Artificial intelligence in disease prediction and diagnosis: a systematic literature review

Abstract

Artificial intelligence (AI) has emerged as a transformative force in healthcare, particularly in disease prediction and diagnosis, offering unprecedented opportunities to improve clinical outcomes and operational efficiency. This systematic literature review examines the current state of AI applications across various medical domains, focusing on its role in general disease diagnosis, cardiovascular diseases, cancer, and other specific conditions such as diabetes and Alzheimer's. We aim to synthesize existing research, identify key trends, and highlight gaps in the literature to guide future investigations. A rigorous methodology was employed to select and analyze relevant studies, ensuring a comprehensive evaluation of AI techniques, their performance, and clinical applicability. The findings reveal that AI models, particularly those based on deep learning and machine learning, demonstrate high accuracy in diagnosing diseases, often surpassing traditional methods. However, challenges such as data heterogeneity, interpretability, and integration into clinical workflows remain significant barriers. In cardiovascular diseases, AI excels in risk stratification and early detection, while in oncology; it enhances tumor classification and prognosis prediction. For chronic and neurodegenerative conditions, AI shows promise in personalized treatment planning. The review concludes that while AI holds immense potential, its widespread adoption requires addressing ethical, regulatory, and technical hurdles. Future research should prioritize robust validation, interdisciplinary collaboration, and real-world implementation to fully realize AI's benefits in healthcare.