A colon polyp image processing method is provided. A value of a blood vessel color feature in an endoscopic image is classified by using an image classification model trained using a neural network algorithm to determine that the endoscopic image is a white light type picture or an endoscope narrow band imaging (NBI) type picture. A polyp in the endoscopic image is detected by using a polyp positioning model based on the determination that the endoscopic image is the white light type picture or the NBI type picture. A polyp type classification detection is performed on the detected polyp in the endoscopic image by using a polyp property identification model, and outputting an identification result.

A colon polyp image processing method is provided. A value of a blood vessel color feature in an endoscopic image is classified by using an image classification model trained using a neural network algorithm to determine that the endoscopic image is a white light type picture or an endoscope narrow band imaging (NBI) type picture. A polyp in the endoscopic image is detected by using a polyp positioning model based on the determination that the endoscopic image is the white light type picture or the NBI type picture. A polyp type classification detection is performed on the detected polyp in the endoscopic image by using a polyp property identification model, and outputting an identification result.

Systems and methods for tracking an evolution of a disease in a colon of a patient over time are configured for operations including receiving video data representing a colon of a patient; segmenting the video data into segments representing portions of a colon of the patient; extracting, from the video data based on the segmenting, a set of features representing locations in the colon of the patient; and registering the feature vector as representing the colon of the patient for tracking the evolution of the disease in the colon of the patient. The system can be configured to predict disease progression predict drug dosage for patients.

Methods and systems are provided for aggregating features in multiple video frames to enhance tissue abnormality detection algorithms, wherein a first detection algorithm identifies an abnormality and aggregates adjacent video frames to create a more complete image for analysis by an artificial intelligence detection algorithm, the aggregation occurring in real time as the medical procedure is being performed.

The present disclosure provides systems and methods for improving detection and location determination accuracy of abnormalities during a gastroenterological procedure. One example method includes obtaining a video data stream generated by an endoscopic device during a gastroenterological procedure for a patient. The method includes generating a three-dimensional model of at least a portion of an anatomical structure viewed by the endoscopic device based at least in part on the video data stream. The method includes obtaining location data associated with one or more detected abnormalities based on localization data generated from the video data stream of the endoscopic device. The method includes generating a visual presentation of the three-dimensional model and the location data associated with the one or more detected abnormalities; and providing the visual presentation of the three-dimensional model and the location data associated with the detected abnormality for use in diagnosis of the patient.

An artificial intelligence-based image processing method includes: obtaining a first sample image of a source domain and a second sample image of a target domain, the first sample image of the source domain carrying a corresponding target processing result; converting the first sample image into a target sample image, the target sample image carrying a corresponding target processing result; training a first image processing model based on the target sample image and the target processing result corresponding to the target sample image, to obtain a second image processing model; and inputting, in response to obtaining a human tissue image of the target domain, the human tissue image into the second image processing model, positioning, by the second image processing model, a target human tissue in the human tissue image, and outputting position information of the target human tissue in the human tissue image.

Implementations include herein are visual navigation strategies and systems for lumen center tracking comprising a high-level state machine for gross (i.e., left/right/center) region prediction and curvature estimation and multiple state-dependent controllers for center tracking, wall-avoidance and curve following. This structure allows a navigation system to navigate even under the presence of significant occlusion that occurs during turn navigation and to robustly recover from mistakes and disturbances that may occur while attempting to track the lumen center. This system comprises a high-level state machine for gross region prediction, a turn estimator for anticipating sharp turns, and several lower level controllers for heading adjustment.

Described herein are systems, methods, and apparatuses for actively and continually fine-tuning convolutional neural networks to reduce annotation requirements, in which the trained networks are then utilized in the context of medical imaging. The success of convolutional neural networks (CNNs) in computer vision is largely attributable to the availability of massive annotated datasets, such as ImageNet and Places. However, it is tedious, laborious, and time consuming to create large annotated datasets, and demands costly, specialty-oriented skills. A novel method to naturally integrate active learning and transfer learning (fine-tuning) into a single framework is presented to dramatically reduce annotation cost, starting with a pre-trained CNN to seek “worthy” samples for annotation and gradually enhances the (fine-tuned) CNN via continual fine-tuning. The described method was evaluated using three distinct medical imaging applications, demonstrating that it can reduce annotation efforts by at least half compared with random selection.

A system and method assist gastroscopic image diagnosis based on artificial intelligence. The processor in the system analyzes each video frame of a gastroscopic image using at least one medical image analysis algorithm and detects whether a finding suspected of being a lesion is present in the video frame. When the finding suspected of being a lesion is present in the video frame, the processor calculates the coordinates of the location of the finding suspected of being a lesion. The processor generates display information, including whether the finding suspected of being a lesion is present and the coordinates of the location of the finding suspected of being a lesion.

A program causes a computer to execute processing of: acquiring a plurality of images captured by an endoscope over a predetermined period; and estimating a future state of an intracorporeal part included in the plurality of images on the basis of the plurality of acquired images.