FIELD: medicine.
SUBSTANCE: group of inventions relates to a method and system for measuring dimensions of neoplasms during endoscopic procedures based on processing a series of endoscopic images in an automatic mode using neural networks. Method and system are proposed, in which by means of a first neural network neoplasm is detected on each separate endoscopic image from a video stream; for each growth using tracking algorithms, it is determined that the detected growth on each separate endoscopic image from the series of several adjacent endoscopic images from the video stream is the same growth, and combining obtained series of several adjacent endoscopic images into a track; one or more tracks are received, where each track contains a series of endoscopic images containing the detected neoplasm, and the obtained one or more tracks are supplied to the input of the second neural network; second neural network is used to process the obtained one or more tracks as follows: for each obtained track, selecting features from each endoscopic image of the track containing the detected neoplasm; for each endoscopic track image, selecting a region of interest in the space of the obtained features, and obtaining spatial coordinates of the selected region of interest; for each endoscopic image track obtaining time coordinates; aggregating features by time and space coordinate; the detected neoplasm is sized inside an organism on the basis of the signs aggregated by time and space coordinate by means of a regression analysis.
EFFECT: group of inventions provides accurate measurement of neoplasm size during endoscopic procedures.
6 cl, 1 dwg
Title | Year | Author | Number |
---|---|---|---|
METHOD FOR TRACKING, DETECTION AND IDENTIFICATION OF OBJECTS OF INTEREST AND AUTONOMOUS DEVICE WITH PROTECTION FROM COPYING AND HACKING FOR THEIR IMPLEMENTATION | 2021 |
|
RU2789609C1 |
METHOD FOR DETECTING A FACE IN ACCESS MONITORING AND CONTROL SYSTEMS | 2021 |
|
RU2765439C1 |
METHOD OF CONSTRUCTING A VIDEO SURVEILLANCE SYSTEM FOR SEARCHING AND TRACKING OBJECTS | 2018 |
|
RU2696548C1 |
SYSTEMS AND METHODS OF TRAINING GENERATIVE-ADVERSARIAL NETWORKS, AS WELL AS USE OF TRAINED GENERATIVE-ADVERSARIAL NETWORKS | 2019 |
|
RU2819619C2 |
SOLID ORAL COMPOSITION CONTAINING COLORS FOR APPLICATION IN ENDOSCOPIC DIAGNOSTICS | 2013 |
|
RU2649783C2 |
METHOD FOR ENDOSCOPIC SUBMUCOSAL DISSECTION OF COLORECTAL FORMATIONS | 2020 |
|
RU2756489C1 |
ENDOSCOPIC LOOP WITH POSSIBILITY OF DISSECTED POLYP ASPIRATION | 2024 |
|
RU2828965C1 |
METHOD FOR ENDOSCOPIC GASTRIC STUDYING | 2003 |
|
RU2253350C2 |
HYBRID LAPARO-ENDOSCOPIC METHOD OF REMOVING COLON TUMORS | 2023 |
|
RU2807566C1 |
METHOD OF ENDOSCOPIC RESECTION OF STOMACH MUCOSA | 2011 |
|
RU2478344C2 |
Authors
Dates
2025-04-28—Published
2022-12-14—Filed