Thyroid Cancer Prediction Using Optimizations
Swati Sharma, Vijay Kumar Sharma, Punit Mittal, Pradeep Pant, Nitin Rakesh · 2025
In the past four decades, we have seen a gradual upsurge in the number of thyroid cancer cases. This alarming diagnosis rate can be implicated in the progressiveness we have achieved in medical imaging techniques augmented with computer-assisted technologies. Given their superior capacity to uncover complex correlations from biological data, machine learning methods are being rapidly included in computer-aided design (CAD) systems. In this paper, we demonstrate how current, non-specialized medical records may be consistently converted into predictive power to help doctors make well-informed recommendations for treatment. A 96.8% accuracy rate in prognostic patient differentiation has been attained. It was achieved by employing data from sizable cohorts of thyroid cancer patients to optimize supervised neural networks, most especially multilayer perceptions appropriately. We also see the possibility of adapting our machine-learning method to other illnesses and objectives related to the malignant nature of organs at the microscopic level.