Automatic detection of diaphragm in temporal motion data using respirator waveforms

Abstract

A diaphragm measurement system (1) comprises at least one electronic processor (13) programmed to perform a diaphragm measurement method (100) comprising receiving ultrasound imaging data (24) of a diaphragm of a patient over a period of time comprising multiple breaths; receiving breathing data of the patient over the time period; calculating a diaphragm thickness metric based on the received ultrasound imaging data of the diaphragm and the received respiration data; and displaying a representation (30) of the calculated diaphragm thickness metric on a display device (14).

Description

[0001] Cross-references to related applications

[0002] This patent application claims priority to U.S. Provisional Application No. 63 / 354,818, filed June 23, 2022, pursuant to Section 119(e) of Title 35 of the United States Code, the contents of which are incorporated herein by reference. Technical Field

[0003] The following generally covers respiratory therapy techniques, mechanical ventilation techniques, ventilator-induced lung injury (VILI) techniques, mechanical ventilation weaning techniques, and related technologies. Background Technology

[0004] The fractional diaphragm thickness (TFdi or TFDI) measured by ultrasound (US) is widely recognized as an assessment of diaphragm function to optimize ventilator support and weaning. TFdi is defined as the percentage increase in diaphragm thickness during tidal breathing relative to end-expiratory thickness. TFdi depends on diaphragmatic activity and reflects the work of breathing (WoB) (i.e., respiratory effort) (see Vivier E, Mekontso Dessap A, Dimassi S et al., “Diaphragm ultrasonography to estimate the work of breathing during non-invasive ventilation” (IntensiveCare Med 2012; 38:796-803); Goligher EC, Fan E, Herridge MS et al., “Evolution of diaphragm thickness during mechanical ventilation. Impact of inspiratory effort” (Am J Respir Crit Care Med 2015; 192:1080-1088)).

[0005] During inspiration and expiration, diaphragmatic thickness and strain can be assessed in the zone of adsorption (ZA) using a 10–15 MHz linear high-frequency sensor. The ZA is the chest wall region where the lower ribs reach the abdominal contents. The probe is positioned between the anterior and mid-axillary lines, perpendicular to the chest wall. Using ultrasound in B mode, the hemidiaphragm below the intercostal muscles is identified as a layer of hypoechoic muscle tissue located between two hyperechoic lines (pleural and peritoneal lines) (e.g., see Fayssoil A, Behind A, Ogna A, et al., “Diaphragm: Pathophysiology and Ultrasound Imaging in Neuromuscular Disorders” (J Neuromuscul Dis. 2018)). Diaphragmatic thickening is assessed using a thickening score (TFdi), which is calculated as the diaphragmatic thickness during tidal breathing relative to the end-expiratory thickness (T). ee The percentage increase in inhalation is calculated according to Equation 1:

[0006]

[0007] Where T ei This refers to the thickness at the end of the intake process.

[0008] Several studies have assessed the correlation between TFdi and respiratory effort (see M. Umbrello et al., “Diaphragm ultrasound as indicator of respiratory effort in critically ill patients undergoing assisted mechanical ventilation: a pilot clinical study” (Crit Care 19(1):161,2015)). In this study, the authors found a correlation coefficient of R = 0.8 between TFdi and the esophageal pressure-time product. diThe correlation coefficient between the diaphragm strain and the time product of diaphragm pressure was R = 0.7. In another study (e.g., see E. Oppersma et al., “Functional assessment of the diaphragm by speckletracking ultrasound during inspiratory loading” (JAppl Physiology, 123(5):1063-1070, 2017)), diaphragm strain can also be measured in real time in the wall-attached region. For example, in this study, the functional assessment of the diaphragm by speckle-tracking ultrasound during inspiratory loading was analyzed. The speckle-tracking ultrasound technique allows the detection and tracking of changes in diaphragm strain over time by analyzing acoustic markers called specks. These specks are formed by the interference of ultrasound waves scattered from physical structures comparable to the wavelength of the ultrasound waves. Both diaphragm strain and diaphragm strain rate are correlated with transdiaphragm pressure Pdi (strain r). 2 =0.72; strain rate r 2 =0.80) and EAdi (strain r) 2 =0.60; strain rate r 2 =0.66) is highly correlated.

