analyzing time series with an underlying piecewise linear trend.
• an auto-correlation process A405 by an auto-correlator is performed on the LI trend filtered ECG signal.
• the output signal from the auto-correlation A405 is coupled into the model merge process A406.
• the model merge process A406 merges the ECG physiological model A409 (see additional explanation in Figure 4C) with the signal processing model.
• the ECG physiological model A409 is applied to the reduced noise LI trend filtered ECG signal in order to generate the final heart rate values A407.
• the heart rate values A407 are shown to the user in a user interface on a display device and/or stored into health records.
• FIG. 4B An example of positioning of the ECG electrodes is shown in Figure 4B.
• the user's index finger/thumb presses on electrode A401f.
• the contact between the finger/thumb and the electrode provides a reliable electrical contact.
• the housing of device B401 is applied against the temple of the user, pushing electrode A401t against the temple or forehead. Due to limited hand stability, small motions made during the measurements. Accordingly, the temple electrode A401t may intermittently lose contact with the temple and generate electrical pollution or noise.
• FIG. 4C illustrates an example of ECG signal collected by electrodes B401f and
• the ECG signal can be decomposed into intervals (e.g., P, Q, R, S T intervals, which are not shown), with C401 pointing to a QRS complex.
• intervals e.g., P, Q, R, S T intervals, which are not shown
• C401 pointing to a QRS complex.
• some additional peaks C402 appear due to the above-mentioned pollution/noise introduced by the sensor-user interaction model.
• those spikes may be mistaken for the QRS complex C401, but experience showed that the spikes generated by these pollutions, are limited in time to approximately 0.02 seconds.
• Figure 4E illustrates an exemplary physiological model for determining heart rates.
• Each of the heartbeats is represented by a spike E401.
• the physiological model of the heart rate provides minimum and maximum heartbeats per minute (e.g., 25 to 250 heartbeats per minute for an adult). Another important physiological limitation comes from the heart rate variability, knowing that heart rates cannot vary faster than a given MAXHEARTBEATVARIATION threshold.
• PPG and Pulse Rate use a noise model and a pulse rate physiology model.
• the noise model is for signals greater than 100 Hertz, for example.
• the pulse rate physiology model defines a valid pulse rate range between upper and lower values.
• the pulse rate signals are expected to be within a range of 0.5 Hz to 40 Hz. Pulse rate variations are expected to less than a maximum threshold.
• PPG1 and PPG2 are positioned against the temple of the user to capture a PPG signal.
• One of the photodiode sensors is an infrared (IR) photodiode while the other photodiode sensor is a red photodiode. Only of the signals from the pair of photodiode sensors needs to be used to determine the pulse rate of the user.
• a selection process A503 selects one of the two raw sensor signals from the photo diode sensors to be processed. While only one signal is shown being processed in Figure 5 A, it is understood that both raw sensor signals may be processed in parallel together for redundancy, noise elimination, and selection of the more reliable pulse rate signal and PPG signal.
• the selected raw sensor signal feeds into a FIR or FFT/IFFT filter A504 removing noise from the data.
• the PPG-patient interaction model A511 of Figure 5 A (see also Figure 5B and 5C) is used to validate the measurements in advance at this stage as well. If the signal measurements are invalid, the other photodiode sensor data may be used. If both signal measurements are invalid, the user may be asked to repeat a signal capture process. If valid measurement data is found the signal data, the intensity of this signal is then normalized by a normalization process A505. After signal normalization, the signal is coupled into a first derivative (gradient) filter A506. The output signal from the first derivative (gradient) filter A506 is coupled into an LI trend filter A507.
• the filtered data is then coupled into an auto- correlator A508 to undergo an auto-correlation process A508 to remove noise from the filter signal data.
• the reduced noise signal is then coupled into a merged model A509 that applies the PPG physiological model A512 (see Figure 5D) to the signal processed signal.
• the merged model A509 is a math algorithm representative of the PPG physiological model A512.
• the resulting output from the merged model A509 are pulse rate values A510 of the user.
• the pulse rate values A510 are displayed to the user by a display device and/or stored into health records.
• Figure 5B illustrates optimal positioning of the photodiode sensors, PPG1 (A501) and PPG2 A502.
• the PPG photodiode sensors A501 and A502 When holding the device housing B501 against the temple, the PPG photodiode sensors A501 and A502 are in direct contact with the temple. It is desirable that ambient interfering light A503 from external light sources is blocked by the direct contact. Due to hand instability and other small motions during the measurements, the detector for PPG photodiodes A501 and A502 may intermittently be perturbed by the interfering light A503. This translates into a saturation of the signal.
• FIG. 5C illustrates a chart of a PPG signal captured with the PPG photodiode sensors A501 and A502 over a period of time.
• the photodiode sensors capture a signal with a peak C501 and a peak C503 corresponding to each pulse rate.
• Ambient light perterbation is minimized by the light sheltering design of the housing B501.
• saturation of the PPG signal may occur as shown at C502.
• the PPG-user interaction model accounts for ambient light perterbation by using an algorithm to detect and filter out the interfering light.
• Figure 5D illustrates an exemplary physiological model for measuring pulse rates.
• Each of the pulse beats is represented by a spike D501 corresponding to one of the C501 and C503 peaks.
