WO2015052519A1

WO2015052519A1 - Drug delivery system

Info

Publication number: WO2015052519A1
Application number: PCT/GB2014/053035
Authority: WO
Inventors: Paulo Alexandre Da Torre Pinheiro; David Pettigrew; Anthony Male; Mark TUCKWELL; Edward Tucker; Paul Christopher Roberts
Original assignee: Sagentia Limited
Priority date: 2013-10-08
Filing date: 2014-10-08
Publication date: 2015-04-16
Also published as: GB201317802D0

Abstract

Drug delivery systems are described that include sensors and processing circuitry that can detect operating events, such as flow rates and drug delivery, in various types of inhalers, such as dry powder inhalers, metered dose inhalers, nasal inhalers and nebulisers. The system has three main parts –a drug delivery device that makes measurements during operation, a user device that receives measurement data transmitted from the drug delivery device; and a remote server that processes the measurement data and determines various operating parameters of the drug delivery device, including compliance data indicating correct of incorrect use of the drug delivery device. At least the compliance data is transmitted back from the remote server to the user device for output to the user to indicate correct or incorrect use of the drug delivery device.

Description

Drug Delivery System

The present invention relates to drug delivery devices, parts thereof and methods. The invention has particular, although not exclusive, relevance to inhalers and to the use of sensing technology to monitor and measure use of various inhalers, such as dry powder inhalers (DPIs), metered dose inhalers (MDIs), breath-actuated inhalers, capsule inhalers, reservoir inhalers, nasal inhalers, nebulisers and the like.

Inhalers are well known drug delivery devices. One of the main concerns about such drug delivery devices is the user's compliance with the intended usage. These medical devices are intended to deliver a drug dose to the lung to treat diseases such as asthma and Chronic Obstructive Pulmonary Disease (COPD). However, if used incorrectly, the discharged drug can end up being deposited at the back of the throat instead of the lung. Examples of incorrect use include mis-timing of inspiration and device triggering, incorrect inspiration rate and incorrect inspiration duration. There is therefore a need to determine how the user is using the device. This can then be used to store usage information that can be subsequently transmitted to a physician or the like; or that can be used to control a user interface to provide feedback to the user, for example indicating correct or incorrect usage. WO201 1/135353 describes an inhaler device that has a microphone mounted on a body thereof and processing circuitry that processes the signal from the microphone to detect operation of the inhaler. The device measures the energy in the signal from the microphone and converts this into a measure of the flow within the inhaler. By monitoring the energy during an inhalation, a flow profile for the inhalation can be determined. Other parameters, such as activation of the breath activated mechanism (BAM) can also be detected by suitable processing of the signal from the microphone. WO'353 also mentions the possibility of the processing circuitry having a communications module to allow the processing results or the data measurements to be transmitted to a remote location where the results can be logged and viewed by a clinician or doctor. The present invention aims to improve on the device described in WO'353.

According to one aspect, the present invention provides drug delivery system comprising: a drug delivery device comprising: a body with a mouthpiece; a sensor for sensing pressure or vibration waves during operation of the drug delivery device; processing circuitry operable to process a signal obtained from the sensor to generate measurement data; and short range communication circuitry for transmitting said measurement data from the processing circuitry; a user communication device comprising: short range communication circuitry for receiving the measurement data transmitted from the short range communication circuitry of the drug delivery device; network transceiver circuitry for transmitting the measurement data to a communications network and for receiving signals from the communications network; and a user interface for outputting information to a user; and a remote server comprising: transceiver circuitry for receiving the measurement data from the user communication device via the communications network and for transmitting data back to the user device via the communications network; processing circuitry for processing the measurement data and for determining operating parameters of the drug delivery device, the operating parameters including compliance data indicating compliance of the operation of the drug delivery device; and a controller for outputting the compliance data to the transceiver circuitry for transmission back to the user communication device; wherein the user communication device is operable to receive the compliance data from the remote server and is operable to perform at least one of: outputting the compliance data to the user via the user interface and transmitting the compliance data back to the drug delivery device using the short range communication circuitry.

The invention also provides a computer program product comprising computer implementable instructions for causing a programmable processor device to become configured as the processing circuitry of the drug delivery device or the processing circuitry of the remote server. The program product may include a CD, a DVD or other recording medium.

In order to aid in the understanding of the present invention, a number of exemplary embodiments will now be described in detail with reference to the accompanying figures in which:

Figure 1 a is a schematic view of an inhaler system having an inhaler, a detachable sleeve, a user communication device and a remote computer server of one embodiment of the application; Figure 1 b is an exploded view illustrating the components of the inhaler illustrated in Figure 1 a; Figure 2 is a block diagram illustrating the main components of processing circuitry that is mounted in the detachable sleeve of the inhaler system shown in Figure 1 a;

Figure 3 is a signal diagram illustrating the way in which a windowing function module of the processing circuitry shown in Figure 2 extracts windows or blocks of samples from an input audio signal;

Figure 4 illustrates the form of measurement data generated by the processing circuitry shown in Figure 2; Figure 5 is a block diagram illustrating the main components of the user device of the inhaler system shown in Figure 1 a;

Figure 6 is a block diagram illustrating the main components of the remote computer server shown in Figure 1 a;

Figure 7 is a plot illustrating the way in which the energy within the sensed acoustic signal varies with the flow rate of air through the inhaler of Figure 1 a;

Figure 8 illustrates a flow profile determined by the remote server using the energy measurements determined by the processing circuitry shown in Figure 2 and received from the processing circuitry via the user device;

Figure 9 is a block diagram illustrating alternative processing circuitry that can be used in the inhaler shown in Figure 1 a;

Figure 10 is a block diagram illustrating alternative processing circuitry that can be used in the inhaler system shown in Figure 1 a; Figure 1 1 illustrates a typical spectrum obtained from an FFT module forming part of the processing circuitry shown in Figure 10; and

Figure 12 illustrates a flow profile obtained by the remote server using measurement data provided by the processing circuitry shown in Figure 10 and illustrating peaks corresponding to potential firings of the BAM mechanism.

