Processing system, image processing method, learning method, and processing device
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12620211-B2
Publication Date
2026-05-05
Expiration Date
Abstract
A processing system includes a processor with hardware. The processor is configured to perform processing of acquiring a detection target image captured by an endoscope apparatus, controlling the endoscope apparatus based on control information, detecting a region of interest included in the detection target image based on the detection target image for calculating estimated probability information representing a probability of the detected region of interest, identifying the control information for improving the estimated probability information related to the region of interest within the detection target image based on the detection target image, and controlling the endoscope apparatus based on the identified control information.
Core Innovation
The invention relates to an endoscope-based processing system that acquires a detection target image based on first control information and calculates a detection result of a region of interest and an estimated probability of the detected result. The system determines whether the estimated probability is equal to or higher than a first threshold and outputs the detection result when the threshold condition is satisfied. When the estimated probability is less than the first threshold, the system performs update of the first control information.
In performing the update, the system identifies second control information and third control information based on the first detection target image and the first control information, acquires a second detection target image based on the second control information, and acquires a third detection target image based on the third control information. It then detects the region of interest based on the second detection target image to calculate a second estimated probability and detects the region of interest based on the third detection target image to calculate a third estimated probability. The system performs an update using control information associated with the higher of the second estimated probability and the third estimated probability as the first control information.
The system further includes training/inference architectures using a first neural network for region-of-interest detection and probability estimation and a second neural network for recommending a priority parameter type, optionally with recommended parameter values. The design rationale described is reducing user burden and oversight errors. The system also overlays or superposes detection results and estimated probability on a display image according to thresholds, with optional behavior related to skipping or storing display content.
Claims Coverage
The independent claims cover a processing system, a processor-implemented method, and a non-transitory computer-readable storage medium that perform region-of-interest detection with an estimated probability, compare the probability to a threshold, output detection results when reliable, and otherwise update control information by testing multiple candidate control informations and selecting the one associated with the higher estimated probability. The independent claims specify an updating branch that evaluates second and third control information by acquiring second and third detection target images and comparing corresponding second and third estimated probabilities, resulting in a higher-probability control information being used for the first control information.
Threshold-gated output of region-of-interest detection result
A processor acquires a first detection target image generated based on a first control information, calculates a detection result of a region of interest and estimated probability based on the first detection target image, determines whether the estimated probability is equal to or higher than a first threshold, and outputs the detection result in a case where the estimated probability is equal to or higher than the first threshold.
Control information update via higher estimated probability among candidate control informations
When the estimated probability is less than the first threshold, the processor performs update of the first control information by identifying a second control information and a third control information based on the first detection target image and the first control information, acquiring a second detection target image based on the second control information, acquiring a third detection target image based on the third control information, detecting the region of interest based on the second and third detection target images to calculate a second estimated probability and a third estimated probability, and performing an update using control information used for outputting one of the second estimated probability and the third estimated probability that is higher as the first control information.
Across the independent claims, the core inventive coverage is the combination of region-of-interest detection with an estimated probability from a detection target image generated using control information, a first-threshold decision that either outputs the detection result or triggers an update, and an update mechanism that tests second and third control information by acquiring corresponding detection target images, calculating corresponding estimated probabilities, and selecting the higher-probability associated control information to update the first control information.
Stated Advantages
Reducing user burden and oversight errors.
Documented Applications
Endoscope-based processing where detection results and estimated probability can be superposed/overlaid on a display image based on thresholds.
Interested in licensing this patent?