• the physiological model of the pulse rate provides minimum and maximum pulse beats per minute, typically thirty-five (35) to two hundred fifty (250) beats per minute for an adult. Another important physiological limitation comes from the pulse rate variability. Heart rates cannot vary faster than a maximum heartbeat variation threshold. The maximum heartbeat variation threshold may be incorporated into the physiological model to exclude pulse rate variations that exceed this threshold. Physiological models may also include degenerated pulse signals due to heart conditions, where the pulse may diminish or even vanish when the heart rate increases. Such phenomena may be measured with the combined use of various sensors.
• Blood oxygenation includes a noise model and a physilogcal model.
• the noise model expects the Sp02 value to be relatively constant over the measurement period.
• the physiological model expects the Sp02 value to be within a range of values to be valid. For example, between 99% and 80%.
• both PPG photodiode sensors A501 and A502 are positioned against the temple of the user. Both signals A602 and A606 are processed in parallel in the following sequence: The signal feeds into a FIR or FFT/IFFT filter (A603 respectively A607) removing noise from the data.
• the PPG-patient interaction model A511 of Figure 5 A (see additional explanation in Figure 5B) is used to validate the
• FIG. 5B Optimal positioning of the PPG diodes is shown in Figure 5B.
• the PPG diodes A501 and A502 are in direct contact with the temple, with features blocking external light interference from A503 light
• PPG diodes A501 and A502 may intermittently be perturbed by interfering ambient light, which will reduce accuracy of Sp02 predictions.
• the ambient light interference may also translate into saturation of the signals.
• FIG 6B illustrates an example of PPG signals B601 and B602 being respectively collected by PPG photodiodes A501 and A502 for various levels of Sp02 saturation.
• the Infrared (IR) PPG photodiode A502 generates signal B602 for the different saturation levels.
• the red photodiode A501 generates signal B601 for the different saturation levels.
• Sp02 is a slow varying parameter for average, alert users, and may sometimes be considered to be constant during a measurement cycle. Other times, variability of Sp02 over time may be used in determination of respiratory rate. Oxygenation should be values between 96% and 99% in healthy individuals. If oxygenation drops to 92%, a user should use oxygen to increase the level. If oxygenation drops to 87%, is important that oxygen be used to avoid impairment. Impairment of mental functions can occur below 85%.
• the Pulse Width Transit Time includes a physiological model and a pair of noise models.
• the first noise model is to use an extracted PWTT value and a cross correlated PWTT value determined by different signal processing methods and fuse them together to determine a more precise and more stable value than that determined from peak detection alone.
• the second noise model is to obtain an extracted PWTT value by using measurements of peak detection through local estimations of PWTT that allows discarding parts of the signal without the peaks.
• the physiological model considers that the PWTT depends on the heart blood pressure and the vascular system.
• the PPG signal S701 is coupled into a finite impulse response (FIR) or fast Fourier transform (FFT/IFFT) filter A702 to remove noise from the data.
• FIR finite impulse response
• FFT/IFFT fast Fourier transform
• the PPG-patient interaction model A511 (see additional explanation in Figure 5B, 5C and 5D) is then used to validate the measurements at this stage.
• the intensity of the PPG signal is then normalized in process A703 and flows to a first derivative (time gradient) filter A704 to form the differentiated PPG signal S704.
• the differentiated PPG signal S704 is further filtered by an LI trend filter A705 forming the LI trend differentiated PPG signal S705 shown in Figure 7B.
• the A706 ECG signal is coupled into a FIR or FFT/IFFT filter A707 to remove noise from the desired data in the ECG signal.
• the ECG-patient interaction model A408 (see additional explanation with reference to Figures 4B, 4C, 4D, 4E) is then used to validate the measurements of the ECG signal at this stage.
• the intensity of ECG signal is then normalized by a normalizer A708 which is then coupled into an LI trend filter A709 to generate the LI trend ECG signal S706t shown in Figure 7B.
• the L 1 trend ECG signal S706t from the L 1 trend filter A709 and the L 1 trend differential PPG signal S705 are cross-correlated by a cross-correlation function A710 to form a maximum ECG QRS complex signal S710.
• the outcome of the cross correlation, the maximum ECG QRS complex signal S710 is coupled into a peak detector to under a peak detection process A711 to calculate a raw PWTT value.
• a model merger process A712 occurs with the PWTT value and a PWTT physiological model A714 to validate and form the final PWTT values.
• the final PWTT values may be shown the user and/or stored into health records. These final PWTT values may be referred to as cross correlation PWTT values.
• FIG. 7B provides a graphical illustration of the PWTT algorithm.
• the ECG signal S706 corresponds to a normalized ECG signal. This signal goes through the LI trend filter to generate the LI Trend ECG signal S706t.
• the LI Trend ECG signal S706t is a first input into the cross-correlation function A710.
• the PPG signal is normalized in A701 goes through a differential function A704 to finally traverse an LI trend filter transforming into signal S705.
• the LI Trend differential PPG signal S705 is a second input into the cross-correlation function A710.
• the signal resulting from the cross-correlation function A710 enters into the Max peak module A711 extracting the PWTT corresponding to the maximum peak of the cross- correlation.
• An ECG extraction process extracts features from the ECG signal S706 to form an ECG QRS complex features signal.
• a PPG extraction process extracts features the PPG signal S701 to form a PPG features signal.
• a further peak extraction process is performed on ECG QRS complex features signal to extract the peak features ECG peaks C701 that correspond to the R feature of the ECG QRS complex.
• a further peak extraction process is performed on the PPG features signal to extract the peak feature PPG peaks C702 (or the rise time) associated with the PPG signal S701.
