Artificial intelligence [AI] has become an indispensable part of scientific and medical research, however, the exponential growth in model complexity and computational requirements has made AI adoption challenging for resource-limited settings. The concept of Frugal AI has emerged to address this gap by designing efficient, accessible, and sustainable AI systems that achieve comparable accuracy with reduced computational, financial, and environmental costs. In the context of medical research, Frugal AI enables data-driven discovery, diagnostics, and decision-making through low-cost, scalable, and adaptable methodologies. This review explores the evolution and influence of Frugal AI on research methodology in biomedical sciences over the past decade. It discusses its principles, technological innovations, and integration into experimental design, data management, and analytical frameworks. Furthermore, the paper critically evaluates the benefits and limitations of Frugal AI in medical research, highlighting its transformative role in improving accessibility, reducing bias, and fostering innovation in low-resource environments. Finally, it underscores the future perspectives of Frugal AI in achieving equitable, reproducible, and sustainable medical research worldwide.