Surgical Video Feed Segmentation for Multiview MIS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Minimally invasive surgery (MIS) is limited by dependence on a single viewpoint from an endoscopic camera, leading to suboptimal image capture and display, which can result in tunnel vision, missed critical parts of the operative field, and reduced multitasking capabilities compared to traditional open surgery.
Innovation Solution
A system that generates multiple daughter video feeds from a single mother video feed, allowing independent panning and zooming based on predefined markers on surgical instruments, enabling multiple surgeons to have simultaneous, optimal views of the operative field without moving the camera, thus reducing image imperfections and enhancing multitasking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a single endoscopic camera is used to capture the operative field, then the surgical procedure can be performed minimally invasively, but the viewpoint is limited and every team member sees the same image which may result in suboptimal capture due to fatigue, tremor, or poor instructions
Solution Approach 1:
The system segments the single camera feed into multiple independent daughter video feeds, each showing a different view of the operative field. This allows multiple surgeons to have simultaneous optimal views without requiring multiple physical cameras, maintaining minimally invasive benefits while improving image capture quality for each user.
Solution Approach 2:
The system creates digital copies of the camera feed at different zoom levels and positions (daughter video feeds) from the single mother video feed. These copies allow multiple users to view different portions of the operative field simultaneously, compensating for camera operator limitations while maintaining the single-camera minimally invasive approach.
2Measurement precision
If the camera provides a panoramic view of the surgical field, then a wide angle is shown, but detail is removed; if it provides a close-up, then detail is shown but tunnel vision occurs and critical parts are missed
Solution Approach 1:
The system divides the single camera feed into multiple daughter video feeds, with each feed showing a different portion of the operative field at different zoom levels. This segmentation allows simultaneous display of both panoramic views (showing wide angle and context) and close-up views (showing detailed instrument work) without losing information about either the overall field or specific details.
3Device complexity
If a single viewpoint is used from the telescope and camera, then the system remains simple, but multitasking is prevented and only one surgical action can occur at any given time
Solution Approach 1:
The system creates multiple digital copies (daughter video feeds) from the single camera source, allowing multiple surgeons to simultaneously view different aspects of the operative field. This enables multitasking where multiple surgical actions can be monitored and performed concurrently, increasing productivity while maintaining the simplicity of the single-camera physical system.
Data Source
AI summary
Data characterizing a mother video feed acquired by an endoscopic video capture device can be received. The mother video feed can be for characterizing an operative field within a patient. One or more predefined markers can be identified within the mother video feed. Each of the one or more predefined markers can be associated with a surgical instrument in the operative field. Using the data characterizing the mother video feed, a daughter video feed comprising a sub-portion of the mother video feed can be generated. At least one of a location of the daughter video feed within the mother video feed and a zoom of the daughter video feed can be based on the identified one or more predefined markers. The daughter video feed can be provided. Related apparatus, systems, techniques, and articles are also described.