CN111511439A

CN111511439A - Thermal therapy with dynamic anatomical boundaries using MRI-based temperature uncertainty maps

Info

Publication number: CN111511439A
Application number: CN201780097974.8A
Authority: CN
Inventors: A·比戈; P·伦纳德; R·库尔茨
Original assignee: Bofang Medical Co ltd
Current assignee: Bofang Medical Co ltd
Priority date: 2017-10-30
Filing date: 2017-10-30
Publication date: 2020-08-07
Also published as: CA3080435A1; WO2019086921A1

Abstract

The temperature uncertainty map is calculated based on a rolling window of temperature maps, which is updated when a new temperature map is generated. The rolling window mitigates the effects of transient motion during the course of thermal therapy. The clinician or automated control system may then update a portion of the anatomical boundary or the thermal therapy applicator center based on the temperature uncertainty map.

Description

Thermal therapy with dynamic anatomical boundaries using MRI-based temperature uncertainty maps

Technical Field

The present invention relates to thermal therapy delivered by a treatment device to target tissue within an anatomical boundary based on a dynamic thermal uncertainty map derived from an MRI thermometry system and data.

Background

The use of Magnetic Resonance Imaging (MRI) to obtain temperature-related data during tissue ablation is discussed, for example, in Chopra (U.S. patent No.7,771,418), which is hereby incorporated by reference. MRI thermometry, the resulting temperature measurements and their temperature uncertainty maps, and related considerations are discussed by the present applicant, for example, in published application US2015/0038883a1, which is also incorporated herein by reference.

In general, temperature measurements using MRI methods are susceptible to errors from various sources known to those skilled in the art. When temperature measurements are used as part of a feedback system for thermal energy delivery, these errors result in accidental heating or lack of heating of the target area. Errors in temperature measurement during treatment using MRI methods include transient motion, such as global patient motion, localized prostate motion (e.g., due to muscle or nerve heating), and/or rectal displacement. For example, transient motion can cause significant errors in temperature measurement that are currently addressed by waiting (e.g., 20 minutes) for the measured body temperature to return to a substantially constant value. This results in a less than ideal treatment course from a patient comfort perspective, as well as reduced patient turnover, or less economical use of MRI thermal therapy treatment facilities, personnel, and equipment.

Disclosure of Invention

The method described herein calculates and displays areas where temperature can be reliably measured. The clinician may then make informed decisions to treat these areas or plan treatments to avoid them based on the sensitivity of surrounding structures to accidental heating.

One aspect of the invention is directed to a method for dynamically delivering thermal therapy to a target volume within a body of a patient. The method comprises the following steps: determining an anatomical boundary corresponding to the target volume for delivering thermal therapy to the target volume; delivering a thermotherapy dose to the target volume using a thermotherapy applicator comprising an ultrasound transducer array; receiving, in a computer, N sets of temperature data for pixels corresponding to a portion of a patient's body, each set of temperature data corresponding to a respective capture time of a phase image captured using a Magnetic Resonance Imaging (MRI) device, wherein N is greater than or equal to M, and M is a rolling capture time window; determining, in the computer, for each of the past M capture times, a corrected temperature at each pixel; calculating, in the computer, for each pixel, a temperature uncertainty based on the corrected temperature at each of the past M capture times; and in the computer, modifying a portion of the anatomical boundary only if a temperature uncertainty of the portion is below a threshold temperature uncertainty.

In one or more embodiments, the temperature uncertainty corresponds to a standard deviation of the corrected temperature at each pixel over the past M capture times. In one or more embodiments, the method further comprises suspending delivery of the thermal therapy dose prior to modifying the portion of the anatomical boundary. In one or more embodiments, the method further comprises modifying the position of the center of the heat therapy applicator.

In one or more embodiments, the method further comprises, in the computer, validating the anatomical boundary to confirm that a temperature uncertainty of the portion of the anatomical boundary is below a threshold temperature uncertainty. In one or more embodiments, the method further comprises generating, in the computer, an alert when the temperature uncertainty of the portion of the anatomical boundary is greater than a threshold temperature uncertainty.

