Artificial Intelligence in CT for Predicting Cervical Lymph Node Metastasis in Papillary Thyroid Cancer Patients: A Meta-analysis

Academic Radiology, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 1, 2025

Language: Английский

Prediction of peripheral lymph node metastasis (LNM) in thyroid cancer using delta radiomics derived from enhanced CT combined with multiple machine learning algorithms

European journal of medical research, Journal Year: 2025, Volume and Issue: 30(1)

Published: March 13, 2025

This study aimed to develop a model for predicting peripheral lymph node metastasis (LNM) in thyroid cancer patients by combining enhanced CT radiomic features with machine learning algorithms. It increased the clinical utility and interpretability of predictions through SHAP (SHapley Additive exPlanation) values nomograms explanation visualization. Clinical image data from 375 confirmed postoperative pathology at Xiangyang No. 1 People's Hospital were collected January 2015 July 2023. Among them, there 88 LNM group 287 non-LNM group. The delta tumours extracted. Various algorithms (such as SVM, GBM, RF, XGBoost, KNN, LightGBM) trained on feature sets used construct reliable prediction model. During training, cross-validation was evaluate performance, optimal selected. In addition, interpret results model, analyse contribution each results, further nomogram visually display results. Univariate analysis that sex, Hashimoto's disease, tumour adjacency capsule, pathological subtype, Delta Radscore, Radscore are risk factors patients. based radiomics performed well test set, achieved high AUC (0.879), sensitivity (0.849), specificity (0.769) values. Through value analysis, importance clarified, providing more detailed intuitive basis decision-making. illustrated process, facilitating understanding application clinicians. successfully constructed combined improved nomograms. not only improves accuracy but also provides scientific decision-making, potential value.

Language: Английский

Predicting central lymph node metastasis in papillary thyroid cancer: A nomogram based on clinical, ultrasound and contrast‑enhanced computed tomography characteristics

Oncology Letters, Journal Year: 2024, Volume and Issue: 28(4)

Published: Aug. 5, 2024

Central lymph node (CLN) status is considered to be an important risk factor in patients with papillary thyroid carcinoma (PTC). The aim of the present study was identify factors associated CLN metastasis (CLNM) for PTC based on preoperative clinical, ultrasound (US) and contrast-enhanced computed tomography (CT) characteristics, establish a prediction model treatment plans. A total 786 confirmed pathological diagnosis between January 2021 December 2022 were included retrospective study, 550 training group 236 enrolled validation (ratio 7:3). Based US CT features, univariate multivariate logistic regression analyses used determine independent predictive CLNM, personalized nomogram constructed. Calibration curve, receiver operating characteristic (ROC) curve decision assess discrimination, calibration clinical application model. As result, 38.9% (306/786) CLNM(-) before surgery had CLNM using postoperative pathology. In analysis, young age (≤45 years), male sex, no presence Hashimoto thyroiditis, isthmic location, microcalcification, inhomogeneous enhancement capsule invasion predictors PTC. integrating these 7 exhibited strong discrimination both [Area under (AUC)=0.826] (AUC=0.818). Furthermore, area ROC predicting features higher than that without (AUC=0.818 AUC=0.712, respectively). addition, appropriately fitted analysis utility nomogram. conclusion, developed novel which could provide basis prophylactic central dissection

Language: Английский