Inhaler System - Overview

Figure 1 a illustrates an inhaler system 1 exemplifying the present invention. The inhaler system 1 includes an inhaler 3, a detachable sleeve 5, a user communication device 7 (such as a smart phone or the like), a base station 8, a communication network 9 and a remote computer server 1 1.

In operation, when a user uses the inhaler 3, a microphone within the sleeve 5 captures the noise and/or vibrations made during the inhalation. This captured signal is then processed by processing circuitry within the sleeve 5 to generate measurement data which is then transmitted wirelessly to the user device 7 by short range communication circuitry (not shown) mounted in the sleeve 5. The user device 7 then forwards the received measurement data to the remote computer server 1 1 via the base station 8 which may be a cellular telephone base station or a WiFi network base station. The remote computer server 1 1 receives the measurement data from the user device 7 and processes the measurements to determine various operating parameters of the inhalation, such as the flow profile of the inhalation, the timing of canister firing or firing of the Breath Activation Mechanism (BAM) (depending on the type of inhaler 3 used), the peak flow rate, the inhalation volume and the like. The remote server 1 1 then transmits at least some of the determined operating parameters back to the user device 7, where the operating parameters are displayed to the user on a display of the user device 7. The user device 7 may also send data back to the inhaler 3 for activating a user interface (not shown) on the detachable sleeve 5 to give the user feedback. For example, the signal from the user device 7 may cause a green light or a red light (on the sleeve 5) to be illuminated depending on whether or not the user has used the inhaler 3 correctly.

The inventors have found that arranging the system in this manner maximises the battery life of a battery used to power the electronic components of the sleeve 5; whilst facilitating use of the system and whilst easing the burden of regulatory requirements (as the user device 7 simply forwards data to the remote server 1 1 and displays data received back from the remote server 1 1 or forwards that data on to the inhaler 3).

As will be explained in more detail below, the measurement data from the processing circuitry is analysed using a set of algorithms that are tailored for the particular inhaler type being used. The information obtained by the remote computer server 1 1 can be used, for example, as a training aid for the user or for providing feedback to clinicians in clinical trials or to doctors or other physicians for patient monitoring.

A more detailed description will now be given of the different components of the inhaler system 1 described above.

Inhaler

Figure 1 b is an exploded view of the inhaler 3 shown in Figure 1 a. The inhaler 3 is a Metered Dose Inhaler (MDI) having a canister 4 containing the drug to be delivered and a body portion 6. The sleeve 5 has two parts - an upper part 5-1 and a base part 5-2 that clip together around the body portion 6 of the inhaler. The sleeve 5 houses a circuit board 13, a battery 15 for powering the electronics coupled to the circuit board 13 and a user switch 17 to power up the electronics. The circuit board 13 has a microphone, processing circuitry and short range communication circuitry. The microphone may be of any conventional type, such as a condenser microphone, MEMS microphone or a vibration based microphone (that senses vibrations through the wall of the inhaler 3). The processing circuitry may include various analogue circuits, but typically will include a microprocessor. The short range communication circuitry may communicate using a standard communication protocol such as Bluetooth, Zigbee, NFC, etc.

Circuit Board

Figure 2 is a block diagram illustrating the main components mounted on or coupled to the circuit board 13 in the first embodiment. As shown, the circuitry includes the microphone 18, the signals from which are input to an analogue to digital converter 21. The digitized samples obtained by the analogue to digital converter 21 are then input to a digital processor 23. The processor 23 may be any suitably programmed microprocessor or ASIC based device.

In this embodiment, the functions performed by the processor 23 are illustrated as processing blocks. These processing blocks may be implemented using hardware circuits but in this embodiment are implemented as software routines run by the processor 23. Thus, as illustrated in Figure 2, the acoustic samples obtained from the analogue to digital converter 21 are firstly processed by a windowing function 25 which divides the samples into discrete blocks of samples by applying a suitable windowing function (such as a Hamming windowing) to reduce the effects of noise added by the windowing process. Figure 3 illustrates the windowing process and shows that the windowing function 25, in this embodiment, extracts blocks 27-1 , 27-2, 27-3, 27-4 of samples which partially overlap each other. In other embodiments, the blocks 27 of samples may be non-overlapping. In this embodiment, the acoustic signal from the microphone 9 is sampled at a sampling rate of 44.1 kHz and the windowing function 25 generates blocks of samples of 50ms duration at a rate of 22 blocks per second. Of course, other sampling rates and windowing rates may be used.