In one or more embodiments, the method further comprises calculating, in the computer, a standard deviation at each point along the anatomical boundary over the past M capture times. In one or more embodiments, the method further comprises generating, in the computer, a temperature uncertainty map comprising the temperature uncertainty for each pixel. In one or more embodiments, the method further comprises displaying the temperature uncertainty map on a display coupled to the computer.

In one or more embodiments, the method further includes detrending the corrected temperatures at each pixel over the past M capture times to form detrended temperature data. In one or more embodiments, the method further comprises performing a linear regression of the corrected temperature at each pixel over the past M capture times. In one or more embodiments, the method further comprises calculating a standard deviation of the detrended temperature data at each pixel. In one or more embodiments, the method further includes determining a temperature uncertainty based on a standard deviation of the de-trended temperature data at each pixel.

In one or more embodiments, the method further comprises: receiving, in the computer, a new set of temperature data for pixels corresponding to the portion of the patient's body; and calculating an updated temperature uncertainty based on the past M capture times, the past M capture times including the new set of temperature data.

Drawings

For a fuller understanding of the nature and advantages of the present invention, reference should be made to the following detailed description of the preferred embodiment taken in conjunction with the accompanying drawings in which:

FIG. 1 illustrates a representation of a cross-section of an MRI temperature uncertainty map showing prostate and target boundaries;

FIG. 2 illustrates an exemplary treatment workflow process;

FIG. 3 illustrates an exemplary process for calculating a temperature uncertainty map;

FIG. 4 is a flow chart of a temperature uncertainty map for dynamically calculating the temperature in the target volume;

5A, 5B, and 5C illustrate examples of temperature uncertainty maps that may be generated in accordance with the flowchart of FIG. 4;

FIG. 6 is a flow chart of a method for updating a prostate boundary;

fig. 7, 8, 9 and 10 illustrate a flow chart of a temperature uncertainty map for dynamically calculating the temperature in the target volume;

FIG. 11 is a graph illustrating the effect of detrending temperature data; and is

Fig. 12 illustrates an example of a coordinate system used in some embodiments.

Detailed Description

The present disclosure provides systems and methods for overcoming the effects of such temperature measurement uncertainties and avoiding errors due to such temperature uncertainties. Thus, improved accuracy and efficiency of delivery of MRI-guided thermal therapy becomes possible. One application for such therapy is the treatment of diseased male prostate.

Embodiments of the present invention relate to dynamically changing and verifying the prostate contour and ultrasound applicator center during treatment. During treatment, the prostate profile and/or applicator center may need to be adjusted (manually or automatically) due to transient motion that may invalidate baseline treatment parameters (e.g., prostate boundary and ultrasound applicator center). Examples of transient motion include global patient motion, localized prostate motion (e.g., due to muscle or nerve heating), and/or rectal displacement. The prostate contour may also need to be adjusted if the noise corrupts some sections of the boundary. For example, there may be areas of low signal due to gas in the rectum or due to transient motion. Furthermore, the prostate profile may need to be adjusted to avoid treatment of the region (e.g., the zone is treated once, and treatment again is undesirable). The ultrasound applicator center may need to be adjusted because the alignment of the ultrasound applicator center is in the treatment plan or because the transient motion is wrong.

To account for transient motion, the temperature and temporal temperature uncertainty at each pixel is calculated retrospectively at a given data capture time over a rolling time window during treatment.

Fig. 1 illustrates a cross-sectional view taken using an imaging modality, such as MRI imaging of a portion of a patient's body, in the vicinity of a treatment target volume. The illustrated scene includes, for example, a visual output device (such as a computer monitor screen 10) or an application window of a computer application program for displaying the image 12. The surface of the patient's body (e.g., the surface of his abdomen) is shown at 110, while various regions 102 in the patient's body are shown with visual representations of their temperature and/or temperature uncertainty within the image 12. The area 102 may be displayed on the screen 10 as a colored outline, a silhouette plot, a gray scale intensity, or other visual representation of temperature uncertainty. The plotted and represented values are determined as described below.

