Systems and methods for tracking surgical items
Described herein are methods and systems for tracking surgical items. The methods may be performed by one or more processors, and may include receiving a first count of surgical items, receiving one or more images, wherein each image is a field of view comprising one or more surgical items, determining a second count of surgical items based at least in part on the one or more received images, and providing a notification based on the comparison between the first count of surgical items and the second count of surgical items.
1. A method comprising:
determining, by one or more processors, a first count of surgical items, the first count quantifying a set of surgical items;
accessing, by the one or more processors, an image that depicts at least a subset of the set of surgical items;
determining, by the one or more processors, a second count of surgical items by characterizing a container that includes a set of compartments depicted in the image; and
providing, by the one or more processors, a notification based on a comparison of the first count of surgical items and the second count of surgical items;
wherein the one or more processors are configured to characterize the container by:
converting, with a counting bag geometry model, the image into a plurality of pocket images, each pocket image corresponding to a portion of the image which includes a compartment of the set of compartments; and
determining the second count of the surgical items based on the pocket images.
2. The method of claim 1 , wherein:
the determining of the first count of surgical items includes identifying the set of surgical items from a further image.
3. The method of claim 1 , wherein:
the providing of the notification based on the comparison of the first count and the second count includes providing an alert in response to the first count not matching the second count.
4. The method of claim 1 , wherein characterizing the container includes classifying, with a pocket classifier model, each pocket image as corresponding to a full compartment or an empty compartment.
5. The method of claim 4 , wherein determining the second count of the surgical items is based on an output from the pocket classifier model and includes determining the second count of the surgical items based on the classification of the pocket images.
6. The method of claim 1 , wherein each pocket image includes a cropped image segment, each cropped image segment including the portion of the image that contains the corresponding compartment.
7. The method of claim 1 , wherein each pocket image includes:
the image; and
a location of the corresponding compartment in the image.
8. The method of claim 1 , further comprising displaying, on a display:
the image; and
a plurality of bounding boxes overlaid on the image, each of the bounding boxes surrounding one of the compartments.
9. A non-transitory machine-readable medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:
determining a first count of surgical items, the first count quantifying a set of surgical items;
accessing an image that depicts at least a subset of the set of surgical items;
determining a second count of surgical items by characterizing a container that includes a set of compartments depicted in the image; and
providing a notification based on a comparison of the first count of surgical items and the second count of surgical items;
wherein characterizing the container includes:
converting, with a counting bag geometry model, the image into a plurality of pocket images, each pocket image corresponding to a portion of the image which includes a compartment of the set of compartments; and
determining the second count of the surgical items based on the pocket images.
10. The method of claim 9 , wherein characterizing the container includes classifying, with a pocket classifier model, each pocket image as corresponding to a full compartment or an empty compartment.
11. The method of claim 10 , wherein determining the second count of the surgical items is based on an output from the pocket classifier model and includes determining the second count of the surgical items based on the classification of the pocket images.
12. The non-transitory machine-readable medium of claim 9 , wherein each pocket image includes a cropped image segment, each cropped image segment including the portion of the image that contains the corresponding compartment.
13. The non-transitory machine-readable medium of claim 9 , wherein each pocket image includes:
the image; and
a location of the corresponding compartment in the image.
14. The non-transitory machine-readable medium of claim 9 , further comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising displaying, on a display:
the image; and
a plurality of bounding boxes overlaid on the image, each of the bounding boxes surrounding one of the compartments.
15. A system comprising:
one or more processors; and
a memory storing instructions that, when executed by at least one processor among the one or more processors, cause the system to perform operations comprising:
determining a first count of surgical items, the first count quantifying a set of surgical items;
accessing an image that depicts at least a subset of the set of surgical items;
determining a second count of surgical items by characterizing a container that includes a set of compartments depicted in the image; and
providing a notification based on a comparison of the first count of surgical items and the second count of surgical items;
wherein characterizing the container includes:
converting, with a counting bag geometry model, the image into a plurality of pocket images, each pocket image corresponding to a portion of the image which includes a compartment of the set of compartments; and
determining the second count of the surgical items based on the pocket images.
16. The method of claim 15 , wherein characterizing the container includes classifying, with a pocket classifier model, each pocket image as corresponding to a full compartment or an empty compartment.
17. The method of claim 16 , wherein determining the second count of the surgical items is based on an output from the pocket classifier model and includes determining the second count of the surgical items based on the classification of the pocket images.
18. The system of claim 15 , wherein each pocket image includes a cropped image segment, each cropped image segment including the portion of the image that contains the corresponding compartment.
19. The system of claim 15 , wherein each pocket image includes:
the image; and
a location of the corresponding compartment in the image.
20. The system of claim 15 , wherein the instructions, when executed by at least one processor among the one or more processors, further cause the system to perform operations comprising displaying, on a display:
the image; and
a plurality of bounding boxes overlaid on the image, each of the bounding boxes surrounding one of the compartments.