As illustrated in Figure 2, the blocks of samples output by the windowing function 25 are then passed to band pass filters 29 and 31. The band pass filter 29 is arranged to pass frequencies between, for example, 3Hz and 10kHz and to block other frequency components outside this range. The filtered samples are then passed to an energy calculator 33 which calculates the energy within each block 27 of samples. The energy value thus calculated is then passed to a controller 35. The band pass filter 31 performs a narrow band filtering in order to extract a peak of the signal's spectrum caused by actuation or firing of the inhaler canister 4. In particular, as noted in WO201 1/135353 (the content of which is incorporated herein by reference), when the inhaler canister 4 fires and the drug is delivered, this produces a notable peak in the spectrum of the signal from the microphone 18 that is characteristic of the inhaler device (in the case of the inhaler 3 used in this embodiment at a frequency of approximately 1.6kHz). Therefore, in this embodiment the band pass filter 31 has a narrow pass-band centred around 1.6 kHz. As shown in Figure 2, the output from the band pass filter 31 is input to a thresholding module 37 which compares the filtered signal against a number of threshold values. In this embodiment, two threshold values are used by the thresholding module 37 - a low threshold value and a high threshold value. The results of the thresholding performed by the thresholding module 37 are passed to the controller 35. In response to receiving an energy value and the thresholding results for a block of samples, the controller 35 generates and outputs a measurement packet 41 (illustrated in Figure 4) comprising the energy value 41-1 obtained from the energy calculator 33, the thresholding results 41-2 from the thresholding module 37 and a time stamp 41-3 generated using a local clock 39. The measurement data 41 also includes a header 41-0 that includes ID data that identifies the processor 23. The processor 23 is associated in advance with a particular type of inhaler 3 and therefore, this ID data identifies the type of inhaler 3 that is used. To minimise power consumption, the controller 35 may include the energy value 41-1 , the thresholding results 41-2 and the time stamp 41-3 for a plurality of blocks 27 of samples within the output measurement data packet 41. The controller 35 outputs the measurement data packet 41 to the short range communications circuitry 42 which transmits the measurement data packet 41 using a low power wireless communication protocol such as low energy Bluetooth that can stream data at a rate of up to 6 kBytes per second to the user device 7. The short range communication circuitry 42 can also receive data back from the user device 7 and, if received, passes this data to the controller 35. This data may be used, for example, to confirm correct or incorrect operation of the inhaler 3. In response to receiving such data from the user device 7, the controller 35 may output to the user visual and/or audible indications via the user interface 45 to confirm correct or incorrect operation of the inhaler 3. This allows the user to get feedback or confirmation when they are using the inhaler correctly or otherwise.

Figure 2 also shows the battery 15 and a power supply module 47 that receives power from the battery and provides electrical power to the different circuit components shown in Figure 2.

User Device

Figure 5 schematically illustrates the main components of the user device 7 shown in Figure 1. The user device 7 will typically be a smartphone or tablet device. As shown, the user device 7 includes a transceiver circuit 71 which is operable to transmit signals to and to receive signals from a base station of a telephone network or a wireless LAN via one or more antennae 73. The user device 7 also includes short range communication circuitry 75 that can communicate with the short range communication circuitry 42 mounted in the detachable sleeve 5. As shown, the user device 7 also includes a controller 77 which controls the operation of the user device 7 and which is connected to the transceiver circuit 71 and the short range communication circuitry 75. The controller 77 is also connected to a loudspeaker 78, a microphone 79, a display 81 , and a keypad 83 (which may be a keypad 83 displayed on the display 81). The controller 77 operates in accordance with software instructions stored within memory 85. As shown, these software instructions include, among other things, an operating system 87 and an inhaler application 89.

The inhaler application 89 includes an inhaler communication control module 91 , a remote server communication control module 93 and a display control module 95. The inhaler communication control module 91 is responsible for controlling communications received from and communications transmitted to the short range communication module 42 on the inhaler circuit board 13 using the short range communication circuitry 75. The remote server communication control module 93 is responsible for controlling communications transmitted to and received from the remote server 1 1 using the transceiver circuit 71 and the antenna 73. The remote server communication control module 93 is programmed in advance with the network address of the remote computer server 1 1. The display control module 95 is responsible for displaying inhaler operating parameters (such as a calculated flow profile) received back from the remote server 11 on the display 81 of the user device 7.

Remote Computer Server

Figure 6 is a block diagram illustrating the main components of the remote computer server 1 1 used in this embodiment. As shown, the remote server 1 1 includes a transceiver circuit 121 which is operable to receive data from and to transmit data to the user device 7 via a network interface 123. The operation of the transceiver circuit 121 is controlled by a controller 127 in accordance with software stored in memory 129. As shown, the memory 129 includes, among other things, an operating system 131 , a user device communications module 133, a flow rate determination module 135, a flow profile determination module 136, a firing detection module 137, a firing timing determination module 139, a compliance determining module 141 and flow profile data 143.

The user device communications module 133 controls the communications with the user device 7. Measurement data 41 received from the user device 7 is sorted (re-ordered) using the time stamp information 41-3 in the received packets - in case packets are delayed through the communications network 9 and received out of order. The energy values in the received packets are passed to the flow rate determination module 135 that determines, for each energy value, the volumetric flow rate corresponding to the determined energy measure. In this embodiment, the flow rate determination module 135 uses a look up table which relates input energy values to corresponding flow rates. The look up table is calibrated in advance by drawing known flow rates through the inhaler 3 and measuring the energy in the corresponding signal obtained from the microphone 18. Figure 7 is a plot illustrating the data obtained for the present inhaler 3 during the calibration process. The same look up table can be used for inhalers of the same design, although different look up tables will be required for inhalers having different acoustic characteristics. Therefore, the remote server 1 1 may store multiple look-up-tables for use with different types of inhalers. The particular look up table used by the flow rate determination module 135 is selected based on the ID data included in the header of the measurement data 41 that identifies the type of inhaler that the measurement data is obtained from. As those skilled in the art will appreciate, instead of using a look up table to represent the measurements obtained during calibration, an equation, such as a quadratic function, may be used to define the relationship between the measured energy and the corresponding flow rate. The quadratic function for the plot illustrated in Figure 7 is also provided on the plot, where x is the measured energy for the current block 27 of samples and y is the corresponding determined flow rate.