• the calculation of the PWTT signal C703 is then based on the time difference between the ECG peaks C701 and the PPG peaks C702.
• the total error on the PWTT is the sum of the errors from peak extraction plus the errors in the peak alignment algorithm. However, the total error is averaged over the number of peaks in the observation window or time window of samples. These PWTT values may be referred to as extracted PWTT values.
• the PWTT value calculated by the extraction process A750 is coupled into the model merge module A712.
• the PWTT value calculated by the extraction process A750 can be combined with the PWTT calculated by the correlation process generated by the maximum peak detection process A711, as well as the PWTT physiological model A714.
• a merging or fusion algorithm can mitigate the risks of error.
• One merging or fusing approach to use in the model merge process A712 is to calculate PWTT based on the peak matching algorithm to get an approximate value, as well as to assess the quality of this signal based on the number of peak detections. Once this signal is validated, the cross-correlation algorithm can be used to perform a more precise determination and validate the overall PWTT result. A valid and accurate PWTT value is important in the computation of blood pressure.
• the calculation of blood pressure includes a physiological model and a noise model.
• the noise model is to use the sum of values for PWTT, pulse rate, heart rate, PPG spectral slope, PPG wave shape, ECG spectral slope, and ECG spread to improve signal quality.
• the physiological model considers that the systolic and diastolic blood pressure should be relatively constant over the measurement period with the values for systolic blood pressure being greater than the values for diastolic blood pressure.
• blood pressure calculation process A800 is now described that includes a PPG/patient interaction model. Both PPG sensors A501,A502 are positioned against the temple of the user. The output signals of the PPG sensors A501 and A502 are processed during a PPG process A801 to determine and select a more reliable and perhaps less noisy PPG signal to undergo further signal processing.
• ECG signals are captured during an ECG capture process A810 by the ECG sensor and its electrodes A401f, A401T making respective contact with a finger and temple of the user.
• the ECG signal from the ECG capture process A810 is coupled into a heart rate signal processing algorithm A400 to generate a heart rate signal A407.
• the ECG signal from the ECG capture process A810 and the PPG signal from the PPG capture process A801 are coupled into the PWTT calculation process A700 to calculate the PWTT result A713.
• the PPG signal is also coupled into a FIR or FFT/IFFT filter A802 removing noise from the data.
• the PPG-patient interaction model A511 (see also Figures 5B and 5C and the description thereof) is used to validate the measurements at this stage.
• the intensity/amplitude of PPG signal is then normalized by a normalization process A803 to form a normalized signal A803 that flows in several directions.
• the normalized PPG signal A803 is coupled into a pulse rate calculator A520 to generate the user's pulse rate A510.
• the normalized PPG signal A803 is coupled into a harmonic analysis A804 to produce the PPG wave-shape A807.
• the normalized PPG signal A803 flows into a latent profile analysis (LP A) A805 to calculate the PPG spec slope A808.
• the normalized PPG signal A803 feeds into a statistical module A806 to calculate standard deviation and absolute deviation, referred to as the PPG spread A809.
• the calculated values A809, A808, A807, A510, A713 and A407 are then all coupled into a function f(.) A822 and a function h(.) A821.
• the function f(.) A822 and the function h(.) A821 apply mathematical transformations to the incoming data values to transform them into components of systolic blood pressure and diastolic blood pressure.
• the resultant component values of these transformations are coupled into the regression algorithm A825.
• patient physiological data A820 e.g. age, sex, weight, size
• function g(.) A824 and the function h(.) A823 apply mathematical transformations to the incoming data values to transform them into components of systolic blood pressure and diastolic blood pressure.
• the resultant component values of these transformations are coupled into the regression algorithm A825.
• the user calibrates the regression algorithm A825 with a set of user parameters to weight the incoming components of systolic blood pressure values and diastolic blood pressure values that are then summed together to achieve known calibration systolic blood pressure values and diastolic blood pressure values. Subsequently, the user parameters for the regression algorithm A825 are used to calculate systolic blood pressure values and diastolic blood pressure values.
• the resultant components of systolic blood pressure values and diastolic blood pressure values from the transformations are coupled into the regression algorithm A825.
• the regression algorithm A825 In response to the user's custom set of parameters and the resultant component values from the transformations, the regression algorithm A825 generates the outputs of diastolic blood pressure (DBP) A826 and systolic blood pressure (SBP) A827.
• DBP diastolic blood pressure
• SBP systolic blood pressure
• Figure 8B illustrates an exemplary blood pressure physiological model that may be used to validate blood pressure measurements determined by the process described in Figure 8A.
• the chart in Figure 8B shows SBP in a range of 60-230mm Hg and DBP in a range of 30- 130mm Hg. Blood pressure measurement outside that range may be considered suspicious.
• Respiration rate can be extracted from either the PPG signal and/or from the ECG signal, with similar algorithms.
• a physiology model and a noise model are used during this process.
• the physiology model illustrates expect ranges of rates for adult, elderly living independently, elderly living in long term care, and a rage of respiration rates that suggest a medical condition (e.g., 10 beats per minute to 30 beat per minute).
• a first noise model uses the ECG peak amplitude and body noise as well as a determination that the ECG data is more stable that the PPG data during the same scan.
• a second noise model uses signal processing to look for patterns in sound signals around 20 Hz.
• a third noise model looks for accelerometer noises.
• a composition of acceleration is formed from three dimensions of acceleration and is compared with expected signals depending upon position of the vital signs scanner.