The image 12 shows the boundary of a target volume, such as a male prostate or a portion 120 thereof. This is a contour line on the image 12 which may be computer drawn or drawn with the aid of an operator on the screen 10. The treatment target boundary 100 is further shown on the image 12, which may be an outline of another color, a dashed outline, or other representation. The target boundary 100 is the intended boundary within which the energy of the thermal treatment process is substantially controlled to a set point temperature (or thermal dose) ensuring rapid and sufficient cell death of the diseased cells within the interior of the volume defined by the target boundary 100. Heat can be conducted outside of the target boundary 100 to the boundary of the prostate 120, which can be measured and controlled to achieve proper thermal therapy while reasonably avoiding damage to non-diseased tissue and organs adjacent to the site of the lesion. Tissues and organs outside the target boundary will not exceed the lethal thermal dose or temperature limit, even if heated.

Methods for determining and controlling the intensity of a thermal therapy treatment as a function of the temperature or desired temperature at such boundary 100 are described by the present inventors in publicly available publications and patent applications, which are hereby incorporated by reference.

In summary, fig. 1 thus shows a temperature uncertainty diagram. The three-dimensional representation of the temperature uncertainty map may be constructed from additional layers, slices, or cross-sectional views (such as the cross-sectional view shown in fig. 1). Without loss of generality, the methods described herein may thus be generalized to three-dimensional space by stacking slices such as shown in fig. 1 side-by-side to form a 3D volume.

Fig. 2 illustrates an exemplary process 20, the exemplary process 20 enabling thermal treatment in an MRI-guided environment, and taking into account temperature uncertainty in the MRI thermometry portion of the process. The process begins at 200 and at step 202, automated or operator driven positioning of a thermal therapy device in or on a patient is performed. In an example, an ultrasound (u/s) heat therapy applicator is inserted transurethrally into a diseased male prostate organ and positioned to deliver heat therapy to the diseased organ. In another aspect, a patient is placed in an MRI imaging volume or machine bore (bore) and a temperature scan using MRI thermometry is acquired slice-by-slice through a target region to generate a thermal image and/or a temperature uncertainty map of the target region.

An anatomical image of the patient or a portion of the patient in the vicinity of the target region is obtained at step 204. At 206, the system may automatically or semi-automatically determine whether the heat therapy applicator is in the correct position to deliver the desired heat therapy to the target area. If the thermotherapy applicator is not in the correct position, the process returns to positioning the thermotherapy applicator at 202.

Once the thermal therapy applicator device is in the correct position, temperature uncertainty images (such as those depicted in fig. 1) are collected at 208. Memory or digital storage may be used to store the data so collected for analysis or other purposes.

The system then calculates and displays a temperature uncertainty map as described above at step 210. These are preferably output to a computer output or display device, such as a computer workstation monitor connected to the imaging and therapy devices throughout the thermal therapy control system.

Using the temperature data and the temperature uncertainty map, at step 212, a thermal therapy treatment plan is determined and a target point or target region is identified.

At step 214, heat therapy itself is delivered from a heat therapy applicator (e.g., an ultrasound transducer array device in or near the desired target area). During thermal therapy, additional temperature uncertainty images are acquired and displayed, as discussed below.

Once the thermal therapy process is complete, the system or operator terminates the process 20 at 216.

Fig. 3 illustrates another set of steps in an exemplary computer-implemented method 30 for acquiring images, making appropriate corrections, and generating outputs for the case of image-guided thermal therapy.

The process starts at 300 and one or more phase images are acquired from a magnetic resonance or MRI apparatus in which the patient is placed. In an embodiment, several (e.g., three to ten) phase images are acquired at step 302 and stored in a machine readable storage device, such as a computer memory device. The MRI apparatus may be configured, arranged, programmed and operated as a run sequence to output magnitude and phase images in real time. The output image is output as required, either via a signal connection or a network connection, for example to a further computer device coupled to the MRI device, where subsequent calculations and processing of the MRI data may be performed.

In an example, an EPI sequence is used to acquire channel uncombined phase images. Other sequences may be used as will be understood by those skilled in the art, for example, GRE sequences.

