Endocrine · 2024
Ultrasound-based radiomics to predict the volume reduction rate of benign thyroid nodules after microwave ablation.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
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.
Chen Fang, Gao Yuxiu, Xue Qingwen, Niu Xiaoyan, Zhang Xiaojuan, Zang Yichen, Zhang Hui, Li Shuao, Zhao Cheng
Аннотация
OBJECTIVE: To evaluate the predictive power of ultrasound-based radiomics models for benign thyroid nodules with a volume reduction rate (VRR) of < or ≥75% at 12 months after microwave ablation.
METHODS: A retrospective study was conducted on 194 individuals with benign thyroid nodules who received ultrasound-guided microwave ablation between November 2019 and June 2023. The clinical and ultrasound features, including age, gender, volume, echogenicity, duration of ablation, and so on were analysed by t-test or chi-square test. Radiomics features were extracted from longitudinal and transverse ultrasound images of the nodules. The features were selected using methods such as least absolute shrinkage and selection operator (LASSO). Radiomics models were established using longitudinal, transverse, and longitudinal + transverse ultrasound images to predict the VRR of benign thyroid nodules after ablation. Decision curve analysis (DCA) and receiver operating characteristic (ROC) curve analysis were used to assess the models' performance.
RESULTS: At 12 months following ablation, the VRR of the nodules was 77.8 ± 19.4% (7.4-98.8%). Statistical analysis revealed that the duration of ablation and the proportion of liquid extracted were significantly correlated with the 12-month VRR (P <0.05). In the radiomics models, Logistic Regression (LR) performed the best. In the training cohorts, the area under the curve (AUC) for the longitudinal, transverse, and combined groups were 0.935, 0.800, and 0.937. The AUC values in the test cohort were 0.820, 0.844, and 0.917.
CONCLUSION: The radiomics models established based on pre-ablation ultrasound images showed good predictive efficacy for the VRR of nodules at 12 months following ablation. The predictive efficacy is best in the combined group. With the models, we can preoperatively predict patients' prognoses and thereby determine whether to proceed with ablation therapy.
METHODS: A retrospective study was conducted on 194 individuals with benign thyroid nodules who received ultrasound-guided microwave ablation between November 2019 and June 2023. The clinical and ultrasound features, including age, gender, volume, echogenicity, duration of ablation, and so on were analysed by t-test or chi-square test. Radiomics features were extracted from longitudinal and transverse ultrasound images of the nodules. The features were selected using methods such as least absolute shrinkage and selection operator (LASSO). Radiomics models were established using longitudinal, transverse, and longitudinal + transverse ultrasound images to predict the VRR of benign thyroid nodules after ablation. Decision curve analysis (DCA) and receiver operating characteristic (ROC) curve analysis were used to assess the models' performance.
RESULTS: At 12 months following ablation, the VRR of the nodules was 77.8 ± 19.4% (7.4-98.8%). Statistical analysis revealed that the duration of ablation and the proportion of liquid extracted were significantly correlated with the 12-month VRR (P <0.05). In the radiomics models, Logistic Regression (LR) performed the best. In the training cohorts, the area under the curve (AUC) for the longitudinal, transverse, and combined groups were 0.935, 0.800, and 0.937. The AUC values in the test cohort were 0.820, 0.844, and 0.917.
CONCLUSION: The radiomics models established based on pre-ablation ultrasound images showed good predictive efficacy for the VRR of nodules at 12 months following ablation. The predictive efficacy is best in the combined group. With the models, we can preoperatively predict patients' prognoses and thereby determine whether to proceed with ablation therapy.