• sensors A501 and A502 are positioned against the temple of the user. Only the signal from one of the PPG sensors needs to be processed to determine respiration rate.
• the PPG signal S901 from the PPG sensor is coupled into a FIR or FFT/IFFT filter A902.
• the FIR or FFT/IFFT filter A902 removes noise from the PPG data signal.
• the PPG-patient interaction model A511 may be used to validate the PPG measurements from the PPG sensor.
• the resulting signal goes into a band pass filter A903 to remove low frequency components as well as high frequency component signals.
• the signal is then processed by a peak detection algorithm A904 that is responsive to the amplitude of the signal. The number of peaks in amplitude detected by the peak signal detector over a period of time is used to determine respiration rate.
• a single ECG sensor A401 with electrodes A401t and A401f closes an electrical circuit with the user's body between a finger/thumb coupled to electrode A401f and the temple of the user coupled to electrode A401t.
• the ECG signal A906 of the ECG sensor is coupled into a FIR or FFT/IFFT filter A907 to remove signal noise from the data signal.
• the PPG-patient interaction model A408 (see additional explanation in Figure 4B, 4C, 4D and 4E) is used to validate the ECG signal measurements from the ECG sensor.
• the resulting signal goes into a band pass filter A908 to remove low frequency components as well as high frequency components from the ECG signal.
• the filtered ECG signal then undergoes a peak detection process with a peak detection algorithm A909 executed by a signal processor.
• the number of peaks detected over a period of time provides the respiration rate. More specifically, distribution of ECG peaks (R-waves) gives information about the heart rate as a function of time. Heart rate variability may be inferred from this function. Respiratory rate can be estimated from the heart rate variability.
• the PPG extraction method or ECG extraction method causes signal A904 and/or A909 to enter into a model merge A910, comparing them to the physiological respiratory rate model A912.
• the physiological respiratory rate model A912 incorporates physiological limitations of respiration rates. For example the breathing rate of an average adult human is between 10 and 30 breaths per minute.
• Figure 9B illustrates details of the exemplary ECG algorithm used to extract respiration rate.
• the ECG signal A906 feeds into a FIR or FFT/IFFT filter A907 to form the filtered signal S907.
• the filtered ECG signal S907 goes through the band pass filter A908 to generate the bandpassed ECG signal S908 shown in Figure 9B.
• the bandpassed ECG signal S908 enters the peak detection algorithm A909 to generate the peak detected ECG signal S909.
• One way of performing peak detection is to set a threshold B902 and count the number of times the signal B901 extends beyond this threshold.
• Other algorithms known in the art can also be used.
• microphone Al 101 is pressed against the chest of the user to measure chest sounds.
• the output signal from the microphone Al 101 is coupled into a signal processor Al 102.
• the signal processor Al 102 executes one or more signal processing algorithms, including an FFT/IFFT algorithm Al 102 to remove some noise from a chest sound signal.
• the processor can then extract an initial respiration rate from the chest sound signal that has some noise removed.
• the signal processor Al 102 further receives a microphone-patient interaction model Al 105 (see additional explanation with reference to Figures 1 IB and 11C).
• the microphone- patient interaction model Al 105 includes an identified noise signal.
• the FFT/IFFT algorithm Al 102 can use the microphone-patient interaction model Al 105 to remove more noise (e.g., the identified noise signal) from the chest sound signal.
• the signal processor Al 102 can then extract an intermediate respiration rate from the chest sound signal that has more noise removed.
• Note the microphone-patient interaction model Al 105 may also be used by the signal processor Al 102 to compare and recognize respiratory conditions in the data signal.
• a model merge Al 103 includes a validation process that compares the intermediate respiration rate is compared to a physiological respiration model A906.
• the comparison includes checks for physiological validity of the intermediate respiration rate.
• the physiological respiration model A906 includes an ECG signal captured from an ECG sensor (see additional explanation described with reference to Figure 9B). If determined to be valid, the intermediate respiration is accepted as the overall respiration rate A904.
• the overall respiration rate A904 is presented to the user and/or stored into health records on a storage device.
• Figure 1 IB illustrates the positioning of the device on the user's chest.
• the housing of the vital signs scanning device Bl 101 is pressed up against the user's chest.
• the microphone Al 101 in the vital signs scanning device Bl 101 records the lung sounds around the chest. If the user is lying on their back, the vital signs scanning device Bl 101 may simply be laid onto the user's chest.
• Figure 11C illustrates typical lung sounds captured by microphones. These lung sounds are classified into well established sound categories, such as vesicular, crackles, wheeze, rhonchi, bronchial, pleural rubs, and bronchovesicular sounds. It is known how to analyze each of these sounds to extract respiration rates from a signal.
• FIG. 10A A simple exemplary process for determining respiration rate is shown in Figure 10A.
• a microphone Al 101 is placed at the user's chest and a sound recording is made into a sound recording system A 1001. The recordings are analyzed and respiration rates are extracted.
• a three dimensional accelerometer A1301 in a housing is pressed against the chest of the user to measure chest motion as the user breathes.
• a 3D accelerometer output signal from the three-dimensional accelerometer A1301 is coupled into a signal processing module A1302.
• the signal processing module A1302 removes some noise from the chest motion signal (e.g., by applying FFT/IFFT).
• the processing module A1302 also detects peaks (e.g., estimates peak-to-peak time intervals) in the chest motion signal by using a moving average peak detector.