In some thermal therapies using an ultrasound transducer system, a plurality of ultrasound transducer elements are deployed in an ultrasound array placed within a diseased tissue volume. For a multi-transducer ultrasound therapy system, multiple image slices may be acquired such that one image slice is acquired by each ultrasound transducer, each therapy applicator system. In yet another aspect, monitoring slice images may be acquired at either end of the imaging slice for adequate monitoring. The sequence is in embodiments arranged to be repeated automatically such that a stack of phase images is generated continuously throughout the thermotherapy treatment.

In step 304, a reference phase image is created using data from the phase image acquired in the previous step. The reference phase image is the phase image before heating is initiated from the thermal therapy process. To increase the signal-to-noise ratio, the reference phase image is calculated as the average phase over several (e.g., 5) reference images for each pixel in the image.

Before and/or during the thermal therapy session, measurement images are collected at step 306. The system then calculates an uncorrected temperature at step 308. In an example, a weighted sum of the phase differences over all channels is calculated and scaled in order to determine the temperature. In one aspect, the MRI apparatus may be programmed to output a combined phase for all coils. In this case, the system need only calculate the phase difference with the reference image to be scaled to output the temperature in the region of interest.

In step 310, the system corrects for drift. As previously mentioned, drift may be due to time variations or drift in the main B0 magnetic field of the MRI machine. Drift can result in erroneous (typically lower) temperature measurements, if not corrected. Thus, according to the present aspect, we correct for such drift effects at one or more regions of the image. The temperature at these regions is assumed to be the temperature of the core temperature of the patient's body, which is substantially constant throughout the therapy treatment. A two-dimensional linear interpolation of drift is calculated for each measured slice image and the interpolation is added to the temperature at each pixel in the image to generate a drift-corrected temperature image.

In step 312, the visual temperature map is displayed on a display coupled to the computer.

Fig. 4 is a flow chart 40 for dynamically calculating a temperature uncertainty map of the temperature in the target volume. In step 400, heat therapy is delivered from a heat therapy applicator (e.g., an ultrasound transducer array device in or near a desired target area), as discussed above. Thermal therapy may be delivered using, for example, a treatment regimen as discussed above with respect to fig. 2. In step 410, MRI phase images are collected from the MRI apparatus during a collection period (e.g., dynamic). The dynamic or collection time period may be based on time (e.g., 3 to 5 seconds) and/or the number of collected phase images (e.g., 25 to 50 phase images). In step 420, the corrected temperature at each pixel is determined by calculating the phase difference between (a) the average phase (average measured phase) during the phase image collection period and (b) the average phase (e.g., phase as discussed above) over several (e.g., 5) reference images for each pixel in the image, and then correcting the drift, similar to the manner described in fig. 3. In step 430, a temperature map is generated and optionally displayed to a user, for example, as discussed above with respect to fig. 3.

In step 440, the computer determines the number of temperature maps stored in memory. If the number of temperature maps (N) is less than M, the flow chart returns to step 410 to collect additional MRI phase images (and generate corresponding temperature maps) during additional collection periods. This process repeats until N is greater than or equal to M, where M is a rolling window of the temperature map used to calculate the temperature uncertainty map, as discussed below. Thus, M is an integer greater than or equal to 2, and preferably at least 5.

When N is greater than or equal to M, the flowchart 40 continues to step 450 where a time temperature uncertainty map is calculated in step 450. The time temperature uncertainty map is formed by calculating the standard deviation of the temperature at each pixel over the last M temperature maps. For example, if there are 10 temperature maps (N-10) and the rolling window of the temperature maps is 5 (M-5), only the last 5 temperature maps are used to calculate the time temperature uncertainty map. Alternatively, each of the past temperature maps is used based on a weighted average, with the most recent temperature map having a higher weight than the older temperature map.

In step 460, the time temperature uncertainty map is visually displayed on a display coupled to the computer. The time temperature uncertainty map may be color coded according to different temperature uncertainty ranges. For example, shades of blue may be assigned a temperature uncertainty below a first threshold (e.g., less than 2℃.), shades of yellow and red are used for temperature uncertainties between the first threshold and a second threshold (e.g., between 2-4℃.), and shades of purple are used for temperature uncertainties greater than the second threshold (e.g., greater than 4℃.).

