Machine learning-based Radiomics analysis for differentiation degree and lymphatic node metastasis of extrahepatic cholangiocarcinoma

• Published in: BMC cancer 2021-11, Vol.21 (1), p.1268-1268, Article 1268
• Main Authors: Tang, Yong, Yang, Chun Mei, Su, Song, Wang, Wei Jia, Fan, Li Ping, Shu, Jian
• Format: Article
• Language: English
• Subjects: Abdomen; Adenocarcinoma - diagnostic imaging; Adenocarcinoma - pathology; Adenocarcinoma - secondary; Adult; Aged; Aged, 80 and over; Algorithms; Area Under Curve; Bile Duct Neoplasms - diagnostic imaging; Bile Duct Neoplasms - pathology; Bile ducts; Cancer; Cell differentiation; Cholangiocarcinoma; Cholangiocarcinoma - diagnostic imaging; Cholangiocarcinoma - pathology; Cholangiocarcinoma - secondary; Classification; Diffusion coefficient; Extrahepatic cholangiocarcinoma; Feature selection; Female; Health aspects; Humans; Learning algorithms; Lymph nodes; Lymphatic metastasis; Lymphatic Metastasis - diagnostic imaging; Lymphatic Metastasis - pathology; Lymphatic system; Machine Learning; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Medical prognosis; Metastases; Metastasis; Middle Aged; Neoplasm Grading; Patients; Performance evaluation; Predictions; Predictive Value of Tests; Radiography, Medical; Radiomics; Retrospective Studies; Risk factors; ROC Curve; Sensitivity and Specificity; Statistical analysis; Technology application; Tomography; Tumors; Variance analysis; Wavelet transforms
• ISSN: 1471-2407; 1471-2407
• DOI: 10.1186/s12885-021-08947-6

Radiomics may provide more objective and accurate predictions for extrahepatic cholangiocarcinoma (ECC). In this study, we developed radiomics models based on magnetic resonance imaging (MRI) and machine learning to preoperatively predict differentiation degree (DD) and lymph node metastasis (LNM) of ECC. A group of 100 patients diagnosed with ECC was included. The ECC status of all patients was confirmed by pathology. A total of 1200 radiomics features were extracted from axial T1 weighted imaging (T1WI), T2-weighted imaging (T2WI), diffusion weighted imaging (DWI), and apparent diffusion coefficient (ADC) images. A systematical framework considering combinations of five feature selection methods and ten machine learning classification algorithms (classifiers) was developed and investigated. The predictive capabilities for DD and LNM were evaluated in terms of area under precision recall curve (AUPRC), area under the receiver operating characteristic (ROC) curve (AUC), negative predictive value (NPV), accuracy (ACC), sensitivity, and specificity. The prediction performance among models was statistically compared using DeLong test. For DD prediction, the feature selection method joint mutual information (JMI) and Bagging Classifier achieved the best performance (AUPRC = 0.65, AUC = 0.90 (95% CI 0.75-1.00), ACC = 0.85 (95% CI 0.69-1.00), sensitivity = 0.75 (95% CI 0.30-0.95), and specificity = 0.88 (95% CI 0.64-0.97)), and the radiomics signature was composed of 5 selected features. For LNM prediction, the feature selection method minimum redundancy maximum relevance and classifier eXtreme Gradient Boosting achieved the best performance (AUPRC = 0.95, AUC = 0.98 (95% CI 0.94-1.00), ACC = 0.90 (95% CI 0.77-1.00), sensitivity = 0.75 (95% CI 0.30-0.95), and specificity = 0.94 (95% CI 0.72-0.99)), and the radiomics signature was composed of 30 selected features. However, these two chosen models were not significantly different to other models of higher AUC values in DeLong test, though they were significantly different to most of all models. MRI radiomics analysis based on machine learning demonstrated good predictive accuracies for DD and LNM of ECC. This shed new light on the noninvasive diagnosis of ECC.