[0009] Furthermore, the use of ultrasound to assess respiratory muscle function (especially the diaphragm) is relatively new and still uncommon due to the difficulty in obtaining sufficient measurements (e.g., see Aarab Y, Jaber S, De Jong A, “Diaphragm Ultrasonography in ICU: Why, How, and When To Use It?” (ICU Management & Practice, Vol. 21-No. 3, 2021); Tuinman, PR, Jonkman, AH, Dres, M. et al., “Respiratory muscle ultrasonography: methodology, basic and advanced principles and clinical applications in ICU and ED patients—an anarrative review” (Intensive Care Med 46, 594–605 (2020)). For example, potential confounding factors that reduce the repeatability and accuracy of daily bedside measurements include varying conditions such as the position of the US probe on the patient each time a measurement is taken, the probe's orientation (angle), the phase point in the patient's respiratory cycle, manually applied probe skin pressure, user-selected ultrasound settings, the fractional thickening that varies with the diaphragm's range, and the workload of manually recording, storing, and processing data points.

[0010] Diaphragmatic thickness and thickness fraction can be assessed using B-mode and M-mode imaging (e.g., see...). BS and Gürsel G, “Does it make difference to measure diaphragm function with M mode (MM) or B mode (BM)?” (J Clin Monit Comput. Vol. 34, 1247–1257, 2020). In the M mode case, a 2D B mode movie is first recorded. A single scan line intersecting the region of interest in the diaphragm is selected from the image. Next, a time-motion image of this scan line is plotted, from which the diaphragm thickness and thickness fraction can be determined.

[0011] Wearable ultrasound patches for non-invasive, continuous assessment of diaphragmatic thickness can be used for several potential applications, such as continuous atrophy detection, diaphragmatic dysfunction detection, precise respiratory rate detection, weaning prediction, asynchronous detection, and proportional ventilation (non-invasive NAVA).

[0012] Ideally, ultrasound patches used to measure diaphragm thickness (or fractional thickening) are designed to operate in M-mode, creating one or more scan lines rather than a full two-dimensional (2D) image. This is to reduce cost, footprint, power consumption, and the amount of data that needs to be processed and transmitted. However, without a full 2D image, it is difficult to identify the diaphragm on a single scan line.

[0013] The following discloses some improvements to overcome these and other problems. Summary of the Invention

[0014] In one aspect, a diaphragm measurement system includes at least one electronic processor programmed to perform a diaphragm measurement method comprising: receiving ultrasound imaging data of a patient's diaphragm over a time period including multiple breaths; receiving respiratory data of the patient during the time period; calculating a diaphragm thickness measure based on the received ultrasound imaging data of the diaphragm and the received respiratory data; and displaying a representation of the calculated diaphragm thickness measure on a display device.

[0015] In another aspect, a diaphragm measurement method includes using at least one electronic processor: receiving ultrasound imaging data of a patient's diaphragm over a time period including multiple breaths; receiving respiratory data of the patient during that time period; calculating a diaphragm thickness measure based on the received ultrasound imaging data of the diaphragm and the received respiratory data; and displaying a representation of the calculated diaphragm thickness measure on a display device.

[0016] One advantage is the use of wearable ultrasound (US) patches to acquire the patient's US imaging data.

[0017] Another advantage is that the location of the patient's diaphragm can be determined based on US imaging data.

[0018] Another advantage is that it allows for the determination of the location of the patient's diaphragm without complete US images.

[0019] Another advantage is the ability to control the settings of the US patch to obtain the patient's US imaging data.

