Automated detection of lung slide to aid in diagnosis of pneumothorax
Methods and apparatuses for performing automated detection of lung slide using a computing device (e.g., an ultrasound system, etc.) are disclosed. In some embodiments, the techniques determine lung sliding using one or more neural networks. In some embodiments, the neural networks are part of a process that determines probabilities of the lung sliding at one or more M-lines. In some embodiments, the techniques display one or more probabilities of lung sliding in a B-mode ultrasound image.
1. A method implemented by a computing device for determining lung sliding, the method comprising:
receiving one or more B-Mode ultrasound images that include a pleural line;
generating a feature list from the one or more B-Mode ultrasound images, the feature list indicating at least one feature of the pleural line; and
generating, from the one or more B-Mode ultrasound images, one or more M-Mode ultrasound images corresponding to one or more M-lines;
generating an M-strip from the one or more M-mode images;
determining if the M-strip has a quality level acceptable for determining lung sliding;
upon determining the M-strip has the quality level acceptable for determining lung sliding, generating reconstructed M-Mode images from the M-strip;
generating, with a neural network implemented at least partially in hardware of the computing device and configured to process the feature list and a B-Mode ultrasound image of the one or more B-Mode ultrasound images, a probability of the lung sliding based on the reconstructed M-Mode ultrasound images.
2. The method as described in claim 1 , wherein the generating the probability of the lung sliding includes activating the neural network to process the feature list and the B-Mode ultrasound image automatically and without user intervention based on the B-Mode ultrasound image having a quality level above a threshold quality level.
3. The method as described in claim 1 , further comprising determining a region of interest in the B-Mode ultrasound image based on a location of the pleural line in the B-Mode ultrasound image;
wherein the generating the probability of the lung sliding is based on pixels of the B-Mode ultrasound image that are included in the region of interest and not based on additional pixels of the B-Mode ultrasound image that are not included in the region of interest.
4. The method as described in claim 3 , wherein the location indicates a distance from an end point of the pleural line to an edge of the B-Mode ultrasound image, and the determining the region of interest is based on the distance.
5. The method as described in claim 1 , further comprising generating an additional probability of the lung sliding based on an additional B-Mode ultrasound image of the one or more B-Mode ultrasound images.
6. The method as described in claim 5 , wherein the generating the additional probability of the lung sliding is based on the feature list.
7. The method as described in claim 5 , further comprising generating an additional feature list from the additional B-Mode ultrasound image, wherein the generating the additional probability of the lung sliding is based on the additional feature list.
8. The method as described in claim 5 , further comprising:
merging the probability and the additional probability to form a merged probability of the lung sliding; and
displaying, in a user interface of the computing device, a representation of the merged probability of the lung sliding.
9. The method as described in claim 1 , further comprising:
determining an additional B-Mode ultrasound image of the one or more B-Mode ultrasound images as redundant to the B-Mode ultrasound image; and
discarding the additional B-Mode ultrasound image from the one or more B-Mode ultrasound images to prevent the neural network from processing the additional B-Mode ultrasound image.
10. The method as described in claim 1 , further comprising:
generating additional B-Mode ultrasound images based on a user adjustment implemented based on an instruction; and
wherein generating, with a neural network implemented at least partially in hardware of the computing device and configured to process the feature list and a B-Mode ultrasound image of the one or more B-Mode ultrasound images comprises generating, with the neural network implemented at least partially in hardware of the computing device and based on one or more of the additional B-Mode ultrasound images, the probability of the lung sliding.
11. The method as described in claim 1 , wherein the B-Mode ultrasound images include a pleural line and quality of the B-Mode ultrasound images is based on a location of the pleural line in the B-Mode ultrasound images.
12. The method as described in claim 1 , wherein the instruction includes:
1) at least one of guidance to move an ultrasound probe,
2) an adjustment of an imaging parameter, and
3) a recommendation for selecting the neural network from a list of neural networks available on the computing device.
13. An ultrasound system for determining lung sliding, the ultrasound system comprising:
a memory to store one or more B-Mode ultrasound images that include a pleural line;
a processor coupled to the memory to:
generate a feature list from the one or more B-Mode ultrasound images, the feature list indicating at least one feature of the pleural line,
generate, from the one or more B-Mode ultrasound images, one or more M-Mode ultrasound images corresponding to one or more M-lines,
generate an M-strip from the one or more M-mode images,
determine if the M-strip has a quality level acceptable for determining lung sliding, and
upon determining the M-strip has the quality level acceptable for determining lung sliding, generate reconstructed M-Mode images from the M-strip; and
a neural network implemented at least partially in hardware of the ultrasound system to process the feature list and a B-Mode ultrasound image of the one or more B-Mode ultrasound images, a probability of the lung sliding based on the reconstructed M-Mode ultrasound images.
14. The ultrasound system as described in claim 13 , wherein the generating the probability of the lung sliding includes activating the neural network to process the feature list and the B-Mode ultrasound image automatically and without user intervention based on the B-Mode ultrasound image having a quality level above a threshold quality level.
15. The ultrasound system as described in claim 13 , wherein the processor is configured to determine a region of interest in the B-Mode ultrasound image based on a location of the pleural line in the B-Mode ultrasound image;
wherein the neural network is configured to generate the probability of the lung sliding based on pixels of the B-Mode ultrasound image that are included in the region of interest and not based on additional pixels of the B-Mode ultrasound image that are not included in the region of interest.
16. The ultrasound system as described in claim 15 , wherein the location indicates a distance from an end point of the pleural line to an edge of the B-Mode ultrasound image, and the determining the region of interest is based on the distance.
17. The ultrasound system as described in claim 13 , wherein the neural network is configured to generate an additional probability of the lung sliding based on an additional B-Mode ultrasound image of the one or more B-Mode ultrasound images.
18. The ultrasound system as described in claim 17 , wherein the neural network is configured to generate the additional probability of the lung sliding based on the feature list.
19. The ultrasound system as described in claim 17 , wherein the processor is configured to generate an additional feature list from the additional B-Mode ultrasound image, wherein the additional probability of the lung sliding is generated based on the additional feature list.
20. The ultrasound system as described in claim 17 , further comprising a computing device, and wherein the processor is configured to:
merge the probability and the additional probability to form a merged probability of the lung sliding; and
to cause display, in a user interface of the computing device, a representation of the merged probability of the lung sliding.