Systems and methods for segmenting surgical videos
AssigneeVerily Life Sciences LLC
InventorsJoëlle K. Barral, Martin Habbecke, Michal Levin, Antons Krumins, Thomas Teisseyre
Life of this patent
Estimating the term…
Abstract
Systems and methods for segmenting surgical videos are disclosed. One example method includes receiving, by a processor of a computing device, surgical video, the surgical video comprising at least a sequence of video frames of a surgical procedure; in response to receiving an identification of a video frame, generating, by the processor, a bookmark based on the video frame; associating, by the processor, the bookmark with the video frame; and storing, by the processor, the bookmark in a non-transitory computer-readable medium.
Claims30
A method comprising:
- receiving, by a processor of a computing device, surgical video, the surgical video comprising at least a sequence of video frames of a surgical procedure;
- recognizing one or more steps of the surgical video by the processor using a trained ML technique;
- identifying one or more video frames of the surgical video corresponding to the recognized steps;
- in response to identification of the one or more video frames, generating one or more bookmarks based on the one or more steps of the surgical video and the one or more video frames;
- associating, by the processor, the one or more bookmarks with the respective one or more video frames;
- and
- storing, by the processor, the one or more bookmarks.
A method comprising:
- receiving, by a processor of a computing device, a surgical video having a bookmark;
- receiving a selection of the bookmark;
- receiving an indication to share a segment of the surgical video and a recipient, the segment of the surgical video based on the bookmark;
- extracting a portion of the surgical video based on the bookmark;
- and
- providing the portion of the surgical video to the recipient.
A system comprising:
- a non-transitory computer-readable medium;
- and
- one or more processor communicatively coupled to the non-transitory computer-readable medium, the processor configured to execute processor-executable instructions stored in the non-transitory computer-readable medium to:
- receive surgical video, the surgical video comprising at least a sequence of video frames of a surgical procedure;
- recognize one or more steps of the surgical video by the processor using a trained ML technique;
- identify one or more video frames of the surgical video corresponding to the recognized steps;
- in response to identification of the one or more video frames, generate one or more bookmarks based on the one or more steps of the surgical video and the one or more video frames;
- associate the one or more bookmarks with the respective one or more video frames;
- and
- store the one or more bookmarks.
A system comprising:
- a non-transitory computer-readable medium;
- and
- one or more processor communicatively coupled to the non-transitory computer-readable medium, the processor configured to execute processor-executable instructions stored in the non-transitory computer-readable medium to:
- receive a surgical video having a bookmark;
- receiving a selection of the bookmark;
- receiving an indication to share a segment of the surgical video and a recipient, the segment of the surgical video based on the bookmark;
- extracting a portion of the surgical video based on the bookmark;
- and
- providing the portion of the surgical video to the recipient.
Dependencies are inferred from the claim text.