[0020] Another advantage is that the settings of the mechanical ventilator can be controlled based on the patient's US imaging data.

[0021] The given embodiments may provide zero, one, two, more or all of the above advantages, and / or may provide other advantages that will be apparent to those skilled in the art upon reading and understanding this disclosure. Attached Figure Description

[0022] This disclosure can be presented in various forms of components and component arrangements, as well as various forms of steps and step arrangements. The accompanying drawings are for illustrative purposes only and should not be construed as limiting this disclosure.

[0023] Figure 1 An illustrative mechanical ventilation system according to this disclosure is shown schematically.

[0024] Figure 2 It shows the result of Figure 1 Example flowchart of the appropriate operations performed by the system. Detailed Implementation

[0025] As used herein, unless the context clearly specifies otherwise, the singular forms “a,” “an,” and “the” include plural references. As used herein, statements of “coupled,” “connected,” or “joined” two or more parts or components should mean that these parts are directly or indirectly (i.e., through one or more intermediate parts or components) connected, operated, or act together, provided a link exists. Directional phrases used herein (such as, for example, but not limited to, top, bottom, left, right, upper, lower, front, rear, and their derivatives) refer to the orientation of elements shown in the accompanying drawings and do not limit the scope of the invention unless expressly indicated herein. The word “comprising” does not exclude the presence of elements or steps other than those listed herein and / or in the claims. In a device consisting of several components, some of these components may be embodied by one or the same hardware.

[0026] refer to Figure 1 A diaphragm measurement device or system 1 is shown. A mechanical ventilator 2 is shown, configured to provide ventilation therapy to the relevant patient P. Figure 1As shown, the mechanical ventilator 2 includes an outlet 4 that is connectable to a patient breathing circuit 5 to deliver mechanical ventilation to a patient P. The patient breathing circuit 5 includes typical components of a mechanical ventilator, such as an inlet line 6, an optional outlet line 7 (which may be omitted if the ventilator uses a single-limb patient circuit), a connector or port 8 for connecting an endotracheal tube (ETT) 16, and one or more respiratory sensors (not shown), such as a gas flow meter, a pressure sensor, an end-expiratory carbon dioxide (etCO2) sensor, etc. The mechanical ventilator 2 is designed to deliver air, an air-oxygen mixture, or other inhalable gases (not shown) to the outlet 4 at a programmed pressure and / or flow rate to ventilate the patient via the ETT. The mechanical ventilator 2 also includes at least one electronic processor or controller 13 (e.g., an electronic processor or microprocessor), a display device 14, and a non-transitory computer-readable medium 15 storing instructions executable by the electronic controller 13.

[0027] Figure 1 The illustration schematically depicts a patient P intubated with an ETT 16 (its lower portion is inside the patient P's body and is therefore shown in dashed lines). A connector or port 8 connects to the ETT 16 to operatively connect a mechanical ventilator 2, thereby delivering breathable air to the patient P via the ETT 16. Mechanical ventilation provided by the mechanical ventilator 2 via the ETT 16 can treat a variety of conditions, such as various types of lung conditions like emphysema or pneumonia, viral or bacterial infections affecting respiration (such as COVID-19 infection or severe influenza), cardiovascular conditions, etc., in which the patient P receives oxygen-rich breathable gas.

[0028] Figure 1 This indicates that patient P has been intubated. That is to say, Figure 1 The image shows a patient after endotracheal intubation to insert an ETT 16. However, for safe endotracheal intubation, an anesthesiologist or other qualified medical professional first assesses the patient (P) to select the appropriate ETT size for the ETT 16, and then inserts the selected size ETT into the patient (P) via the endotracheal intubation procedure.

