Ultrasonic image segmentation method and system based on deep learning

The invention provides an ultrasonic image segmentation method and system based on deep learning, and the method comprises the steps: carrying out the coarse segmentation of a to-be-segmented ultrasonic image through a backbone network, separating a plurality of organs from one ultrasonic image, and...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: LUO JIAJIA, WANG JIANLIU, GENG JING, SUN XIULI, ZHANG CHENXIN, ZUO JIANWEI
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention provides an ultrasonic image segmentation method and system based on deep learning, and the method comprises the steps: carrying out the coarse segmentation of a to-be-segmented ultrasonic image through a backbone network, separating a plurality of organs from one ultrasonic image, and obtaining a plurality of coarse segmentation result images and corresponding organ labels, and each coarse segmentation result image only comprises one organ segmentation result with noise, so that a good data basis is provided for inputting different organ labels into different branch networks for fine segmentation subsequently, the calculation amount during fine segmentation of the branch networks is reduced, and the overall segmentation efficiency is improved. Each branch network carries out fine segmentation on the coarse segmentation result image corresponding to the matched organ label according to the mapping relation between the organ label and the branch network, so that the plurality of branch networks c