The flow rates determined by the flow rate determination module 135 for the received measurement data packets 41 for an inhalation, are passed to the flow profile determination module 136 which uses the received flow rates to obtain a flow profile for the inhalation. Figure 8 schematically illustrates the resulting flow profile 44 that is typically obtained for an MDI type inhaler 3. In particular, during an inhalation, it is typically desirable that the flow rate increases from zero to a maximum desired flow rate and remains at this maximum flow rate for a period of time before decreasing back to zero as the inhalation ends. The desired peak flow rate is usually much lower than that achievable by most users - and too strong an inhalation is one of the many faults users have with using the inhaler 3. If the user inhales too strongly, then this typically results in the medicament not being inhaled into the lungs but instead lining the back of the user's throat. Similarly, if the user inhales too softly and the peak of the flow profile 41 is too low, then the air flow may not be sufficient to draw the medicament into the user's lungs. Therefore, the compliance determining module 141 compares the obtained flow profile 44 with stored flow profile data 143 (representing an ideal flow profile) and outputs an indication as to whether or not the user inhaled properly when using the inhaler 3. The user device communications module 133 sends the determined compliance indication back to the user device 7, where the compliance/non-compliance is displayed to the user on the user device's display 81. The indication may also be transmitted from the user device 7 back to the circuitry on the inhaler 3 so that the controller 35 can output user feedback via the user interface 45.

The threshold results 41-2 in the received measurement data are passed to the firing detection module 137, which uses the threshold results to detect firing of the inhaler 3. The threshold results identify whether the signal level output from the band pass filter 31 was above or below the low threshold value and whether it was above or below the high threshold value. If the signal level was below the low threshold, then the firing detection module 137 determines that no firing of the inhaler 3 occurred in the corresponding block 27 of samples. If the signal level was above the low threshold value but below the high threshold value, then the firing detection module 137 uses this to identify a faulty firing - perhaps because the inhaler canister 4 is nearly empty or because there is a partial blockage of the metering valve. If the signal level exceeds the high threshold value then the firing detection module 137 determines that the inhaler 3 did fire during the corresponding block 27 of samples. Typically, the sound of the firing of the inhaler 3 will last approximately 200ms and so the firing detection module 137 should identify the firing of the inhaler 3 within the measurement data 41 for a number of consecutive blocks 27 of samples. The firing detection module 137 reports possible inhaler firings to the firing timing detection module 139, which processes the reports to determine the actual timing of the firing during the inhalation. In this way, spurious firing reports can be ignored and an accurate determination can be made as to exactly when the firing occurred during the inhalation.

The firing timing determination module 139 uses the time stamp associated with the threshold results corresponding to the firing report to determine the timing during the inhalation that the firing occurred. Figure 8 shows the determined inhalation profile 44 and three peaks 43-1 , 43-2 and 43-3. These peaks 43 correspond to possible timings when the inhaler 3 fires. Ideally, the inhaler 3 should be fired just before the time at which the flow rate of the inhalation peaks at time ti . Therefore, if the inhaler 3 is fired at the time corresponding to peak 43-1 then this is too early in the inhalation and may result in improper delivery of the medicament. Similarly, if the inhaler 3 is fired at the time corresponding to the peak 43-3, then this is at a time well after the peak flow rate of the inhalation has been achieved and this may also result in the incorrect delivery of the medicament to the user. Therefore, in this embodiment, the firing timing determination module 139 passes the determined firing timing to the compliance determining module 141 which compares the timing of the inhaler firing with the determined flow profile 44 to determine whether or not the firing has occurred too early or too late in the inhalation or at the perfect timing, just before the peak of the inhalation flow profile 44. In this embodiment, the compliance determining module does this by integrating the flow profile 44 over the duration of the inhalation to determine the total displaced volume of air drawn by the inhalation and determines the ratio of the air drawn before the inhaler fired to the air drawn after the inhaler fired. If the ratio is above a first threshold, then the firing was too late and if the ratio is below a second lower threshold, then the firing was too early. If the ratio is between the two thresholds, then the compliance determining module 141 determines that the firing occurred at the correct timing. The compliance determining module 141 then sends a compliance indication back to the user device 7 for display on the display 81 or for transmission back to the circuitry on the circuit board 13, indicating if the firing was at the correct timing.

Alternatively, the compliance determining module 141 can simply compare the determined firing timing (relative to the flow profile 44) with stored data (defining a desired or optimum firing timing) and compliance or non-compliance is determined based on the comparison result. The user device communications module 133 may also return the determined flow profile 44 for the inhalation and the determined firing timing relative to the flow profile, to the user device 7 for display on the display 81. In this way, the user can see the determined flow profile for their inhalation together with the firing timing. This information may be displayed together with a desired flow profile and a desired firing timing in order to give the user feedback on the correct usage of the inhaler 3.

Modifications and alternative embodiments

An embodiment has been described above that illustrates the way in which signals obtained from a microphone may be processed to determine various operational events during the use of an inhaler device. Various alternatives and modifications can be made to this embodiment and a number of these will now be described.

