Detecting and Classifying Tumors with Artificial Intelligence

Received: 9 September 2025; Revised: 5 November 2025; Accepted: 12 November 2025; Published: 19 November 2025

Abstract:

Tumor detection and classification play a crucial role in the early diagnosis and treatment planning of cancer, significantly improving patient outcomes. The rapid growth of medical imaging data has necessitated the development of intelligent and automated solutions for accurate interpretation. In this study, a novel artificial intelligence‑based method is proposed for detecting and classifying tumors using advanced machine learning ar‑ chitectures. The approach integrates convolutional and fully connected neural layers, structured to analyze key diagnostic features from imaging and clinical datasets. The model is fine‑tuned through evolutionary optimization to enhance its learning parameters and minimize classification errors. The system operates in two primary phases: data‑driven feature refinement and multi‑class tumor classification. Evaluations were carried out across multiple benchmark datasets, including Computer Tomography (CT) and Magnetic Resonance Imaging (MRI) scans, with the proposed method demonstrating a high classification accuracy and robustness in distinguishing between benign and malignant tumor types. Comparative analyses show that the proposed model achieves superior performance over conventional classifiers and hybrid systems. This framework not only supports accurate and early tumor di‑ agnosis but also holds potential for adaptation across various tumor categories beyond those initially tested. The study underscores the capacity of artificial intelligence to support clinical decision‑making in oncology through reliable, efficient, and scalable diagnostic tools.