After step 460, the flowchart 40 returns to step 410 to collect additional MRI phase images during the next collection period. In the next iteration through the flowchart 40, a new temperature map (N +1) is generated and a temperature uncertainty map is calculated based on the temperature maps in the current rolling window of the temperature map M. In other words, in the next iteration, the current rolling window of temperature map M includes the newest temperature map (N +1), but does not include the oldest temperature map used in the last iteration. Alternatively, as discussed above, all temperature maps are used based on a weighted average.

In some embodiments, a linear regression is performed on the temperature at each pixel over the rolling window M, which may reduce the effect of heating (or cooling) on the temperature uncertainty map. The de-trended data is then used to calculate a temperature uncertainty map in step 450.

The rolling window M may reduce the effect of transient motion on the temperature uncertainty map. For example, transient motion may cause a shift in temperature in a given temperature map because, for example, the ultrasound applicator center has moved relative to the baseline image. However, the effect of such an offset may be reduced over time by comparing the temperature map of the offset with subsequent temperature maps that may also have an offset in temperature.

Examples of temperature uncertainty maps that may be generated according to flowchart 40 are illustrated in fig. 5A-5C. Fig. 5A illustrates a first temperature uncertainty map 50A corresponding to a first time collection period (e.g., period 10). In the temperature uncertainty map 50A, there are several regions of high temperature uncertainty 500. The remainder of the temperature uncertainty map 50A has a low temperature uncertainty. The region of high temperature uncertainty 500 is disposed outside of the prostate boundary 510 and inside the prostate boundary 510 at the flame 520, the flame 520 corresponding to the heat therapy generated by the applicator 530.

Fig. 5B illustrates a temperature uncertainty map 50B corresponding to a second time collection period (e.g., period 20) that occurs after the transient motion. It can be seen that the region of high temperature uncertainty 500 is greater in the temperature uncertainty map 50B than the temperature uncertainty map 50A. In addition, a region of high temperature uncertainty 500 is disposed adjacent to the prostate boundary 510. The system or operator may modify any location of the prostate boundary 510 or applicator center 530 with computer verification that the modified location is below a threshold (e.g., 2 ℃) as a constraint.

Fig. 5C illustrates a temperature uncertainty map 50C corresponding to a third time collection period (e.g., period 30). It can be seen that the region of high temperature uncertainty 500 decreases in the temperature uncertainty map 50C after a period of time due to the scrolling of the time window M discussed herein.

Fig. 6 is a flow chart 60 of a method for updating prostate boundaries. In step 600, a time temperature uncertainty map is displayed on a display coupled to a computer. In optional step 610, the operator manually or the computer automatically pauses the treatment. The treatment may be suspended, for example, to provide time for additional time collection periods to reduce temperature uncertainty (e.g., as discussed above). In step 620, the operator manually or a computer automatically modifies the position of the prostate boundary and/or the ultrasound applicator center (e.g., to compensate for transient motion). In optional step 630, the operator manually or the computer automatically resumes treatment. In step 640, the computer verifies the new prostate boundary to confirm that the prostate boundary has not been modified at the location of the high temperature uncertainty.

Fig. 7 is a flow chart 70 for dynamically calculating a temperature uncertainty map of the temperature in the target volume. In step 700, heat therapy is delivered from a heat therapy applicator (e.g., an ultrasound transducer array device in or near a desired target area), as discussed above. Thermal therapy may be delivered using, for example, a treatment regimen as discussed above with respect to fig. 2. In step 702, MRI phase images are collected from the MRI apparatus during a collection period (e.g., dynamic). The dynamic or collection time period may be based on time (e.g., 3 to 5 seconds) and/or the number of collected phase images (e.g., 25 to 50 phase images). In step 704, the phase images collected during the dynamic are processed to form a temperature map (e.g., a temperature map as described above with respect to fig. 4). In step 706, the temperature map is stored in a buffer having a width of M temperature maps (corresponding to M dynamics), where M is the temperature map or dynamic rolling window used to calculate the temperature uncertainty map. Thus, M is an integer greater than or equal to 2, and preferably at least 5.

If the number of temperature maps or dynamics (N) is less than or equal to M, the flow chart returns to step 702 to receive additional dynamics and the corresponding temperature map is processed in step 704, which is then added to the buffer in step 706. The process repeats until N is greater than M in step 708.