[0029] Figure 1Medical imaging device 18 (also referred to as an image acquisition device, imaging device, etc.) is also shown. As primarily described herein, medical imaging device 18 includes an ultrasound (US) medical imaging device 18. In other embodiments, image acquisition device 18 may be a computed tomography (CT) image acquisition device, a C-arm imager or other X-ray imaging device; a magnetic resonance (MR) image acquisition device; or another modality of medical imaging device. As described herein, medical imaging device 18 is used to acquire images of patient P. In some embodiments, medical imaging device 18 may include a wearable US imaging device 18. In other embodiments, medical imaging device 18 may include a US imaging probe 18.

[0030] In a more specific example, medical imaging device 18 includes an ultrasound patch 20 that can be worn by patient P (e.g., as shown in the image). Figure 1 As shown, the US patch 20 is located on the abdomen or chest of patient P for imaging the patient's diaphragm. The US patch 20 includes one or more transducers and is positioned to acquire US imaging data (i.e., US images) 24 of patient P's diaphragm. For example, the US patch 20 is configured to acquire imaging data of patient P's diaphragm while patient P is receiving mechanical ventilation using a mechanical ventilator 2, and more specifically, to acquire US imaging data relating to the thickness of patient P's diaphragm during inspiration and expiration. An electronic processor 13 controls an ultrasound imaging device 18 to receive ultrasound imaging data 24 of patient P's diaphragm from the handheld US patch 20. In other embodiments, the electronic processor 13 may be implemented in the ultrasound imaging device 18 and may be configured to control the operation of both the ultrasound imaging device 18 and / or the mechanical ventilator 2. In yet another embodiment, the electronic processor 13 may be implemented in a separate electronic processing device (not shown), such as a computer, smart tablet, etc.

[0031] In some embodiments, the non-transitory computer-readable medium 15 stores an artificial neural network (ANN) model 22 configured to determine a diaphragm thickness measurement based on ultrasound imaging data 24 and the patient's respiratory data. The non-transitory computer-readable medium 15 also stores instructions executable by the electronic controller 13 to perform the diaphragm measurement method or process 100.

[0032] refer to Figure 2 And continue to refer to Figure 1 An illustrative embodiment of the diaphragm measurement method 100 is schematically shown in the form of a flowchart. At operation 101, ultrasound imaging data 24 of the diaphragm of patient P is received over a time period including multiple breaths. For example, an electronic processor 13 controls an ultrasound imaging patch 20 to acquire ultrasound imaging data 24 of the diaphragm of patient P.

[0033] At operation 102, the patient's respiratory data for that time period is received. The respiratory data may be, for example, airway pressure (from a pressure sensor), airway flow (from a flow sensor), or the output of a respiratory monitor (such as the breathing belt worn by patient P).

[0034] In operation 103, a diaphragm thickness metric can be calculated based on the received US imaging data 24 of the diaphragm of patient P and the received respiratory data. In one example, the diaphragm thickness metric includes a diaphragm thickening rate, which indicates the diaphragm thickness during inspiration relative to the diaphragm thickness during expiration. In another example, the diaphragm thickness metric includes the average diaphragm thickness over multiple respiratory cycles.

[0035] In operation 104, a representation 30 of the calculated diaphragm thickness measurement is displayed on the display device 14 of the mechanical ventilator 2. In some embodiments, at operation 105, the mechanical ventilator 2 can be controlled to adjust one or more parameters of the mechanical ventilation therapy delivered to the patient based on the calculated diaphragm thickness measurement.