• the processor module A 1302 can use the detected peaks to extract an initial respiration from the chest motion signal.
• the signal processing module A1302 further receives a second input (e.g., noise signal) from an accelerometer-patient interaction model A1305.
• a second input e.g., noise signal
• the accelerometer-patient interaction model A1305 is further described herein with reference to Figure 13B and 13C.
• the peak detection signal from the signal processing block A1302 is coupled into a peak-to-peak interval estimator A1303 that performs more peak-to-peak time interval detection.
• the peak-to-peak interval is a measurement that corresponds to the intermediate respiration rate of the user.
• a comparator A1304 compares the measured peak-to-peak time intervals against a physiological respiration model A906.
• the physiological respiration model A906 is described with reference to Figure 9B.
• the comparator A1304 checks the measured peak-to-peak time intervals for physiological validity. If the measured peak-to-peak time intervals are
• the overall respiration rate is shown to the user on a display device and/or stored into a storage device with a health records database.
• Figure 13B illustrates positioning the vital signs scanning device Bl 101 on the user's chest.
• the housing of the device Bl 101 is simply pressed against the chest of the user.
• the 3D accelerometer A1301 in the device Bl 101 records chest motion. If the user is lying on their back to obtain a better reading, the device may simply be laid on the chest of the user.
• FIG. 13C illustrates waveforms of an exemplary accelerometer-user interaction model.
• the waveform signal CI 301 represents the normal or perpendicular motion of the chest wall while breathing.
• the model represents the expected normal motion of a user's chest as he/she breathes. However, the frequency of motion of the user's chest still needs to be determined. Chest movement is measured by the accelerometer A1301.
• the signal generated by the accelerometer A1301 as a result of the chest movement detects, for example a vector component normal to the chest, or a vectorial sum of the X, Y, and Z axis of accelerations of the chest.
• the peak detection algorithm A1302 executed by the signal processor in response to the model A1305, extracts the breath inhalation frequency of the user over time.
• the signal processor further analyses the peak-to-peak time intervals in the breath inhalation frequency and converts it using the respiration rate algorithm A1303 into a measured respiration rate.
• the comparison algorithm A1304 compares the measured respiration rate to the respiration rate physiological model A906 to be sure the measured respiration rate is valid, before generating the output respiration rate A904.
• the raw recorded data of the chest movement could be analyzed later in some cases. If this is the case, a simpler embodiment such as shown in Figure 12A may be used to record and store chest movement.
• the vital signs scanning device Bl 101 with the three dimensional accelerometer A1301 is placed against the chest of the user.
• the output signals from the accelerometer A1301 are recorded over time by a motion recording system A 1201.
• the recorded output signals can be stored in a storage device for later analysis and comparison.
• the system described herein performs multiple measurements in parallel. Since sensors are all closely positioned in the same housing of the vital signs scanner 100, they receive similar noise, such as vibrations or motion.
• a housing or enclosure A1401 of the vital signs scanner includes a plurality of sensors, including an accelerometer A1301.
• the accelerometer A1301 in the housing records the vibrations and motions A 1201 of the scanner 100.
• the vibrations and motions A 1201 of the accelerometer may be shared with the other sensors.
• thermometer A301 thermometer A301
• PPG sensors A501 and 502 ECG (40 It A401f) are indirectly perturbed by vibrations. They do not directly react to vibrations or motion, but the quality of their signal depends on the sensor/patient interaction model.
• thermometer In the case of the thermometer, vibrations have a lesser effect due to the constant air gap maintained by mechanical design between the sensor and the temple. However relative motion between the temple and the sensor may show differences if the sensor moves over the temporal artery, because blood in the temporal artery is warmer than other blood in the temple.
• the ECG signal from the ECG sensor shows noise pollution spikes in response to a small vibration.
• the PPG signal from the PPG sensor shows signal saturation when external light breaks the seal between the device and the temple.
• An example of an improvement gained from using multiple sensors is an algorithm for the respiration rate that combines the results of accelerometer A1301, microphone Al 101 and PPG-ECG (A501, A502, A401t and A401f) sensor data.
• the fusion algorithm initially considers the results of the accelerometers measuring the chest motion since this is a simple sinusoidal motion in a single dimension. Maximum chest motion (highest speed) corresponds to the peaks in the breathing action.
• Minimum chest motion corresponds to full inhalation and full exhalation. This provides the initial determination of the respiration rate, with the precise moments corresponding to maximum inhalation time point t m and maximum exhalation time point t ex .
• time points t m and t ex By providing the time points t m and t ex to the microphone respiration rate algorithm and to the PPG-ECG respiration rate algorithm, it can predict alternate measures t' m and t' ex for maximum inhalation time point and maximum exhalation time point to verify the prediction for respiration rate.
• the final determination of respiration rate A1401 may be a combination of the respiration rates determined from the measures for maximum inhalation time point t m and maximum exhalation time point t ex , and alternate minimum inhalation time point t'm and alternate maximum exhalation time point t' ex .
• BP blood pressure
• HR heart rate
• PR pulse rate
• Sp02 blood oxygenation
• core temperature is also improved by recognition of a noise signature in the signals from the accelerometers.
• the noise signature in signals from the accelerometer can be transmitted to a user interface for the scanner 100 to provide feedback to the user.
• the user interface in response to the noise signature, can encourage the user to improve the grip and positioning of the scanning device. For example, detection of device oscillations may mean the device needs to be pressed a little harder against the user's temple.
