Kasap Zeliha Aydın, Kurt Burçin, Güner Ali, Özsağır Elif, Ercin Mustafa Emre
Karadeniz Technical University, Faculty of Medicine, Department of Biostatistics and Medical Informatics, Trabzon, Turkey.
Karadeniz Technical University, Faculty of Medicine, Department of Biostatistics and Medical Informatics, Trabzon, Turkey.
Artif Intell Med. 2025 Sep;167:103186. doi: 10.1016/j.artmed.2025.103186. Epub 2025 May 30.
Atypia of Undetermined Significance (AUS), classified as Category III in the Bethesda Thyroid Cytopathology Reporting System, presents significant diagnostic challenges for clinicians. This study aims to develop a clinical decision support system that integrates structural equation modeling (SEM) and machine learning to predict malignancy in AUS thyroid nodules. The model integrates preoperative clinical data, ultrasonography (USG) findings, and cytopathological and morphometric variables. This retrospective cohort study was conducted between 2011 and 2019 at Karadeniz Technical University (KTU) Farabi Hospital. The dataset included 56 variables derived from 204 thyroid nodules diagnosed via ultrasound-guided fine-needle aspiration biopsy (FNAB) in 183 patients over 18 years. Logistic regression (LR) and SEM were used to identify risk factors for early thyroid cancer detection. Subsequently, machine learning algorithms-including Support Vector Machines (SVM), Naive Bayes (NB), and Decision Trees (DT) were used to construct decision support models. After feature selection with SEM, the SVM model achieved the highest performance, with an accuracy of 82 %, a specificity of 97 %, and an AUC value of 84 %. Additional models were developed for different scenarios, and their performance metrics were compared. Accurate preoperative prediction of malignancy in thyroid nodules is crucial for avoiding unnecessary surgeries. The proposed model supports more informed clinical decision-making by effectively identifying benign cases, thereby reducing surgical risk and improving patient care.
意义未明的非典型性病变（AUS），在贝塞斯达甲状腺细胞病理学报告系统中被归类为III类，给临床医生带来了重大的诊断挑战。本研究旨在开发一种临床决策支持系统，该系统整合结构方程模型（SEM）和机器学习，以预测AUS甲状腺结节的恶性程度。该模型整合了术前临床数据、超声检查（USG）结果以及细胞病理学和形态学变量。这项回顾性队列研究于2011年至2019年在特拉布宗技术大学（KTU）法拉比医院进行。数据集包括从183例18岁以上患者的204个经超声引导细针穿刺活检（FNAB）诊断的甲状腺结节中得出的56个变量。使用逻辑回归（LR）和SEM来确定早期甲状腺癌检测的风险因素。随后，使用包括支持向量机（SVM）、朴素贝叶斯（NB）和决策树（DT）在内的机器学习算法来构建决策支持模型。在通过SEM进行特征选择后，SVM模型表现最佳，准确率为82%，特异性为97%，AUC值为84%。针对不同情况开发了其他模型，并比较了它们的性能指标。准确术前预测甲状腺结节的恶性程度对于避免不必要的手术至关重要。所提出的模型通过有效识别良性病例来支持更明智的临床决策，从而降低手术风险并改善患者护理。