[0036] In some embodiments, US imaging data 24 includes M-mode US imaging data. For example, the diaphragm can be automatically detected in a single-channel ultrasound echo mode, thereby utilizing the correlation between airway pressure and diaphragm thickness in actively breathing patients receiving pressure-supported ventilation, or pressure- or volume-controlled ventilation. For example, in pressure-supported ventilation mode, airway pressure is correlated with diaphragm thickness, such as... Figure 1 As shown in Illustration A. In such an embodiment, calculating a diaphragm thickness measure based on received M-mode ultrasound imaging data 24 of the diaphragm and received respiratory data includes identifying the component of the M-mode ultrasound imaging data 24 corresponding to the diaphragm of patient P based on the respiratory data of patient P, and calculating the diaphragm thickness measure based on the identified component of the M-mode ultrasound imaging data 24 corresponding to the diaphragm of patient P. For this purpose, hyperechoic lines in the M-mode ultrasound imaging data 24 are identified and grouped into pairs. Hyperechoic lines can be identified, for example, by thresholding, edge detection, or any other suitable technique. Alternatively, this is done on the raw ultrasound data 24 and then converted to a grayscale image by identifying the high-amplitude portions of the echo pattern (i.e., strong echoes reflected from hyperechoic structures). Pairs of hyperechoic lines represent possible boundaries of the diaphragm. Preferably, adjacent lines are grouped together because the diaphragm should appear as a layer of low-echoic muscle tissue located between two adjacent hyperechoic lines (i.e., the pleural line and the peritoneal line).

[0037] For each pair of hyperechoic lines, the distance between the pair of hyperechoic lines is determined as a function of time, and the correlation between the distance between the hyperechoic lines in the pair determined as a function of time and the patient P's respiratory data is determined. Based on the determined correlation, the components of the M-mode ultrasound imaging data 24 corresponding to the patient P's diaphragm are identified as one pair of hyperechoic lines. Since airway pressure is related to diaphragmatic thickening for pressure support ventilation, the pair with the largest correlation coefficient is selected as the two hyperechoic lines containing the diaphragm. For example, the patient's respiratory rate over a period of time is identified from the patient P's respiratory data, and the M-mode ultrasound imaging data 24 is filtered using a bandpass filter 26 (e.g., implemented in the non-transitory computer-readable medium 15 of the mechanical ventilator 2) to extract the components of the M-mode ultrasound imaging data 24 corresponding to the patient P's diaphragm, the passband of which is centered on the identified respiratory rate.

[0038] In the case of multiple scan lines, the optimal scan line can be selected by choosing the one with the highest correlation coefficient. Furthermore, multiple scan lines can be selected (e.g., for those with correlation coefficients exceeding a predetermined threshold level), and the average or median diaphragmatic thickness can be calculated. If (multiple) correlation coefficients do not exceed the predetermined threshold level, the user / caregiver can be alerted (e.g., to reposition patch 20 or to check for partial or complete (paralyzed) diaphragmatic dysfunction). This procedure can be repeated periodically (e.g., daily).

[0039] In some embodiments, US imaging data 24 of the diaphragm of patient P is received over a time period including multiple breaths, for example, during inspiration and expiration of patient P during mechanical ventilation with mechanical ventilator 2. In such embodiments, electronic controller 13 is configured to determine the phase shift between bandpass-filtered M-mode ultrasound imaging data 24 and patient P's respiratory data. Patient ventilator asynchrony is determined based on the phase shift, and an indication of determined patient ventilator asynchrony is displayed on display device 14 of mechanical ventilator 2. For this purpose, US imaging data 24 and respiratory data are acquired, and the frequency of mechanical ventilator 2 is determined. The US imaging time-motion data is filtered, and the phase shift between the muscle thickening signal and the ventilator pressure signal from the US time-motion data is determined.

[0040] In some embodiments, the diaphragm measurement method 100 can be repeated for consecutive sessions based on adjustments to system 1. In one example, one or more settings of the mechanical ventilator 2 are adjusted, and method 100 is repeated for each adjustment. For example, the diaphragm is identified in an active breathing patient receiving pressure support by changing ventilator settings and utilizing the negative correlation between diaphragm thickening fraction and pressure support level to identify the diaphragm in the echo pattern. For this purpose, a ventilator support level is set. Hyperechoic lines in time-motion data are identified, and the distance between each pair of adjacent high-amplitude portions is determined. For each breath, timestamps of the end of inspiration and the end of expiration are determined based on the pressure or flow waveform. For each pair, a thickness fraction is calculated using the timestamps. These operations are repeated at several support levels. For each pair, a correlation coefficient between support level and maximum thickness is determined. The pair with the largest correlation coefficient is selected as the two hyperechoic lines where the diaphragm is located.