Figure 9 illustrates alternative processing circuitry that can be used with the inhaler device 3 shown in Figure 1. As can be seen by comparing Figure 9 with Figure 2, the main difference, in this embodiment, is that the band pass filter 31 is replaced with a cosine transform module 61. This cosine transform module 61 is programmed to calculate the cosine transform of each block 27 of samples at the characteristic frequency of the inhaler 3. With the inhaler illustrated in Figure 1 , the characteristic frequency is 1.6 kHz and therefore, the cosine transform module 61 only needs to calculate the cosine transform at this frequency. The output from the cosine transform module 61 represents the amplitude of the signal at the characteristic frequency. This amplitude value is then passed to the thresholding module 37 as before. The inhaler 3 described above was an MDI type of inhaler. The present invention is also applicable to other types of inhalers, such as dry powder inhalers (DPIs). DPI inhalers typically have a swirl chamber that has a plurality of tangential inlets through which air is drawn when a user inhales through a mouthpiece of the inhaler. During the inhalation process, a breath actuation mechanism (BAM) is activated which releases the medicament into one of the inlets and the active drug particles are de-agglomerated from carriers (usually lactose) to create a free vortex within the swirl chamber. This swirling airflow is then concentrated through a smaller outlet in the mouthpiece, increasing the tangential velocity of the airflow. This highly swirling airflow through the inhaler produces a dominant acoustic frequency which is dependent upon the volumetric flow rate. Therefore, it is possible to determine the volumetric flow rate and other parameters (as will be described below) by performing a tonal analysis of the sound made by the inhaler 3.

Figure 10 illustrates the circuitry coupled to or mounted on the circuit board 13 that can be used with a DPI type of inhaler to collect measurements that allow the remote server 1 1 to determine the volumetric flow rate through the inhaler and to detect events such actuation of the breath actuation mechanism (BAM) of the inhaler. As shown in Figure 10, the acoustic signal picked up by the microphone 18 is converted into digital data by the analogue to digital converter 21 and the samples are then divided into blocks 27 of samples by the windowing function 25 (as per the first embodiment). In this embodiment, each block 27 of samples is then passed to a Fast Fourier Transform (FFT) module 173 which performs a Fast Fourier Transform on each block 27 of samples individually. Figure 1 1 illustrates a typical FFT spectrum obtained by the FFT module 173. As shown, the spectrum for a block 27 of samples includes a dominant frequency component 175 resulting from the swirling airflow as well as other harmonic components 177. As explained in WO201 1/135353, it has been found that for DPI type inhalers, the peak frequency varies approximately linearly with the flow rate through the inhaler 3.

Therefore, as illustrated in Figure 10, the spectrum obtained from the FFT module 173 is input to a maximum detector 181 which identifies the frequency corresponding to the peak 175 in the spectrum. The determined frequency is then passed to the controller 35 for subsequent transmission to the user device 7.

As explained in WO201 1/135353, a breath actuation mechanism (BAM) triggers when the user inhales through the mouthpiece, releasing the powdered medicament into the swirling airflow. The actuation of the BAM does not affect the tonal characteristics of the spectrum obtained from the FFT module 173. However, the actuation of the BAM produces a loud "click" sound which is more predominant in the higher frequencies of the spectrum. Therefore, in this embodiment, the spectral output from the FFT module 173 is also input to a mean signal level detector 85 which determines the mean signal level of the spectrum in the higher frequency range (for example above 15kHz). The determined mean signal level is then passed to the controller 35 for subsequent transmission to the user device 7. Once the peak frequency and the mean signal level have been determined for a block 27 of samples and passed to the controller 35, the controller 35 generates measurement data 41 ' comprising a header (with identification data identifying the processor 23 - and hence the type of inhaler 3), the peak frequency, the mean signal level and a time stamp and outputs the measurement data 41 ' to the short range communications circuit 42 for transmission to the user device 7. The user device 7 transmits the measurement data 41 ' on to the remote computer server 1 1 , which processes the measurement data. In particular, the remote computer 1 1 has a peak frequency to flow function module (not shown) which converts the peak frequency into a corresponding flow rate. This may be achieved using a look up table or using an equation. As before, this look-up-table or equation may be determined from suitable calibration of the inhaler 3 by passing known flow rates through the inhaler 3 and obtaining corresponding measurement data 41 ', as per the first embodiment described above. The remote server 1 1 can then use the determined flow rates to determine the overall flow profile for the inhalation. This flow profile can then be used to train the user to use the inhaler 3 correctly as per the first embodiment described above. The desired flow profile for a DPI inhaler is normally different to the desired flow profile for an MDI type of inhaler. This is because, the energy within the user's inhalation is used to de-agglomerate the medicament and so a shorter and stronger inhalation is typically required. A typical inhalation flow profile 80 for a DPI type of inhaler is illustrated in Figure 12. Figure 12 also shows possible timings for the firing of the BAM mechanism with the peaks 43. Ideally, the firing will occur just before the peak inhalation corresponding to peak 43-2 in Figure 12. The remote server 1 1 processes the received mean signal level for a block of samples to determine whether or not the BAM has been activated during that block of samples. This can be determined using a number of different techniques. For example, the remote server 1 1 can consider the change in the mean signal level from one block of samples to the next. If the change in the mean signal level exceeds a predetermined threshold value (determined in advance during a calibration routine for the particular type of inhaler), then the remote server 1 1 can infer that the BAM has been activated in the time period corresponding to the current block 27 of samples being processed. It is important to consider the difference between the mean signal levels in adjacent (or near adjacent) blocks 27 of samples in order to take into account the variation in the mean signal levels caused by the variation in the flow rates during the normal inhalation. In an alternative technique, calibration data may be stored in the remote computer server 1 1 identifying typical mean signal levels for different flow rates. In this case, the remote server 1 1 can use the flow rate determined by the peak frequency to flow function for the current block 27 of samples, to determine from the calibration data what the corresponding mean signal level is for this flow rate. The remote server 1 1 can then compare this calibration mean signal level with the mean signal level obtained in the received measurement data 41 '. If the mean signal level obtained from the received measurement data 41 ' exceeds the calibration mean signal level by more than a predetermined amount, then the remote server 1 1 can infer that the BAM has been actuated in the current block 27 of samples.