• Another possible improvement in signal quality may be made detecting steady motion of the scanner 100.
• a steady motion in the scanning device 100 can cause some measurements, such as temperature from the temperature sensor, to capture less accurate data and generate less accurate results.
• a message may be presented to the user by the user interface to hold scanner steady.
• Training sessions can also be provided by the user interface software application of the vital signs scanner 100.
• a training session can show where the scanner 100 is to be positioned against the body.
• a training session can be used to show how the scanner 100 is to be gripped.
• a training session by the user interface can propose exercises to address common mistakes that may be made using the device.
• sounds may be emitted by vital signs scanner and its user interface to train users in reaching the optimal use of the device.
• the device can be used to statistically collect accelerometer measurements to assess usage improvement of the vital signs scanner by the user and pinpoint opportunities for further improvement in its use.
• the device For chest measurements the device performs two measurements in parallel. Since the sensors are closely positioned in the device housing they share noise, in particular the vibrations and motion.
• the device enclosure A 1401 is represented by a shape containing all sensors.
• the accelerometer A1301 records all the accelerations of the body enclosure, whereas the microphone Al 101 records all the sounds. Since both are mechanical vibrations, they can be used to correct each other and provide more accurate respiration rates.
• respiration inhalation and the exhalation correspond respectively to the maximum and the minimum in the chest motion. This provides the initial determination of the respiration rate, with the precise moments corresponding to maximum inhalation t ln and maximum exhalation t ex .
• n and t ex to the microphone respiration rate algorithm, simple verification techniques on the sound track can be made, instead of making more complex determinations.
• the final determination of respiration rate A1501 may come either from the microphone Al 100, or from the accelerometer A1300.
• the device may be positioned on their chest. This reduces signal pollutions as the accelerometers are just sensing the up and down chest motions and not movement from the user swaying, fidgeting, etc. Sensor fusion
• the vital signs scanner 100 concurrently performs multiple measurements in parallel with multiple sensors during a measurement period under control of the same processor. Due to these parallel operations over the same timeline, some measurements from one or more sensors may be used to correct/validate others and improve the quality of the vital sign measurements generated by the device.
• the device housing A 1401 is represented by a shape containing multiple sensors.
• An accelerometer A1301 records accelerations of the housing, which can be shared with other sensors.
• the sensor data or signals from the multiple sensors in the vital signs scanning device 100 may be fused together, perhaps with other data as well, to improve measurement quality of the vital signs.
• thermometer A301 with its signal processing A300, the PPG sensors A501 and 502 are connected to all of the following - pulse rate A500, Sp02 rate A600, blood pressure A800 and, respiration rate A900.
• the ECG sensor with single lead ECG with an electrode A401t on the temple and an electrode A401f on the finger/thumb is connected to all of the following - heart rate A400, respiration rate A900 and, blood pressure A800.
• the sensor data from the microphone Al 101 and the sensor data from the accelerometer A1301 captured concurrently may be fused together by a signal processor to generate a higher quality of measurement of respiration rate A900 for a user.
• one or more signals of sensor data may be used to validate other data or vital sign measurements in a measurement validation process A 1601.
• a data fusion algorithm may be used to fuse different data together to validate or enhance the quality of the vital sign measurements.
• a fusion algorithm receives a measure of respiration rate from three sources - PPG- ECG, microphone, and accelerometers - and compares them to assess the validity of their measurements.
• the different measures of respiration rate from each may be weighted based on a perceived quality in their data. For example, if the signal from the accelerometers used to determine respiration rate was not too noisy, its signal and measurement may be given greater weight in a weighted sum of the outputs.
• the fusion algorithm may raise the confidence of both ECG and PPG and give it a greater weight in the weighted sum of the outputs. With the respiration rate from fusing PPG- ECG sensor data being comparable to the respiration rate determined from the accelerometer and microphone, the fusion algorithm may also raise the confidence of their use in the blood pressure calculation process A800 by the signal processor.
• Another example of a fusion algorithm to fuse different data from different sensors is to compare the pulse rate and heart rate that has been determined in the signal processing processes A500 and A400. In most cases, these values are very similar. With similar pulse rate and heart rate being determined in the respective processes A500 and A400 by the signal processor, the trust of the ECG sensor data and PPG sensor data can be raised and weighted more heavily due to the sensor data being of higher quality. Because ECG and PPG are also used to determine the vital signs of blood oxygenation and pulse rate, and their quality of sensor data is determined to be higher, the trust in the vital sign measurements for Sp02 and pulse rate determined in signal processing steps A600 and A500 are higher.
• Motions A 1201 of the scanning device 100 can be used to validate and invalidate sensor data and the vital sign measures. For example, core temperature determined from the sensor data measured by the temperature sensor A301 can be trusted when there is little motion of the scanning device 100 near the temple of the user. If the motion A 1201 sensed by the one or more accelerometers A1301 is nearly zero, the core temperature determination is higher in quality and can be validated by the lack of motion. If the motion A 1201 sensed by the one or more accelerometers A 1301 is substantial, the core temperature determination is of lower quality and invalidated during the measurement validation process A 1601.
• the measurement validation algorithm and signal processing process A 1601 may validate the data in an all or nothing mode in some cases. All vital sign measurements may be invalidated in some cases. For example, if a plurality of sensors capture no data, or have constant data with no change, the scanner 100 may be improperly and so all vital sign data should be considered to be of low quality and perhaps discarded. Alternatively in other cases, the measurement validation algorithm and signal processing process A 1601 may partially validate the data, updating only the validated vital sign measurements and setting blank or zero values for the output vital sign values for invalidated vital sign measurements.
