Research indexJournal article
Precision Thyroid Detection using Explainable ML
Narayanswamy, Kumar, L. and 2 others
International Journal of Data Mining Techniques and Applications31 Dec 2025
NobleIDark:/48914/w1/20261105/6070F7BADOI10.20894/ijdmta.102.014.002.004
Precision Thyroid Detection using Explainable ML
Narayanswamy, Kumar, L. and 2 others · International Journal of Data Mining Techniques and Applications · 2025 · Subscription
NobleIDark:/48914/w1/20261105/6070F7BADOI10.20894/ijdmta.102.014.002.004
The full text is at the publisher
You can read the abstract here, or open it at the publisher.
Abstract
Thyroid disorders occur due to hormonal imbalance imbalances involving triiodothyronine (T3), thyroxine (T4), and thyroid-stimulating hormone (TSH), which affect imbalances, metabolic regulation and overall body function. Early and accurate detection of thyroid dysfunction is crucial to minimize complications and ensure timely treatment. This paper presents a comparative machine learning framework for classifying thyroid diseases such as hypothyroidism, hyperthyroidism, and euthyroidism. A real-world dataset was preprocessed to remove missing values and normalized for efficient model training. Various supervised algorithms, Logistic Regression, affect logistic regression, Random Forest, logistic regression, random forest, Support Vector Machine random forest, support vector machine (SVM), Naïve Bayes, k- Nearest Neighbor support vector machine nearest neighbors (KNN), and artificial neural network nearest neighbors (ANN), were implemented using Python. The performance of each model was evaluated using metrics such as accuracy, precision, recall, and F1-score. Results show that the ANN achieved the highest accuracy of 96.75%, followed by the logistic regression and random forest models. The proposed model demonstrates that AI-based approaches can effectively classify thyroid dysfunctions, providing an efficient diagnostic support system for healthcare professionals.