[0041] In the case of multiple scan lines, the optimal scan line can be selected by repeating these operations for each scan line and choosing the one with the highest correlation coefficient. Furthermore, multiple scan lines can be selected (e.g., for those with correlation coefficients exceeding a predetermined threshold level), and the average or median diaphragm thickness can be calculated. Additionally, if (multiple) correlation coefficients do not exceed the predetermined threshold level, the user / caregiver can be alerted (e.g., patch 20 is repositioned).

[0042] In another example, one or more settings of the ultrasonic patch 20 (such as frequency, time-varying gain, swivel angle, depth of focus, and aperture size and position (i.e., selection of active elements in the ultrasonic array) are adjusted, and method 100 is repeated for each adjustment. The ultrasonic patch 20 is equipped with a transducer array that allows swivel, and several scan lines can be recorded by sweep angle. The optimal scan lines can be selected, or the average thickness (fraction) can be calculated. Similarly, the optimal settings for frequency, depth of focus, etc., can be found.

[0043] In another embodiment, calculating the diaphragm thickness metric may include inputting M-mode ultrasound imaging data 24 and respiratory data into an ANN model 22 to determine the diaphragm thickness metric based on the M-mode ultrasound imaging data 24 and the respiratory data of patient P. During training of the ANN model 22, the correlation between pressure and diaphragm thickness is implicitly learned. To this end, the M-mode ultrasound imaging data 24 and respiratory data are acquired and input into the ANN model 22, which is then trained to reproduce the real diaphragm on the ground.

[0044] In some embodiments, the mechanical ventilator (MV)2 provides several types of waveforms (e.g., pressure, flow, work of breathing, volume, etc.). For each pair of the US line and MV signals, the correlation is determined, and the optimal correlation selection for the MV waveform is determined. In cases of partial or complete diaphragmatic dysfunction, the diaphragmatic thickness fraction can be negative under pressure or volume-controlled ventilation due to passive stretching (e.g., see Santana PV, Cardenas LZ, Albuquerque ALP, Carvalho CRR, Caruso P., "Diaphragmatic ultrasound: a review of its methodological aspects and clinical uses" (JBras Pneumol. 2020 Nov 20)). Furthermore, other factors can also cause a negative diaphragmatic thickness fraction, such as excessively high support levels. Therefore, when a negative diaphragmatic thickness fraction is detected, a warning can be issued to caregivers, or the ventilation settings can be automatically adjusted.

[0045] Although described in the context of mechanical ventilation therapy, System 1 can be used in any suitable setting, including cardiac surgery, heart failure monitoring, asynchronous detection, home ventilation, bladder and prostate enlargement monitoring, carotid artery stenosis (e.g., stroke, cardiac output, A-fib, etc.), myocardial function and cardiac output in acute care settings (i.e., heart attack), etc.

[0046] This disclosure has been described with reference to preferred embodiments. After reading and understanding the above detailed description, others may make modifications and variations. It is intended that the exemplary embodiments be interpreted to include all such modifications and variations, provided they fall within the scope of the appended claims or their equivalents.

Claims

1. A diaphragm measurement system, comprising: At least one electronic processor programmed to perform a diaphragm measurement method comprising: receiving ultrasound imaging data of a diaphragm of a patient over a period of time including a plurality of breaths; receiving respiratory data of the patient during the time period; calculating a diaphragm thickness measure based on the received ultrasound imaging data of the diaphragm and the received respiration data; and A representation of the calculated diaphragm thickness metric is displayed on a display device. 2 . The system of claim 1 , wherein the diaphragm thickness metric comprises a diaphragm thickening ratio indicating diaphragm thickness during inspiration relative to diaphragm thickness during expiration.

