Pathological image similarity detection method and device

Abstract

The invention discloses a pathological image similarity detection method and device, and the method comprises the steps of obtaining a pathological image, labeling at least one region of interest on the pathological image, and classifying all labeled regions of interest; dividing all labeled regions of interest into two parts, one part serving as a training image, and the other part being used for establishing a pathological image library; segmenting the foreground of the training image into a plurality of image blocks, and allocating the same category as the regions of interest to the image blocks; extracting the features in the image blocks with the category labels by adopting a deep convolutional neural network model, and classifying the features, thereby training the deep convolutional neural network model; and inputting the test image into the trained deep convolutional neural network model, outputting the category to which the test image belongs by the deep convolutional neural network model, and according to the category to which the test image belongs, searching the pathological image of the corresponding category in the pathological image library. The method can quickly improve the radiograph reading capability of the pathologists and the credibility of the radiograph reading results.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a pathological image similarity detection method and detection device. Background technique [0002] In clinical medicine, pathology is one of the most important methods for diagnosing diseases. Pathological examination is also often referred to as the gold standard for disease diagnosis. Clinicians generally make an empirical diagnosis based on experience and knowledge, and the final diagnosis often requires pathologists to make a combined experiment. It often requires taking a piece of tissue from the patient's lesion, preparing slices, observing the morphology of cells and tissues under an electron microscope, and finally issuing a pathology report and giving a diagnosis opinion. The training process of pathologists is long and boring. To become a qualified pathologist often takes years or even ten years of rigorous training to complete. A pathologist needs to carefu...

AI Technical Summary

Problems solved by technology

The training process of pathologists is long and boring

Images