Duan Bo, Pan Li-Li, Chen Wen-Xia, Qiao Zhong-Wei, Xu Zheng-Min
Department of Otolaryngology-Head and Neck Surgery, Children's Hospital of Fudan University, Shanghai, China.
Department of Radiology, Children's Hospital of Fudan University, Shanghai, China.
Front Pediatr. 2022 Aug 9;10:809523. doi: 10.3389/fped.2022.809523. eCollection 2022.
This study aimed to conduct an in-depth investigation of the learning framework used for deriving diagnostic results of temporal bone diseases, including cholesteatoma and Langerhans cell histiocytosis (LCH). In addition, middle ear inflammation (MEI) was diagnosed by CT scanning of the temporal bone in pediatric patients.
A total of 119 patients were included in this retrospective study; among them, 40 patients had MEI, 38 patients had histology-proven cholesteatoma, and 41 patients had histology-proven LCH of the temporal bone. Each of the 119 patients was matched with one-third of the disease labels. The study included otologists and radiologists, and the reference criteria were histopathology results (70% of cases for training and 30% of cases for validation). A multilayer perceptron artificial neural network (VGG16_BN) was employed and classified, based on radiometrics. This framework structure was compared and analyzed by clinical experts according to CT images and performance.
The deep learning framework results vs. a physician's diagnosis, respectively, in multiclassification tasks, were as follows. Receiver operating characteristic (ROC) (cholesteatoma): (0.98 vs. 0.91), LCH (0.99 vs. 0.98), and MEI (0.99 vs. 0.85). Accuracy (cholesteatoma): (0.99 vs. 0.89), LCH (0.99 vs. 0.97), and MEI (0.99 vs. 0.89). Sensitivity (cholesteatoma): (0.96 vs. 0.97), LCH (0.99 vs. 0.98), and MEI (1 vs. 0.69). Specificity (cholesteatoma): (1 vs. 0.89), LCH (0.99 vs. 0.97), and MEI (0.99 vs. 0.89).
This article presents a research and learning framework for the diagnosis of cholesteatoma, MEI, and temporal bone LCH in children, based on CT scans. The research framework performed better than the clinical experts.
本研究旨在深入调查用于得出颞骨疾病诊断结果的学习框架，这些疾病包括胆脂瘤和朗格汉斯细胞组织细胞增多症（LCH）。此外，通过对儿科患者颞骨进行CT扫描来诊断中耳炎症（MEI）。
本回顾性研究共纳入119例患者；其中，40例患有MEI，38例经组织学证实患有胆脂瘤，41例经组织学证实患有颞骨LCH。119例患者中的每一位都与三分之一的疾病标签进行匹配。该研究包括耳科医生和放射科医生，参考标准为组织病理学结果（70%的病例用于训练，30%的病例用于验证）。采用多层感知器人工神经网络（VGG16_BN）并基于放射测量学进行分类。临床专家根据CT图像和性能对该框架结构进行了比较和分析。
在多分类任务中，深度学习框架结果与医生诊断结果分别如下。受试者操作特征曲线（ROC）（胆脂瘤）：（0.98对0.91），LCH（0.99对0.98），MEI（0.99对0.85）。准确率（胆脂瘤）：（0.99对0.89），LCH（0.99对0.97），MEI（0.99对0.89）。敏感性（胆脂瘤）：（0.96对0.97），LCH（0.99对0.98），MEI（1对0.69）。特异性（胆脂瘤）：（1对0.89），LCH（0.99对0.97），MEI（0.99对0.89）。
本文提出了一种基于CT扫描诊断儿童胆脂瘤、MEI和颞骨LCH的研究与学习框架。该研究框架的表现优于临床专家。