A different sound is produced when the BAM is activated and dry powder is released into the swirl chamber than when the BAM is activated and no powder is released into the swirl chamber. In particular, when powder is released into the swirl chamber upon activation of the BAM, the peak frequency 175 described above temporarily drops in frequency. This is because the addition of the powder adds to the mass of the swirling air, which reduces the acoustic frequency of the swirl. However, this actually makes it easier to detect the activation of the BAM using the second method described above as the reduction in the peak frequency will have the effect of lowering the flow rate determined by the peak frequency to flow function, which will in turn equate to a lower mean signal level determined using the calibration data. Therefore, by using different thresholds, the remote server 1 1 is able to distinguish between the situation where the BAM is activated and no dry powder is released into the swirl chamber and the situation where the BAM is activated and dry powder is released into the swirl chamber. Consequently, the remote server 1 1 is able to detect the misuse or misfiring of the inhaler 3 and transmit a warning back to the user device 7 or back to the circuitry on the inhaler 3 to warn the user (via the display 81 or the user interface 45).

In the DPI alternative described above, the processing circuitry was arranged to detect the dominant frequency component in the acoustic signal. This was achieved by performing a Fast Fourier Transform of the signal obtained from the microphone. In an alternative embodiment, the dominant frequency component may be determined using time domain techniques, for example by detecting zero crossings of the microphone's signal or by using banks of band pass filters and comparison circuits. ln the above embodiment, the band pass filter 29 is arranged to pass frequencies between 3Hz and 10kHz and to block other frequency components outside this range. Filters with other pass bands could of course be used, depending on where the useful information in the received signal is located within the frequency spectrum.

In the above embodiments, the inhaler circuitry performed some processing of the raw sensor samples obtained from the analog to digital converter. The amount of processing performed by the inhaler circuitry can vary. For example, in one embodiment, the raw sensor samples could be packetized and sent to the remote server 1 1 via the user device 7; or additional processing may be performed in the inhaler circuitry, such as the conversion of the energy value into a flow rate value before the flow rate value is then packetized and sent to the remote server 1 1 for compliance determination and data aggregation from multiple inhalers.

In the above embodiments, the remote computer server 1 1 was arranged to detect the flow profile of the airflow drawn through the inhaler during an inhalation. As those skilled in the art will appreciate, other characteristics and metrics may be calculated. For example, the remote computer server 1 1 may be arranged to calculate the inhaled volume, the peak inspiratory flow rate, the maximum lung capacity, the rate of change of inspiratory flow rate, the inhalation duration, the sustained average flow rate etc. Logging parameters such as these throughout a treatment period may provide valuable information about the efficacy of the treatment. Further, if for example, the peak inspiratory flow rate suddenly decreases during a treatment period, the remote server 1 1 can inform a doctor of the problem. As a further example, with DPI type inhalers, where the energy in the user's inhalation is used to aerosolise the medicament, an important parameter is the rate of change of the flow rate. An inhalation that has a large rate of change of flow rate before the peak inhalation is better for aerosolising the medicament that one having a low rate of change of flow rate. Therefore, measuring the rate of change of the flow rate can also be used to determine if the inhaler is used correctly.

In the DPI embodiment described above, the flow rate was determined by determining the dominant frequency and relating this through stored calibration data to the flow rate. As the harmonics of the dominant frequency also vary with the flow rate, the processing electronics could be arranged to identify one or more of the harmonics as well as or instead of the dominant frequency, and use these to determine the flow rate. As an improvement to the DPI embodiment, the energy in the acoustic signal may also (or instead) be measured by the circuitry mounted on the inhaler and reported as part of the measurement data 41 '. This allows the remote server 1 1 to determine a coarse measure of the flow rate (for example using the technique used in the first embodiment). In particular, whilst the tonal analysis described above provides an accurate measure of the flow rate, it is most accurate for flow rates above about 20 l/min. Therefore, for lower flow rates, the flow rate can be determined using the energy in the acoustic signal. The energy signal may also be used as a check when the BAM actuates. In particular, as discussed above, when the BAM activates and medicament is added to the swirling air flow, this reduces the peak frequency which reduces the calculated flow rate. However, it is also possible that the user hiccupped during the inhalation and this is what caused the dip in the measured flow rate. By considering the measured flow rate using the energy measure, the remote server 1 1 can distinguish between a drop in measured flow rate caused by a hiccup and a drop in measured flow rate caused by the firing of the BAM mechanism. In the embodiments described above, the remote server 1 1 determined the flow profile for the inhalation. In other embodiments, the remote server 1 1 may only determine the flow profile for a part of the inhalation - for example the initial part until the drug has been released and the peak flow rate etc has been calculated. In the above embodiments, the remote server 1 1 is able to process the measurement data 41 obtained from the inhaler and detect if the delivery mechanism is activated during the inhalation and, if it is, to detect if the drug is also delivered by the mechanism. The remote server 1 1 may maintain a count of the number of times that the delivery device is activated and the drug is successfully delivered and the number of times that the delivery device is activated but no drug is delivered. This information may be useful for subsequent diagnosis by the clinician or physician. Additionally, the remote server 1 1 can report this information back to the user device to give substantially real time feedback to the user so that they know if the drug was actually delivered. Very often with inhaler devices, users take too much of the drug because they do not realise that the drug is dispensed during one or more of their inhalations.

