Medical image focus detection positioning method and device, electronic device and storage medium

Abstract

The invention provides a medical image focus detection positioning method and device, an electronic device and a storage medium. The method comprises the following steps: registering a medical image template to a to-be-detected medical image to obtain a registration transformation relation, and mapping anatomical position information in the medical image template to the to-be-detected medical image according to the registration transformation relation to obtain anatomical position information corresponding to different pixel points in the to-be-detected medical image; then, the abnormal shooting suspected focus in the bone whole body gamma plain film image is intelligently detected. Therefore, a suspected focus area in the to-be-detected medical image is automatically detected; Anatomicalposition information corresponding to a suspected focus area is intelligently provided, and auxiliary diagnosis of suspected focus properties by clinicians is facilitated by combining image characteristics such as contrast and shape of a suspected focus and referring to other related information such as inspection and clinic of a patient.

Description

technical field [0001] The invention relates to the technical field of medical images, in particular to a method for detecting and locating a lesion in a medical image, a method for detecting and locating a lesion in a medical image, a device, an electronic device, and a storage medium. Background technique [0002] Whole-body gamma film images of bones, also known as whole-body bone imaging or whole-body bone scan, detect the radioactive distribution of bones throughout the body through radioactive imaging instruments (such as gamma cameras, ECT). If the absorption of radioactivity in a certain bone increases or decreases abnormally, there is abnormal concentration or sparseness of radioactivity, and the abnormal absorption of radioactivity in bone in bone scan is just a reflection of abnormal bone metabolism. It is generally believed clinically that bone scans can detect tumor metastases earlier than X-ray or CT examinations, which can be as early as 3 to 6 months. [000...

AI Technical Summary

Problems solved by technology

However, in the process of realizing the present invention, the inventors found that the above-mentioned way of detecting lesions has at least the following problems: for the way of BP neural network to detect lesions on whole-body gamma film images of bones, the algorithm uses a single pixel as Judging the object requires a large amount of calculation, and the convergence speed of the BP neural network is slow, and does not give the anatomical position information corresponding to a single pixel

Based on the kernel density estimation and K-means clustering algorithm, the lesion detection of whole-body gamma film images of bones is performed. The initial clustering center of this algorithm depends on the calculation of the Gaussian kernel density estimation function. The detection accuracy of lesions is not high, and there are also no The problem of giving information about the anatomical location of the lesion

Images