• the vital signs scanning device 100 may concurrently perform two measurements in parallel.
• sound measurement or capture of sounds by microphone Al 101
• the measurement or capture of accelerations by accelerometers A1301
• the sensors contained within the same housing of the vital signs scanning device 100 the sensors may be submitted to similar stimuli but capture it in different manners.
• the signals of the sensors including sound signals captured by the microphone, share the same timeline. Accordingly, features of the sound in a sound signal can be compared, time-wise to the features of a motion signal captured by the accelerometers, for example.
• a fusion algorithm for the respiration rate combines the results or signals of an accelerometer and a microphone, for example.
• the results of the accelerometers measuring the chest motion A1300 are initially considered, because their model is a simple sinusoidal motion in a single dimension.
• the respiration inhalation and the exhalation correspond respectively to a maximum or peak chest position and a minimum or lowest chest position. This provides the initial determination of the respiration rate, with the precise moments in time of the peak chest position and the lowest chest position corresponding to a maximum inhalation time n and maximum exhalation time t ex .
• the time points of maximum inhalation time t ln and maximum exhalation time t ex are coupled into the microphone respiration rate algorithm Al 100.
• the microphone respiration rate algorithm Al 100 can use simple verification techniques with the sound track of the recorded sounds using the maximum inhalation t ln and maximum exhalation te X , instead of making more complex determinations.
• a measurement validation A1701 including a final validation algorithm, can be implemented by checking that the microphone determined respiration rate Al 100 is equal or close enough to the accelerometer respiration rate A1300 determined from the accelerometers A1301 to validate the displayed respiration rate A 1702. If the accelerometer respiration rate A1300 and microphone determined respiration rate Al 100 are substantially different then the displayed respiration rate A 1702 may be set to a blank value.
• the vital signs scanning device 100 may be positioned on their chest to measure respiration rate. This reduces signal pollutions as the accelerometers are just sensing the up and down chest motions, and avoid measuring other user movements, such as from the user swaying, fidgeting, etc.
• the vital signs scanner 100 was introduced by Figures 1 and 2A-2B and the description thereof.
• Figure 18 illustrates further details regarding the vital signs scanner 100.
• U.S. Patent Application No. 12/292,820 filed on May 30, 2014 by Wenyi Zhao et al. and incorporated by reference, further describes the vital signs scanner 100.
• the electronic circuitry 800 of the portable wireless vital signs scanner 100 includes a processor 840 at the heart of the system.
• the processor 840 may be a reduced instruction set processor operating with embedded operating system software.
• the processor is an ARM processor operating with MICRIUM's embedded real time operating system (RTOS).
• a wireless radio 870 is coupled to the processor 841.
• the wireless radio 870 is coupled to an antenna 871 that could be internal, as part of an overall radio system, or external to the wireless radio 870.
• An optional light emitting diode 848 used as a wireless connection indicator, is coupled to the wireless radio to indicate a successful pairing with the personal portable wireless digital multifunction device 104.
• the electronic system 800 includes an infrared thermometer 812, an accelerometer 885, a pulse oximetry sensor and a pulse oximetry circuit 880, and analog electrocardiogram circuitry 860.
• Coupled to the electrocardiogram circuitry 860 is the bottom or top electrode 806B, the front electrode 711, bottom/top electrode connection, and the front electrode connection 806F. As shown in Figure 1 A, a portion of a human body is coupled to the front electrode 711 and the top/bottom electrode 806B to form a circuit.
• the pulse oximetry circuit 880 is coupled to a pair of light emitting diodes 808A- 808B. Each of these emit light patterns that are reflected off of the user's forehead/temple internally. The reflected light is captured by a photodiode 810 and coupled to the circuit 880. That is, incident light 891 from the light emitting diodes 808A-808B reflects internally off the user's head 116 as reflective light 892 which is received by the photodiode (PD) 810.
• PD photodiode
• the infrared thermometer 812 detects the surface temperature of a use's
• Blackbody radiation 813 emanating from the head 116 (or other body portion to which the scanner is pressed) of a user.
• a rechargeable battery 850 and a voltage regulator and battery charge controller 854 are coupled together into the circuits in the system 800 when the switch 852 is closed.
• the battery charge controller 854 is coupled to power pins of a serial connector 856 to receive an external DC voltage supply.
• the external voltage supply may be used to recharge the battery and power the system 800 when it is connected.
• the rechargeable battery 850 may hold a charge for a period of seven days, even while scanning multiple times during each day, due to the low power consumption of the circuitry and the limited period of time needed to perform a scan of the vital signs of a user. That is, the vital signs scanner 100 is not expected to be continuously powered on during a day, but powered up periodically to perform the scans as needed.
• the processor 840 may include a processor memory 841 to store system instructions to control the circuitry in the system to obtain the scans and process the information obtained through those scans into a proper user format.
• a nonvolatile memory 844 is coupled to the processor 840.
• the nonvolatile memory 844 may be soldered to a printed circuit board with the processor 840.
• a connector 845 is provided so that the nonvolatile memory 844 is a removable memory card so that a user's data may be transferred from one scanner to the next, if needed.
• a power LED 851 may be coupled to the processor 840 to provide an indication that the electronic system 800 is powered up.
