Deep learning (DL) and machine learning (ML) techniques are used in AI-based "normal" vs. "abnormal" data classification, especially in healthcare and industrial applications, to differentiate between standard, healthy, or routine states and deviations that indicate illness, injury, or malfunctioning operations. This technique is commonly utilized to increase productivity, decrease human error, and speed up decision-making in signal processing, behavior monitoring, and imaging. One of the most recent developments in medical imaging is thermal imaging of breast cancers. Conventional techniques make it more difficult for doctors to diagnose patients quickly and accurately since they take longer to evaluate every image. Research on computer-aided techniques for rapidly and precisely diagnosing all produced pictures has been prompted by this. It is time-consuming and prone to inaccuracy to manually detect and categorize breast cancers. The system makes use of artificial intelligence algorithms to compute certain metrics for a given region or to identify the best configuration of picture attributes for categorization. This study focused on the use of machine learning in the diagnosis and classification of breast tumors using thermal imaging. The results of the current research indicate that the algorithms used in the proposed classification system perform well based on image quality metrics. The accuracy of classifying benign from malignant breast tumors reached 90.25% and 86.65%, using DNN and RF algorithms respectively.