3. The system of claim 1, wherein the diaphragm thickness metric comprises an average diaphragm thickness over a plurality of respiratory cycles.

4. The system according to claim 1, further comprising: An ultrasound imaging patch wearable by the patient, wherein the at least one electronic processor controls the ultrasound imaging patch to obtain the ultrasound imaging data of the diaphragm of the patient.

5. The system of claim 1 , wherein the ultrasound imaging data comprises M-mode ultrasound imaging data, and calculating the diaphragm thickness metric based on the received M-mode ultrasound imaging data of the diaphragm and the received respiration data comprises: identifying a component of the M-mode ultrasound imaging data corresponding to the diaphragm of the patient based on the respiration data of the patient; as well as The diaphragm thickness metric is calculated based on the identified component of the M-mode ultrasound imaging data corresponding to the diaphragm of the patient.

6. The system of claim 5, wherein the at least one electronic processor is programmed to identify the component of the M-mode ultrasound imaging data corresponding to the diaphragm of the patient by: identifying a hyperechoic line in the M-mode ultrasound imaging data; grouping the hyperechoic lines into pairs of hyperechoic lines; for each pair of hyperechoic lines, determining the distance between the hyperechoic lines in the pair as a function of time; for each pair of hyperechoic lines, determining a correlation between a distance between the hyperechoic lines in the pair determined as a function of time and the respiration data of the patient; and The component of the M-mode ultrasound imaging data corresponding to the diaphragm of the patient is identified as one of the pairs of hyperechoic lines based on the determined correlation.

7. The system of claim 5, wherein identifying the component of the M-mode ultrasound imaging data corresponding to the diaphragm of the patient based on the respiratory data of the patient comprises: identifying, from the respiratory data of the patient, a respiratory rate of the patient during the time period; as well as The M-mode ultrasound imaging data is filtered using a bandpass filter having a passband centered at the identified respiratory rate to extract the component of the M-mode ultrasound imaging data corresponding to the diaphragm of the patient.

8. The system of claim 7, wherein the patient receives mechanical ventilation during the time period, and the electronic processor is further programmed to: determining a phase shift between bandpass filtered M-mode ultrasound imaging data and the respiration data of the patient; determining patient-ventilator asynchrony based on the phase shift; and An indication of the determined patient-ventilator asynchrony is displayed on the display device.

9. The system of claim 1, wherein the respiratory data of the patient comprises airway pressure, airway flow, or output of a respiratory monitor.

10. The system of claim 1, wherein the patient receives mechanical ventilation during the time period, and the electronic processor is programmed to: One or more settings of the mechanical ventilator are adjusted, and the diaphragm measurement method is repeated for each adjustment.

11. The system of claim 1 , wherein the electronic processor is programmed to: One or more settings of an ultrasound patch configured to acquire the ultrasound imaging data are adjusted, and the diaphragm measurement method is repeated for each adjustment.

12. The system of claim 1 , wherein calculating the diaphragm thickness metric based on the received ultrasound imaging data of the diaphragm and the received respiration data comprises: The ultrasound imaging data and the respiration data are input to an artificial neural network (ANN) model configured to determine the diaphragm thickness metric based on the ultrasound imaging data and the respiration data of the patient.

13. The system of claim 1, wherein the at least one electronic processor is configured to: An associated mechanical ventilator is controlled to adjust one or more parameters of mechanical ventilation therapy delivered to the patient based on the calculated diaphragm thickness metric.

14. The system of claim 13, further comprising: A mechanical ventilator is configured to deliver mechanical ventilation therapy to the patient.

15. A method of measuring diaphragm muscle comprising using at least one electronic processor: receiving ultrasound imaging data of a diaphragm of a patient over a period of time including a plurality of breaths; receiving respiratory data of the patient during the time period; calculating a diaphragm thickness measure based on the received ultrasound imaging data of the diaphragm and the received respiration data; as well as A representation of the calculated diaphragm thickness metric is displayed on a display device.

Images