ai
Researchers Develop Enhanced Machine Learning Model for Thyroid Disease Classification
A study published in Scientific Reports presents an enhanced extreme learning machine model that incorporates Drop-Connect regularization for thyroid disease detection. The model was tested on a four-class classification task using publicly available data.
May 19, 8:00 PM·1m read1 source
ndtv.comAudio version
Tap play to generate a narrated version.
Developing·Limited corroboration so far. This page will refresh as more sources emerge.
Researchers have introduced an Enhanced Extreme Learning Machine model that applies Drop-Connect regularization to improve performance in thyroid disease classification. The model was evaluated on a four-class task covering hypothyroidism, hyperthyroidism, sick-euthyroid, and normal cases. Under 10-fold cross-validation, it reached an average accuracy of approximately 82 percent.
89 percent accuracy. Performance was measured using accuracy, precision, recall, specificity, sensitivity, F1-score, ROC, and AUC metrics. 05. The framework follows a seven-step pipeline that includes data preprocessing, model building, training, and evaluation. All datasets used were obtained from publicly available sources.
The work received funding from Princess Nourah bint Abdulrahman University in Saudi Arabia, the Ministry of Higher Education in Malaysia, and Multimedia University. Corresponding authors are Sikandar Ali and It Ee Lee. The study was published online on 20 May 2026.
