Wang Yi, Gao Jiening, Yin Zhaolin, Wen Yue, Sun Meng, Han Ruoling
Department of Ultrasound, The Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei, China.
Front Oncol. 2024 May 13;14:1384105. doi: 10.3389/fonc.2024.1384105. eCollection 2024.
The pathological classification and imaging manifestation of parotid gland tumors are complex, while accurate preoperative identification plays a crucial role in clinical management and prognosis assessment. This study aims to construct and compare the performance of clinical models, traditional radiomics models, deep learning (DL) models, and deep learning radiomics (DLR) models based on ultrasound (US) images in differentiating between benign parotid gland tumors (BPGTs) and malignant parotid gland tumors (MPGTs).
Retrospective analysis was conducted on 526 patients with confirmed PGTs after surgery, who were randomly divided into a training set and a testing set in the ratio of 7:3. Traditional radiomics and three DL models (DenseNet121, VGG19, ResNet50) were employed to extract handcrafted radiomics (HCR) features and DL features followed by feature fusion. Seven machine learning classifiers including logistic regression (LR), support vector machine (SVM), RandomForest, ExtraTrees, XGBoost, LightGBM and multi-layer perceptron (MLP) were combined to construct predictive models. The most optimal model was integrated with clinical and US features to develop a nomogram. Receiver operating characteristic (ROC) curve was employed for assessing performance of various models while the clinical utility was assessed by decision curve analysis (DCA).
The DLR model based on ExtraTrees demonstrated superior performance with AUC values of 0.943 (95% CI: 0.918-0.969) and 0.916 (95% CI: 0.861-0.971) for the training and testing set, respectively. The combined model DLR nomogram (DLRN) further enhanced the performance, resulting in AUC values of 0.960 (95% CI: 0.940- 0.979) and 0.934 (95% CI: 0.876-0.991) for the training and testing sets, respectively. DCA analysis indicated that DLRN provided greater clinical benefits compared to other models.
DLRN based on US images shows exceptional performance in distinguishing BPGTs and MPGTs, providing more reliable information for personalized diagnosis and treatment plans in clinical practice.
腮腺肿瘤的病理分类和影像学表现复杂，而术前准确鉴别对临床治疗及预后评估至关重要。本研究旨在构建并比较基于超声（US）图像的临床模型、传统放射组学模型、深度学习（DL）模型及深度学习放射组学（DLR）模型在鉴别腮腺良性肿瘤（BPGT）与腮腺恶性肿瘤（MPGT）方面的性能。
对526例术后确诊为腮腺肿瘤（PGT）的患者进行回顾性分析，将其按7:3随机分为训练集和测试集。采用传统放射组学及三种DL模型（DenseNet121、VGG19、ResNet50）提取手工放射组学（HCR）特征及DL特征，随后进行特征融合。结合逻辑回归（LR）、支持向量机（SVM）、随机森林、极端随机树、XGBoost、LightGBM和多层感知器（MLP）这七种机器学习分类器构建预测模型。将最优模型与临床及US特征整合以制定列线图。采用受试者操作特征（ROC）曲线评估各模型性能，同时通过决策曲线分析（DCA）评估临床实用性。
基于极端随机树的DLR模型表现出色，训练集和测试集的AUC值分别为0.943（95%CI：0.918 - 0.969）和0.916（95%CI：0.861 - 0.971）。联合模型DLR列线图（DLRN）进一步提升了性能，训练集和测试集的AUC值分别为0.960（95%CI：0.940 - 0.979）和0.934（95%CI：0.876 - 0.991）。DCA分析表明，与其他模型相比，DLRN具有更大的临床益处。
基于US图像的DLRN在区分BPGT和MPGT方面表现卓越，为临床实践中的个性化诊断和治疗方案提供了更可靠的信息。