Some inhaler devices that are provided include a mechanical counter that increments each time the drug is dispensed. Such mechanical counters typically make a clicking sound when they change value and this can also be detected by the remote server 1 1.

Some dry powder inhalers use capsules to store the drug and the capsules have a foil which has to be pierced before the drug can be delivered. In such embodiments, the sound made by the piercing of the foil may be detected by the microphone and the remote server may detect the time when the foil is pierced. This time may then be recorded together with the time that the drug is subsequently delivered so that information about the time gap between piercing the foil and drug delivery can be determined. This information may be important, for example, if it is known that the drug deteriorates once the foil has been pierced and the drug is open in contact with the atmosphere.

In the embodiments described above, the measurement data was transmitted in an uncompressed format. If desired, the measurement data may be compressed or encrypted before being transmitted. In this case, the controller 35 would have a compression module and/or an encryption module for compressing and/or encrypting the measurement data.

In the above embodiments, the circuitry mounted in the sleeve of the inhaler performed some limited processing of the data obtained from the microphone. As those skilled in the art will appreciate, the data obtained directly from the microphone (or other sensor) may be transmitted directly to the user device for onward transmission to the remote server for processing.

In the above embodiments, the electronics and microphone were mounted within a sleeve that was detachable from the inhaler body. The sleeve was a two part sleeve. As those skilled in the art will appreciate, the sleeve could have more than just two parts. Additionally, the sleeve could just have a single part - such as a "C" shaped sleeve that clips on to the side of the inhaler body. The use of such a sleeve is preferred as it allows the electronics to be used with different inhalers and to be used with standard inhalers - as not all inhalers will be provided with the sensing circuitry (due to cost).

In the above embodiments, the remote server processed all the measurement data received from the inhaler. The remote server may be a cloud based server or group of servers, such that a different processing of the measurement data may be performed by a different processor in the cloud.

As the remote server will receive measurement data from many different inhalers being used by many different users, the remote server can aggregate the compliance data from all the different users to determine general information about the different types of inhalers - such as general ease of use of the inhaler and the like. For example, if it is found that all users struggle to comply with the desired operating parameters of the inhaler, then this can provide valuable feedback to a product manufacturer that that particular inhaler is difficult to use.

In the above embodiments, a microphone was used that detected acoustic signals. Any microphone or sensor can be used that can detect any kind of pressure wave or vibration. Additionally, an ultrasound type of sensor could also be used that has an ultrasound emitter and receiver and that determines the time of flight of the sound signal from the emitter to the receiver. This time of flight measurement will depend on the flow rate through the inhaler. In this case, the inhaler circuitry would, for example, calculate the time of flight through the inhaler once every 50ms or so and would report this time of flight information in a time stamped packet to the remote server via the user device as before. The remote server can then process the time of flight information and determine a flow profile in a similar manner to the first embodiment described above. Introduction of the drug into the inhaler (either by canister firing or through BAM activation) can then be determined either by detecting a reduction in the detected signal level or by detecting a Doppler shift (change in frequency of the received ultrasound signal) due to scattering from the drug particles. Additional measurement data may be generated by the inhaler circuitry, such as temperature data that can be used by the remote server to select appropriate temperature compensated calibration data that relates the measured time of flight information to flow rates. Rather than having just one ultrasonic transmitter and receiver, two transmitters/receivers may be provided - one for transmitting an ultrasound signal "upstream" against the flow caused by the user's inhalation and the other for transmitting an ultrasound signal downstream so that it travels along the same direction as the flow caused by the user's inhalation. The difference between the two time of flight measurements can then be used to provide a more accurate measurement of the flow through the inhaler. Of course, other signal parameters of the received ultrasonic signals may be measured and transmitted to the remote server from which time of flight information and hence flow rates can be calculated. For example, the signal samples defining the received signal waveform or the temporal locations of the zero-crossings of the received waveform may be packetized and transmitted to the remote server via the user device.

In summary, the invention applies to all kinds of inhaler (e.g. pressurised metered dose inhalers, breath-actuated inhalers, nebulisers, capsule inhalers, reservoir inhalers, etc.) and embodiments can include the following elements: 1. An embedded sensing technology capable of detecting and quantifying the

manoeuvre and operation of the inhaler. Such sensor technologies include, but are not limited to:

• Acoustic microphone

• Vibration microphone

· Pressure sensor

• Ultrasound time of flight emitter and receiver

These sensors should not interfere with the air and drug flow through the inhaler and should ideally be completely non-invasive - i.e. should be mounted on the external surface of the inhaler caseworks rather than come into contact with the air and drug flow.

The raw data collected by the sensor should be capable of being converted into key parameters such as:

• Air flow rate

• Peak inspirational flow

· Flow duration

• Triggering events

• Acceleration

• Etc.

