A computer-implemented method for evaluating colonoscopy performance includes: (S1) splitting a video acquired during a colonoscopy examination into a plurality of colonoscopy images; (S2) assigning each of the colonoscopy images into a fold-inspection group or a non-fold-inspection group according to a first classification criterion and a second classification criterion, wherein the first classification criterion comprises at least one of clarity, exposure, level of tissue wrinkling, and level of occlusion in each of the colonoscopy images; and the second classification criterion comprises at least one of an amount of haustrum, an amount of colonic lumen, and a position of the colonic lumen in each of the colonoscopy images; and (S3) determining a performance rating of the colonoscopy examination according to an elapsed time of the fold-inspection group. The method classifies colonoscopy images more accurately and reliably, thereby providing an effective tool for quality assessment and guidance of colonoscopy examinations.

Disclosed is an image recognition based workstation for evaluation on quality check of colonoscopy, relating to the technical field of intelligent healthcare. The workstation comprises an algorithm module, a timing module, a data transmission module, a display device, a colonoscopy device, and a computer host. The colonoscopy device is connected with the data transmission module, and the data transmission module is connected with the computer host through the algorithm module and the timing module; and the display device is used to display the results of the computer host. The described workstation can evaluate different techniques of doctors during each colonoscopy check by means of different image recognition algorithms. During the checking process, the workstation determines whether the operation of the doctor is appropriate and gives the corresponding reference suggestions, which is responsible for patients and allows the doctor to continuously improve his ability during the checking process, thereby greatly reducing the pressure on doctors, and allowing doctors to focus more on other more creative tasks, and besides bringing huge economic and social benefits.

An endoscope system includes a storage medium that stores a plurality of correspondence between an imaging condition and a plurality of index values relating to a plurality of structures of an observation object, wherein the plurality of index values including a first index value acquirable under the first imaging condition and a second index value non-acquirable under the first imaging condition but acquirable under a second imaging condition, a monitor and a processor, coupled to the storage medium and the monitor. The processor is configured to: acquire an image of the observation object by using an endoscope; acquire a first imaging condition which represents an imaging condition of the image; refer to the plurality of index values and the imaging condition in the storage medium and extract the second imaging condition; and display guidance indicating that the second index value is acquirable under the extracted second imaging condition.

A deep learning-based examination and diagnosis assistance system and apparatus for early digestive tract cancer comprising a feature extraction network, an image classification model, an endoscope classifier, and an early cancer recognition model. The feature extraction network is used for performing initial feature extraction on endoscope images based on a neural network model; the image classification model is used for performing extraction on the initial features to acquire image classification features; the endoscope classifier is used for performing feature extraction on the initial features to acquire endoscope classification features and classify gastroscope/colonoscope images; the early cancer recognition model is used for splicing the initial features, the endoscope classification features, and the image classification features to acquire the probability of early cancer lesions in white light images, electronic dye images or chemical dye images of a corresponding site or acquire a flushing prompt or position recognition prompt for the corresponding site.

The present invention provides a transfer learning based capsule endoscopic images classification system. The system removes the capsule endoscopic images with an average brightness value beyond the preset threshold, and removes the capsule endoscopic images without details based on image brightness standard deviation and image brightness gradient. The system also removes similar images from the capsule endoscopic images using optical flow method, classifies the capsule endoscopic images according to the corresponding anatomical structure, and obtains the classified capsule endoscopic images list arranged in chronological order. The system further determines and labels the position of the first image of each specific anatomical structure in the classified capsule endoscopic images list arranged in chronological order.

The present invention relates to image analysis of pathology images. In order to improve reliability in image analysis of pathology images, a method is provided for providing support in identifying at least one feature of a tissue sample in a microscopic image. The method comprises the steps of providing a first image of a first microscopy 5 modality representing an area of the tissue sample, providing a second image of a second microscopy modality representing the said area of the tissue sample, generating a first high intensity image by applying a first high intensity filter to the first image or a first low intensity image by applying a first low intensity filter to the first image to obtain first information of the at least one feature, generating a second high intensity image by applying 10 a second high intensity filter to the second image or a second low intensity image by applying a second low intensity filter to the second image to obtain second information of the at least one feature, calculating a correlation of an image pair comprising one of the first high intensity image and the first low intensity image and one of the second high intensity image and the second low intensity image for correlating the first information and the second 15 information of the at least one feature, and outputting the calculated correlation for providing support in identifying the at least one feature of the tissue sample.

An information processing system including units configured to: acquire a pathologic tissue image of a patient having a target disease; divide the pathologic tissue image into region images; input each of the region images to each of a plurality of feature prediction models constructed one-to-one for types of histopathological features; sort a plurality of region images of which respective combinations of presence or absence of the histopathological features based on the acquisition match a previously set combination of presence or absence of the histopathological features at time of sorting; input each of the region images selected by the sorting to each of gene mutation prediction models constructed one-to-one for types of gene mutations, the gene mutation prediction models each having a combination of presence or absence of the histopathological features; and output a prediction result of presence or absence of at least one gene mutation in the patient.

A parasite analysis system includes a pressure vessel configured to store a biological sample, an imaging cell connected to the pressure vessel, and a waste depository connected to the imaging cell. An input valve controls whether biological sample can flow from the pressure vessel into the imaging cell and an output valve controls whether biological sample can flow from the imaging cell into the waste depository. The parasite analysis system also includes a camera that captures a chronological set of images of a portion of the biological sample in the imaging cell and an image analysis system that analyzes the chronological set of images to generate an estimate of a number of parasites in the portion of the biological sample. Estimates for multiple portions of the biological sample may be generated and sampling techniques used to estimate the number of parasites in the entire biological sample.

A method of visualising the three-dimensional internal surface of a lumen in real-time comprising using various combinations of image, motion and shape data obtained from an endoscope with a trained neural network and a curved lumen model. The three-dimensional internal surface may be unfolded to form a two-dimensional visualisation.

Systems and methods for improving endoscopy procedures are described that provide not only a conventional real time image of the view obtained by an endoscope, but in addition, a near real time 3D model and/or a 2D flattened image of an interior surface of an organ, which model and image may be processed using AI software to highlight potential tissue abnormalities for closer examination and/or biopsy during the procedure. A navigation module interacts with other system outputs to further assist the endoscopist with navigational indicia, e.g., landmarks and/or directional arrows, that enhance the endoscopists' spatial orientation, and/or may provide navigational guidance to the endoscopist to assist manipulation of the endoscope.