When N is greater than M, the flowchart 70 continues to step 710 where the oldest temperature map (corresponding to the oldest dynamic) is discarded from the buffer in step 710. Thus, the buffer contains only the last M temperature maps or dynamics. After step 710, flowchart 70 proceeds to placeholder A, which also appears in FIG. 8. Note that the new dynamic acquisition and processing occurs throughout the flowchart 70, so the temperature uncertainty map may be dynamically updated during any step of the flowchart 70.

Starting with placeholder a on fig. 8, flowchart 70 proceeds to step 712 to perform a linear regression (e.g., a first order linear regression) for each pixel, slice, and dynamic in the buffer heap. The first order linear regression may use the formula T_EstimatingCalculated as (x, y, z) — a1(x, y, z) t + b1(x, y, z) +, this formula estimates that the temperature rises for each pixel over the last M dynamics as a linear trend. In this equation, x and y refer to the coordinates of the pixel, z refers to the coordinates (e.g., number of slices) on the dynamic volume made up of N slices, a1 corresponds to the slope of the first order regression, b1 corresponds to the intercept of the first order regression, corresponding to the noise of the data. The coordinates x, y and z are also illustrated in fig. 12.

In step 714, according to the disclosureFormula T_{Trend reduction}(x，y，z)＝T(x，y，z)-T_Estimating(x, y, z) to detrended the data, where T (x, y, z) is the temperature measured by MRI thermometry, T_Estimating(x, y, z) is calculated in step 712. An example of a graph illustrating the effect of de-trending the temperature data is illustrated in FIG. 11, where line 1110 represents measured heating data for a pixel, line 1120 represents a first order fit to line 1110, and line 1130 represents de-trended temperature data from a pixel. It can be seen that line 1130 does not include the heating component of line 1110, thus improving the standard deviation calculation.

In step 716, the standard deviation of the detrended data is calculated for each pixel over the last M dynamics. The standard deviation for each pixel is then displayed as a temperature uncertainty map in step 718.

In step 720, the computer determines whether the user has attempted to modify the prostate boundary or the ultrasound applicator center position. In some embodiments, the prostate boundary may be modified regardless of the temperature uncertainty at a given point or pixel. If so, the flow diagram 70 continues to placeholder B, which also appears in FIG. 9. If not, the flowchart 70 proceeds to step 724 to determine if there are any indications that the prostate boundary may be too uncertain (e.g., due to motion or noise). If so, the flow diagram 70 continues to placeholder B. Additionally, if the system determines in step 724 that there is any indication that the prostate boundary may be too uncertain, the system may trigger an alarm or suspend treatment. If not, the flow diagram 70 continues to placeholder C, which also appears in FIG. 10.

Starting with placeholder B on fig. 9, flowchart 700 continues to step 728 to calculate a standard deviation for each point of the prostate boundary, similar to the manner described above. In step 730, the computer displays (e.g., on a color-coded map) the prostate boundary segment with a high temperature uncertainty (e.g., greater than 2 ℃).

In step 732, the user is allowed to modify any point on the prostate boundary and/or move the ultrasound applicator center. In step 734, the modified section of the prostate boundary and/or the new position of the ultrasound applicator is displayed.

In step 736, the user is asked to confirm the changes made in step 732 (i.e., modifications to the prostate boundary and/or the ultrasound applicator center). If the user does not confirm the change, the flowchart 70 returns to step 702 to receive a new dynamic. If the user confirms the change, the flowchart 70 proceeds to step 738 where the standard deviation of the temperature in the modified section of the prostate boundary is calculated in step 738. After step 738, the flowchart 70 continues to placeholder D, which appears in FIG. 10.

