Method for detecting skull abnormity of fetus based on machine learning and system

Abstract

The invention discloses a method for detecting skull abnormity of a fetus based on machine learning. Particularly, whether a major tissue structure of a skull is abnormal or not and which related skull diseases will be caused by abnormity are detected during the development of the skull of the fetus. According to the method, whether the major tissue structure of the skull is abnormal or not is mainly detected through acquiring data of standard sections of skulls of different gestational week series, preprocessing the data and training a model. The method comprises the steps: extracting features by adopting a deep convolutional network, generating candidate regions by a region proposal network RPN, collecting an input characteristic graph and the candidate regions by an interested pooling layer, carrying out classifying and regression detection by using a softmax classifier, and finally analyzing whether a primary structure is abnormal or not by using a detection result; and judging that the primary structure is normal if no abnormity is detected. The method and the system have the aim of aiding diagnosis by using a computer, and whether the skull is abnormal or not is diagnosed inan aided manner under the condition that doctors or people have no need of excessively participating in diagnosis.

Description

technical field [0001] The invention belongs to the technical field of computer-aided diagnosis, and more specifically relates to a method and system for detecting fetal brain abnormalities based on machine learning. Background technique [0002] During the development of the whole body of the fetus, the healthy development of the brain is of great significance. Brain hypoplasia will directly affect the intelligence of the fetus, and seriously cause cerebral palsy, mental retardation, mental retardation, and epilepsy. In view of this, it is necessary to conduct a detailed examination of the fetal brain by prenatal ultrasound. [0003] However, there are some non-negligible shortcomings in the existing ultrasound for fetal brain detection: first, because the fetal brain detection process is quite complicated, and there is a serious shortage of sonographers with rich clinical experience and good at prenatal detection of fetal brain abnormalities, so Greatly increased the work...

AI Technical Summary

Problems solved by technology

[0004] In view of the above defects or improvement needs of the prior art, the present invention provides a method and system for detecting fetal brain abnormalities based on machine learning. Ultrasound image data of normal and abnormal fetal brain solves the above-mentioned technical problems existing in the detection of fetal brain based on ultrasonic imaging. The present invention combines the gestational age and medical history of the fetus during the real-time ultrasonic scanning process of the fetus. Real-time and uninterrupted monitoring and analysis of fetal cranial ultrasound images intelligently identify a series of standard sections of fetal brain development, automatically acquire and store the intelligently identified standard sections, and finally automatically measure and analyze fetal brain development parameters, and identify Possible fetal brain abnormalities

Images