• the system can be manually shut down via the scanning software application 140 so that the scanner 100 powers off. However, the scanner 100 can also automatically shut off after a predetermined period of time to conserve power and a charge on the rechargeable battery 850. The user then just needs to press the power switch 852, once again, to turn the system back on and scan for vital signs of a user.
• the processor 840 includes one or more analog digital converters 842 in order to receive analog signals from the infrared thermometer 812, accelerometer 885, pulse oximetry circuits 880, and ECG analog circuits 860.
• Electronic system 800 may further include a stereo microphone 875 consisting of a top microphone 875T and a bottom microphone 875B each coupled to a stereo microphone amplifier 874.
• the stereo microphone amplifier may have its own analog to digital converter, or the processor's analog digital converter 842 may be used to convert analog signals into digital signals.
• an ECG analog signal may be converted into digital signals with the analog digital converter 842 of the processor.
• the stereo microphone 875 captures audio signals near the wireless vital signs scanner 100.
• the accelerometer 885 captures movement of the portable wireless vital signs scanner 100.
• the combination of the audio information and the movement information may be utilized to determine the quality of the scanning information being obtained by the vital signs capturing circuitry.
• the stereo microphone 875 may be used to capture noise from a user talking and plot that on a graph indicating noise spikes, or noise lines 330, such as shown in Figure 3A. This provides feedback to a user about the quality of the scan at these intervals.
• the accelerometer 885 and the motion information may be similarly used to make a judgment about the quality of the vital signs scanned information being captured by the vital signs circuitry of the infrared thermometer 812, the pulse oximetry circuits 880, and the ECG analog circuits 860.
• the microphones 875 in the portable wireless scanner 120 may also used to capture body sounds such as shown in Figures IE- IF and store the captured body sounds in memory 844 as a potential symptom of a medical condition of the users body.
• body sounds such as shown in Figures IE- IF
• heart beat sounds 155 may be captured by the microphones 875 when the scanner 100 is positioned against skin of the chest 114 near ones heart 156, as is illustrated in Figure IE.
• lung or breathing of air entering and exiting ones lungs, respiration sounds 157 may be captured by the microphones 875 when the scanner 100 is positioned against skin of the chest 114 near a lung 158 in ones body, as is illustrated in Figure IF.
• scan quality algorithm monitor the vita signs scanning process and can provide feedback (visual and/or audible) to the user, such as through the multifunction device 104.
• An optional audible sound generator 847 in the scanner 100 may be coupled to the processor 840 to provide audible user feedback to the user during the scanning process.
• the user feedback may help the user to perform better vital signs scan with the wireless vital signs scanner 100 and acquire a higher quality of vital signs measurements.
• the audible sound generator 847 may generate alert sounds indicating when the scanning process begins and ends. It may also generate an error signal indicating to the user that he is not properly using the scanner 100 and provide instructions.
• the relative relationships between them are also contemplated as having been herein disclosed as additional beneficial aspects (even if the specific values of the relative dimensions change).
• certain lengths, widths, and/or depths of particular components shown and described for a particular assembly provide overall geometries, which may be varied by changing certain sub-sets of such dimensions, but may also be fixed relative to the ratios of these values despite the values changing (as long as their general relationship remains constant).
• such dimensions of different component parts also have similar relative relationships, which are similarly contemplated, also as apparent to one of ordinary skill.
• the elements of the embodiments of the invention are essentially the code segments or instructions to perform the functional tasks described herein.
• the code segments or instructions are executable by a processor, such as a signal processor 206 shown in Figure 2B, and can be stored in a storage device or a processor readable storage medium, such as memory 208, awaiting execution.
• the processor readable storage medium may include any medium that can store information. Examples of the processor readable storage medium include an electronic circuit, a semiconductor memory device, a read only memory (ROM), a flash memory, an erasable programmable read only memory (EPROM), a floppy diskette, a CD-ROM, an optical disk, a hard disk.
• the code segments or instructions may be downloaded via computer networks such as the Internet, Intranet, etc. into the processor readable storage medium.

Landscapes

• Health & Medical Sciences (AREA)
• Life Sciences & Earth Sciences (AREA)
• Cardiology (AREA)
• Engineering & Computer Science (AREA)
• Physics & Mathematics (AREA)
• Molecular Biology (AREA)
• Public Health (AREA)
• Pathology (AREA)
• Veterinary Medicine (AREA)
• Biomedical Technology (AREA)
• Heart & Thoracic Surgery (AREA)
• Medical Informatics (AREA)
• Biophysics (AREA)
• Surgery (AREA)
• Animal Behavior & Ethology (AREA)
• General Health & Medical Sciences (AREA)
• Physiology (AREA)
• Pulmonology (AREA)
• Signal Processing (AREA)
• Vascular Medicine (AREA)
• Spectroscopy & Molecular Physics (AREA)
• Optics & Photonics (AREA)
• Artificial Intelligence (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Psychiatry (AREA)
• Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
• Measuring And Recording Apparatus For Diagnosis (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (1)

Family

ID=54392922

Family Applications (1)

Country Status (3)

Cited By (13)

Families Citing this family (9)

Citations (5)

Family Cites Families (10)

• 2015
  • 2015-05-05 WO PCT/US2015/029330 patent/WO2015171667A1/en active Application Filing
  • 2015-05-05 CN CN201580023418.7A patent/CN106255449A/zh active Pending
  • 2015-05-05 EP EP15789393.4A patent/EP3139825A4/en not_active Withdrawn

Patent Citations (5)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events