PloS one · 2025
A severity classification model of cervical spondylotic radiculopathy symptoms based on MRI radiomics: A retrospective study.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Wang Xi, Tao Qiaoli, Liu Huanwen, Lin Huangbo, Zhang Honglai
Аннотация
OBJECTIVE: To develop a severity classification model for symptoms of cervical spondylotic radiculopathy (CSR) based on magnetic resonance imaging (MRI) radiomics and to evaluate the predictive value of MRI radiomics features in the classification of symptoms severity, providing an objective basis for personalized therapeutic interventions.
METHODS: This retrospective study included 99 patients diagnosed with CSR, admitted between August 2022 and April 2023. Symptom severity was assessed using the neck disability index (NDI) scale, which facilitated the categorization of participants into mild and severe symptoms groups. A comprehensive set of 3,404 quantitative radiomics features was extracted from four predefined regions of interest (ROIs) using the 3D Slicer software. The least absolute shrinkage and selection operator (LASSO) regression analysis was used to identify the optimal subsets of radiomics features, which were subsequently used to develop a support vector machine (SVM) classification model. Model performance was evaluated using receiver operating characteristic (ROC) curve analysis with area under the curve (AUC) calculations, complemented by accuracy, precision, sensitivity and F1 score evaluations.
RESULTS: Analysis of cervical T2-weighted MRI resulted in the extraction of 3,404 radiomics features from four ROIs. Using the LASSO regression for feature selection, 96 radiomics features were retained for model construction. The most discriminatory characteristics were in the intervertebral discs at levels C4/5, C5/6, and C6/7 in the mid-sagittal plane. The model demonstrated an AUC of 0.91. The accuracy of the model is 0.917, the precision is 0.979, the sensitivity is 0.833, and the F1 score is 0.890.
CONCLUSION: The severity classification model of CSR symptoms based on MRI radiomics demonstrates robust predictive performance in assessing the severity of CSR symptoms, serving as an effective decision-support tool for guiding personalized therapeutic strategies in clinical practice.
METHODS: This retrospective study included 99 patients diagnosed with CSR, admitted between August 2022 and April 2023. Symptom severity was assessed using the neck disability index (NDI) scale, which facilitated the categorization of participants into mild and severe symptoms groups. A comprehensive set of 3,404 quantitative radiomics features was extracted from four predefined regions of interest (ROIs) using the 3D Slicer software. The least absolute shrinkage and selection operator (LASSO) regression analysis was used to identify the optimal subsets of radiomics features, which were subsequently used to develop a support vector machine (SVM) classification model. Model performance was evaluated using receiver operating characteristic (ROC) curve analysis with area under the curve (AUC) calculations, complemented by accuracy, precision, sensitivity and F1 score evaluations.
RESULTS: Analysis of cervical T2-weighted MRI resulted in the extraction of 3,404 radiomics features from four ROIs. Using the LASSO regression for feature selection, 96 radiomics features were retained for model construction. The most discriminatory characteristics were in the intervertebral discs at levels C4/5, C5/6, and C6/7 in the mid-sagittal plane. The model demonstrated an AUC of 0.91. The accuracy of the model is 0.917, the precision is 0.979, the sensitivity is 0.833, and the F1 score is 0.890.
CONCLUSION: The severity classification model of CSR symptoms based on MRI radiomics demonstrates robust predictive performance in assessing the severity of CSR symptoms, serving as an effective decision-support tool for guiding personalized therapeutic strategies in clinical practice.