Multi-modality deep learning model reaches high prediction accuracy in the diagnosis of ovarian cancer
Zimo Wang,
Shuyu Luo,
Jing Chen,
Yang Jiao,
Chen Cui,
Siyuan Shi,
Yang Yang,
Junyi Zhao,
Yitao Jiang,
Yujuan Zhang,
Fanhua Xu,
Jinfeng Xu,
Qi Lin,
Fajin Dong
Affiliations
Zimo Wang
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China
Shuyu Luo
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China
Jing Chen
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China
Yang Jiao
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China
Fanhua Xu
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China
Jinfeng Xu
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China; Corresponding author
Qi Lin
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China; Corresponding author
Fajin Dong
Second Clinical College of Jinan University, Department of Ultrasound, Shenzhen People’s Hospital, First Affiliated Hospital of Southern University of Science and Technology, Shenzhen Medical Ultrasound Engineering Center. Shenzhen, Guangdong 518020, China; Corresponding author
Summary: We evaluated the diagnostic performance of a multimodal deep-learning (DL) model for ovarian mass differential diagnosis. This single-center retrospective study included 1,054 ultrasound (US)-detected ovarian tumors (699 benign and 355 malignant). Patients were randomly divided into training (n = 675), validation (n = 169), and testing (n = 210) sets. The model was developed using ResNet-50. Three DL-based models were proposed for benign-malignant classification of these lesions: single-modality model that only utilized US images; dual-modality model that used US images and menopausal status as inputs; and multi-modality model that integrated US images, menopausal status, and serum indicators. After 5-fold cross-validation, 210 lesions were tested. We evaluated the three models using the area under the curve (AUC), accuracy, sensitivity, and specificity. The multimodal model outperformed the single- and dual-modality models with 93.80% accuracy and 0.983 AUC. The Multimodal ResNet-50 DL model outperformed the single- and dual-modality models in identifying benign and malignant ovarian tumors.