Starting with placeholder D on fig. 10, flowchart 70 proceeds to step 740 to determine whether the standard deviation of each pixel is less than 2 ℃. If so, the controller is updated to use the new prostate boundary and/or the new ultrasound applicator center modified in step 732. If the standard deviation of any pixel is greater than or equal to 2℃ at step 740, then the flowchart 70 determines whether the user has confirmed and acknowledged this large standard deviation at step 742. If the user has confirmed and acknowledged the large standard deviation, the flowchart 70 proceeds to step 744 to update the controller with a new prostate boundary and/or a new ultrasound applicator center, as discussed above. If the user has not confirmed and acknowledged the large standard deviation in step 742, the flowchart 70 returns to placeholder B in FIG. 9, at which point the standard deviation for each point of the prostate boundary is calculated in step 728. If the segment of the modified prostate boundary is too uncertain (i.e., greater than 2 ℃), the user can either (a) wait for the ultrasound applicator beam to pass if the user has modified the segment currently being heated; (b) pause the treatment and wait for the temperature uncertainty map to stabilize; (c) redrawing the prostate boundary to a different location (e.g., to avoid high temperature uncertainty regions); (d) recognizing and confirming that at least some sections of the prostate boundary have high temperature uncertainty; or (d) discard the changes to the prostate boundary and continue using the original prostate boundary.

After the controller is updated in step 744, the flowchart 70 proceeds to step 746 for the controller to perform the thermal therapy treatment based on the new boundary and/or new UA-center (if incoming from step 744) or based on the existing boundary and/or UA-center (if incoming from placeholder C). As discussed above, flowchart 70 also proceeds from placeholder C reached after step 724 to step 746.

Therefore, the present invention should not be considered limited to the particular embodiments described above. Various modifications, equivalent processes, as well as numerous structures to which the present invention may be applicable will be readily apparent to those of skill in the art to which the present invention is directed upon review of the present disclosure.

Claims

1. A method for dynamically delivering thermal therapy to a target volume within a body of a patient, the method comprising:

determining an anatomical boundary corresponding to the target volume to deliver thermal therapy to the target volume;

delivering a thermotherapy dose to the target volume using a thermotherapy applicator comprising an ultrasound transducer array;

receiving, in a computer, N sets of temperature data for pixels corresponding to a portion of a patient's body, each set of temperature data corresponding to a respective capture time of a phase image captured using a Magnetic Resonance Imaging (MRI) device, wherein N is greater than or equal to M, and M is a rolling capture time window;

determining, in the computer, for each of the past M capture times, a corrected temperature at each pixel;

calculating, in the computer, for each pixel, a temperature uncertainty based on the corrected temperature at each of the past M capture times; and

in the computer, a portion of the anatomical boundary is modified only if a temperature uncertainty of the portion is below a threshold temperature uncertainty.

2. The method of claim 1, wherein the temperature uncertainty corresponds to a standard deviation of the corrected temperature at each pixel over the past M capture times.

3. The method of claim 1, further comprising pausing delivery of the thermal therapy dose prior to modifying the portion of the anatomical boundary.

4. The method of claim 1, further comprising modifying the position of the center of the heat therapy applicator.

5. The method of claim 1, further comprising validating, in the computer, the anatomical boundary to confirm that the temperature uncertainty of the portion of the anatomical boundary is below the threshold temperature uncertainty.

6. The method of claim 5, further comprising generating, in the computer, an alert when the temperature uncertainty of the portion of the anatomical boundary is greater than the threshold temperature uncertainty.

7. The method of claim 5, further comprising calculating, in the computer, a standard deviation at each point along the anatomical boundary over the past M capture times.

8. The method of claim 1, further comprising generating, in the computer, a temperature uncertainty map comprising the temperature uncertainty for each pixel.

9. The method of claim 8, further comprising displaying the temperature uncertainty map on a display coupled to the computer.

10. The method of claim 1, further comprising detrending the corrected temperature at each pixel over the past M capture times to form detrended temperature data.

11. The method of claim 10, further comprising performing a linear regression of the corrected temperature at each pixel over the past M capture times.

12. The method of claim 10, further comprising calculating a standard deviation of the detrended temperature data at each pixel.

13. The method of claim 11, further comprising determining the temperature uncertainty based on a standard deviation of the detrended temperature data at each pixel.

14. The method of claim 1, further comprising:

receiving, in the computer, a new set of temperature data for pixels corresponding to the portion of the patient's body; and

calculating an updated temperature uncertainty based on the past M capture times, the past M capture times including the new set of temperature data.