A novel hybrid approach for thyroid disease detection: Integrating cuttlefish algorithm and simulated annealing for optimal feature selection

Effective treatment relies on a timely diagnosis, which is critical in the case of thyroid disorder—one of the chronic endocrine disorders alongside diabetes and obesity—with profound health concerns. Thyroid disorders occur due to the malfunctioning of the thyroid gland, which may result in an imbalanced metabolic rate due to inappropriate hormone levels synthesis. An overactive gland results in hyperthyroidism, whereas an underactive or sluggish thyroid lead to hypothyroidism. Both disorders, if not detected and managed timely, can lead to severe health complications. Early identification is crucial to delay or avoid debilitating complications and achieve a better quality of life through the right medical interventions and precise hormonal readjustments. The proposed hybrid algorithm method finds the best features for finding thyroid disease uses performance measures such as accuracy, F1-score, precision, and recall. The research demonstrates promising results with an accuracy of 98.91 % and an F1-score of 94.83, showcasing the robustness of the proposed algorithms on a benchmark dataset. The findings hold potential to improve clinical decision-making processes. This study advances medical diagnostics by combining machine learning algorithms with nature-inspired optimization techniques to detect thyroid illnesses in their early stages. • This article proposes a novel hybrid algorithm that combines the Cuttlefish Optimization Algorithm (CFA) and Simulated Annealing (SA) to find the best features for finding thyroid disease. • The study uses machine-learning models for classification. • The integration of machine learning and nature-inspired optimization significantly enhances the diagnostic capabilities of healthcare systems, enabling prompt diagnosis and treatment planning for thyroid disorders.

Abstract licence: CC BY-NC

30 more references, whose wording the paper’s licence keeps private.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.