Skip to main content

ORIGINAL RESEARCH article

Front. Oncol.
Sec. Head and Neck Cancer
Volume 14 - 2024 | doi: 10.3389/fonc.2024.1417330

De e p le arning-as s is te d diagnos is of be nign and malignant parotid gland tumors bas e d on automatic s e gme ntation of ultras ound image s : a multice nte r re tros pe ctive s tudy

Provisionally accepted
Wei Wei Wei Wei 1*Jingya Xu Jingya Xu 1Fei Xia Fei Xia 2Jun Liu Jun Liu 3Zekai Zhang Zekai Zhang 4Jing Wu Jing Wu 1Tianjun Wei Tianjun Wei 1Huijun Feng Huijun Feng 1Qiang Ma Qiang Ma 1Feng Jiang Feng Jiang 1Xiangming Zhu Xiangming Zhu 1Xia Zhang Xia Zhang 1
  • 1 First Affiliated Hospital of Wannan Medical College, Wuhu, China
  • 2 Second People's Hospital of Wuhu, Wuhu, Anhui Province, China
  • 3 Linyi Cancer Hospital, Linyi, Shandong Province, China
  • 4 Zibo Central Hospital, Shandong, China

The final, formatted version of the article will be published soon.

    Obje ctive s : To construct deep learning-assisted diagnosis models based on automatic segmentation of ultrasound images to aid radiologists in differentiating benign and malignant parotid tumors. A total of 582 patients histopathologically diagnosed with PGTs were retrospectively recruited from 4 centers, and their data were collected for analysis. The radiomics features of six deep learning models (ResNet18, Inception_v3, etc.) were analyzed based on the ultrasound images obtained under the best automatic segmentation model (Deeplabv3, UNet++, and UNet). The performance of three physicians was compared when the optimal model was used and not. The Net Reclassification Index (NRI) and Integrated Discrimination Improvement (IDI) were utilized to evaluate the clinical benefit of the optimal model. The Deeplabv3 model performed optimally in terms of automatic segmentation. The ResNet18 deep learning model had the best prediction performance, with an area under the receiver-operating characteristic curve of 0.808 (0.694-0.923), 0.809 (0.712-0.906), and 0.812 (0.680-0.944) in the internal test set and external test sets 1 and 2, respectively. Meanwhile, the optimal model-assisted clinical and overall benefits were markedly enhanced for two out of three radiologists (in internal validation set, NRI: 0.259 and 0.213 [p = 0.002 and 0.017], IDI: 0.284 and 0.201 [p = 0.005 and 0.043], respectively; in external test set 1, NRI: 0.183 and 0.161 [p = 0.019 and 0.008], IDI: 0.205 and 0.184 [p = 0.031 and 0.045], respectively; in external test set 2, NRI: 0.297 and 0.297 [p = 0.038 and 0.047], IDI: 0.332 and 0.294 [p = 0.031 and 0.041], respectively). The deep learning model constructed for automatic segmentation of ultrasound images can improve the diagnostic performance of radiologists for PGTs.

    Keywords: Automa tic s e gme nta tion, De e p le a rning, P a rotid gla nd tumors, Ultra s ound, Ne t Re cla s s ifica tion Inde x, Inte gra te d Dis crimina tion Improve me nt

    Received: 14 Apr 2024; Accepted: 18 Jul 2024.

    Copyright: © 2024 Wei, Xu, Xia, Liu, Zhang, Wu, Wei, Feng, Ma, Jiang, Zhu and Zhang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: Wei Wei, First Affiliated Hospital of Wannan Medical College, Wuhu, China

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.