Multi-mode cervical cancer MRI image automatic identification and segmentation method and system

The invention relates to a multi-mode cervical cancer MRI image automatic identification and segmentation method and system. The method comprises the following steps: firstly, acquiring nuclear magnetic resonance images of three modalities of T2WI, DWI and CE-T1WI, then classifying the nuclear magne...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: ZHANG XIAOYAN, KONG XIANGXING, WANG ZHINAN, XIA SHAOJUN, LIN TIANYE, SON EUNG-SIL, LI XIAOTING, ZHU HAITAO, YANG QINGMO
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 relates to a multi-mode cervical cancer MRI image automatic identification and segmentation method and system. The method comprises the following steps: firstly, acquiring nuclear magnetic resonance images of three modalities of T2WI, DWI and CE-T1WI, then classifying the nuclear magnetic resonance images of the three modalities by adopting an image classification network based on a ConvNeXt network, and predicting whether a tumor area exists in the images; and then, an image segmentation network based on a Double U-Net network is adopted to segment the nuclear magnetic resonance images of three modalities of T2WI, DWI and CE-T1WI with a tumor region, and an accurately segmented focus region image is obtained. The multi-modal cervical cancer MRI image recognition and segmentation integrated model is constructed based on the deep learning method, an automatic sketching tool is provided for clinical application, and the technical problems of poor robustness, low efficiency and the like existing in