Electronic hardware and software located on the inhaler the purpose of which is to:

• Power and control the embedded sensor

• Collect data in real time and time stamp the data by means of an internal clock (for event sensing) • Optionally, carry out compression of the data

• Optionally, encrypt the data

• Optionally, carry out calculations on the data

• Optionally, present the data back to the patient on the inhaler by means of a simple user interface (e.g. LED traffic light indicator)

• Package data for wireless transmission

• Optionally, change the flow characteristics of the inhaler in some way so as to optimise patient compliance. For example, a message may be sent back to the user device 7 or to the inhaler 3 indicating that a flow restrictor of the inhaler 3 should be changed in order to increase or decrease the resistance to flow - which might help to correct for non-compliance due to excessively high flow rate.

• Receive incoming data from the user device (see element #4)

Wireless data link between inhaler and user device

• The wireless protocol used will depend on what data is sent. If the raw

sensor samples are processed in the inhaler circuitry to determine energy or flow rate measures, then low energy Bluetooth is preferably used for real time data streaming at up to 6 kBytes/s. If the raw sensor samples are transmitted, then a protocol with a higher data throughput is required, such as Bluetooth.

User device - typically a smart phone (e.g. Apple or Android)

• Configured to receive low energy Bluetooth data

• Hosting a bespoke inhaler application for processing and displaying data to the patient (e.g. flow profile)

• Optionally providing feedback to patient to indicate success or otherwise of the inhalation or to assist with training

Wireless data link between the user device and the remote server

• Preferably 3G, 4G, LAN is used for non-real time relaying of data to the

remote server or cloud

Remote server or cloud application

• Capable of carrying out complex calculations on the measurement data

• The results of said calculations may be used to:

• Inform the patient and/or the patient's physician about levels of compliance over a course of treatment (by sending data/messages back to the user device or to a hospital or clinic information system)

• Aggregate data over multiple patients for the purpose of tracking product performance These and various other modifications and alternatives will be apparent to those skilled in the art.

Claims

Claims:

1. A drug delivery system comprising:

a drug delivery device comprising:

a body with a mouthpiece;

a sensor for sensing pressure or vibration waves during operation of the drug delivery device;

processing circuitry operable to process a signal obtained from the sensor to generate measurement data; and

short range communication circuitry for transmitting said measurement data from the processing circuitry;

a user communication device comprising:

short range communication circuitry for receiving the measurement data transmitted from the short range communication circuitry of the drug delivery device;

network transceiver circuitry for transmitting the measurement data to a communications network and for receiving signals from the communications network; and

a user interface for outputting information to a user; and

a remote server comprising:

transceiver circuitry for receiving the measurement data from the user communication device via the communications network and for transmitting data back to the user device via the communications network;

processing circuitry for processing the measurement data and for determining operating parameters of the drug delivery device, the operating parameters including compliance data indicating compliance of the operation of the drug delivery device; and

a controller for outputting the compliance data to the transceiver circuitry for transmission back to the user communication device; wherein the user communication device is operable to receive the compliance data from the remote server and is operable to perform at least one of: outputting the compliance data to the user via the user interface and transmitting the compliance data back to the drug delivery device using the short range communication circuitry.

2. A system according to claim 1 , wherein the processing circuitry of the remote server is operable to determine, using said measurement data, a flow profile during use of the drug delivery device and wherein the remote server is operable to transmit the determined flow profile back to the user device for display to the user via a display of the user device.

3. A system according to claim 1 or 2, wherein said processing circuitry of the drug delivery device is operable to compress and/or encrypt the measurement data before said short range communication circuitry transmits the measurement data to the user device.

4. A system according to claim 1 , 2 or 3, wherein said processing circuitry of the drug delivery device is operable to divide the signal obtained from the sensor into a sequence of blocks of signal samples and is operable to determine an energy value of the signal within each block of samples and wherein said measurement data comprises said energy values and a time stamp associated with the corresponding block of samples.

5. A system according to any of claims 1 to 4, wherein the measurement data includes identification data identifying a type of the drug delivery device and wherein the remote server is operable to use the identification data in the received measurement data to select predefined calibration data for use in processing the measurement data to determine said compliance data.

6. A system according to any of claims 1 to 5, wherein the processing circuitry of the drug delivery device is arranged to packetize signal samples obtained from the sensor to form said measurement data.

7. A system according to any of claims 1 to 5, wherein the sensor comprises a microphone or an ultrasonic sensor.

8. A drug delivery method comprising:

providing a drug delivery device having a body with a mouthpiece and processing circuitry;

sensing pressure or vibration waves at the drug delivery device during operation of the drug delivery device; using the processing circuitry to process a signal obtained from the sensing to generate measurement data; and

using short range communication circuitry to transmit said measurement data from the processing circuitry;

receiving, using short range communication circuitry of a user communication device, the measurement data transmitted from the short range communication circuitry of the drug delivery device;

transmitting the measurement data from the user communication device to a communications network and for receiving signals from the communications network; and receiving, at a remote server, the measurement data transmitted from the user communication device via the communications network

processing the measurement data at the remote server to determine operating parameters of the drug delivery device, the operating parameters including compliance data indicating compliance of the operation of the drug delivery device; and

transmitting the compliance data from the remote server back to the user communication device;

receiving the compliance data from the remote server; and

performing at least one of: outputting the compliance data to the user via a user interface of the user device and transmitting the compliance data back to the drug delivery device using the short range communication circuitry.