Explainable Machine Learning Techniques To Predict Amiodarone-Induced Thyroid Dysfunction Risk: Multicenter, Retrospective Study With External Validation

Abstract

Background Machine learning offers new solutions for predicting life-threatening, unpredictable amiodarone-induced thyroid dysfunction. Traditional regression approaches for adverse-effect prediction without time-series consideration of features have yielded suboptimal predictions. Machine learning algorithms with multiple data sets at different time points may generate better performance in predicting adverse effects. Objective We aimed to develop and validate machine learning models for forecasting individualized amiodarone-induced thyroid dysfunction risk and to optimize a machine learning–based risk stratification scheme with a resampling method and readjustment of the clinically derived decision thresholds. Methods This study developed machine learning models using multicenter, delinked electronic health records. It included patients receiving amiodarone from January 2013 to December 2017. The training set was composed of data from Taipei Medical University Hospital and Wan Fang Hospital, while data from Taipei Medical University Shuang Ho Hospital were used as the external test set. The study collected stationary features at baseline and dynamic features at the first, second, third, sixth, ninth, 12th, 15th, 18th, and 21st months after amiodarone initiation. We used 16 machine learning models, including extreme gradient boosting, adaptive boosting, k-nearest neighbor, and logistic regression models, along with an original resampling method and 3 other resampling methods, including oversampling with the borderline-synthesized minority oversampling technique, undersampling–edited nearest neighbor, and over- and undersampling hybrid methods. The model performance was compared based on accuracy; Precision, recall, F1-score, geometric mean, area under the curve of the receiver operating characteristic curve (AUROC), and the area under the precision-recall curve (AUPRC). Feature importance was determined by the best model. The decision threshold was readjusted to identify the best cutoff value and a Kaplan-Meier survival analysis was performed. Results The training set contained 4075 patients from Taipei Medical University Hospital and Wan Fang Hospital, of whom 583 (14.3%) developed amiodarone-induced thyroid dysfunction, while the external test set included 2422 patients from Taipei Medical University Shuang Ho Hospital, of whom 275 (11.4%) developed amiodarone-induced thyroid dysfunction. The extreme gradient boosting oversampling machine learning model demonstrated the best predictive outcomes among all 16 models. The accuracy; Precision, recall, F1-score, G-mean, AUPRC, and AUROC were 0.923, 0.632, 0.756, 0.688, 0.845, 0.751, and 0.934, respectively. After readjusting the cutoff, the best value was 0.627, and the F1-score reached 0.699. The best threshold was able to classify 286 of 2422 patients (11.8%) as high-risk subjects, among which 275 were true-positive patients in the testing set. A shorter treatment duration; higher levels of thyroid-stimulating hormone and high-density lipoprotein cholesterol; and lower levels of free thyroxin, alkaline phosphatase, and low-density lipoprotein were the most important features. Conclusions Machine learning models combined with resampling methods can predict amiodarone-induced thyroid dysfunction and serve as a support tool for individualized risk prediction and clinical decision support.

Designing Interpretable ML System to Enhance Trustworthy AI in Healthcare: A Systematic Review of the Last Decade to A Proposed Robust Framework

“…This process involved identifying crucial risk factors for predicting subsequent parental stress. SHAP techniques were developed to predict the risk of amiodarone-induced thyroid dysfunction; which served as a tool for personalized risk assessment and helped in clinical decision-making [17]. An interpretable model using the SHAP approach enhanced mental stress detection using heart rate variability data.…”

“…XAI has numerous applications across healthcare and medicine. Currently, interpretability is the primary focus in the medical domain to provide easy-to-understand results for patients and caregivers and enhance their trust in specific applications such as drug discovery [16], thyroid [17], parental stress [18], wearable health trackers and biosensors [19], and respiratory disease [20]. The black-box issue in AI arises when a system faces challenges to provide a clear explanation for how the model achieved a decision; which black-box, gray-box, and white-box terms are used to describe different levels of transparency regarding the inner workings of machine learning algorithms.…”

Lipoprotein alterations in endocrine disorders - a review of the recent developments in the field

“…Recently, machine learning models, which can predict the occurrence of amiodarone-induced hyperthyroidism, were described. This study found that the most important baseline risk factors for amiodarone-induced thyroid dysfunction were higher HDL-C, TSH, lower free thyroxine(FT4), alkaline phosphatase, and LDL-C, and shorter duration of treatment (18). Such tools are still far from being used in routine clinical practice, but they could significantly improve patients' safety if available.…”

Predicting Colorectal Cancer Survival Using Time-to-Event Machine Learning: Retrospective Cohort Study

“…Model transparency is critical to the application of models in the medical domain. Therefore, to make time-to-event ML models more transparent, we introduced SHAP, which is a model-agnostic post hoc explanation algorithm that has been widely applied to explain ML models [10,38,39].…”