Endoscopic Lesion Detection During Removal Step
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The insertion of an endoscope is technically challenging, requiring a doctor to concentrate, making it difficult to confirm lesion detection during insertion and increasing the likelihood of overlooking lesions during removal.
Innovation Solution
An endoscopic examination assistance device that acquires endoscopic images and position information, detects lesions, determines whether the endoscope is being inserted or removed, and provides notifications based on the positional relationship between the endoscope and detected lesions during the removal step.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the doctor concentrates on endoscope insertion, then the insertion operation is performed accurately, but the doctor cannot confirm the detection result of the lesion at the time of insertion
Solution Approach 1:
The system performs preliminary lesion detection during the insertion step and stores the detection results with their corresponding positions. This allows the lesion detection to be completed in advance during insertion, so the results are available for confirmation during the removal step without requiring the doctor to divide attention during insertion.
Solution Approach 2:
The system provides feedback by notifying the doctor of detected lesions during the removal step based on the stored detection results from insertion. The notification includes positional information that helps the doctor confirm whether lesions were present, creating a feedback loop that compensates for the inability to confirm during insertion itself.
2Device complexity
If the doctor only confirms the detection result at the time of removal, then the insertion process remains simple, but the lesion may be overlooked
Solution Approach 1:
The system performs preliminary lesion detection and position recording during insertion, storing this information for later reference. This preliminary action ensures that lesion detection occurs during both insertion and removal phases, increasing reliability while maintaining procedural simplicity.
Solution Approach 2:
The system provides feedback during removal by notifying the doctor of lesions detected during insertion, using the stored positional information. This feedback mechanism ensures that lesions are not overlooked by cross-referencing the stored detection results with the current position during removal.
3Measurement precision
If the notification is provided during insertion, then the lesion detection accuracy is improved, but the doctor's concentration is disrupted
Solution Approach 1:
The system performs lesion detection as a preliminary action during insertion and stores the results, but delays the notification until the removal step. This separates the detection function from the notification timing, maintaining detection accuracy while avoiding disruption to the doctor's concentration during the critical insertion phase.
Data Source
AI summary
An endoscopic examination assistance device according to an aspect of the present disclosure includes: a memory storing instructions; and at least one processor configured to execute the instructions to: acquire an endoscopic image captured by an imaging unit of an endoscope and imaging unit position information indicating a position of the imaging unit of the endoscope within a lumen at a time when the endoscopic image is captured; detect a lesion from the endoscopic image; determine whether it is an insertion step for inserting the endoscope to a turnaround point in the lumen or a removal step for removing the endoscope from a turnaround point in the lumen; and cause a notification device to give notification according to a positional relationship between the imaging unit position information and lesion position information indicating a position of a lesion, in a case where a result of the determination indicates the removal step.