Reinforcement Learning (RL) has emerged as a powerful AI paradigm for optimizing complex decision-making processes in healthcare. Unlike traditional machine learning methods, RL enables adaptive learning from real-time feedback, allowing healthcare systems to dynamically adjust treatment strategies, allocate resources efficiently, and improve clinical decision-making. The ability of RL to model sequential decision-making under uncertainty makes it particularly well-suited for personalized medicine, automated diagnostics, and intelligent healthcare interventions. This paper explores the role of RL in enhancing real-time decision-making for adaptive patient management and clinical workflow automation. By leveraging deep reinforcement learning (DRL) models, healthcare systems can optimize dynamic treatment regimes, including chemotherapy cycle planning, personalized insulin dosing, and sepsis management. These AI-driven treatment strategies enable precision medicine by continuously adapting to individual patient responses, thereby minimizing adverse effects and improving therapeutic outcomes. Furthermore, RL plays a critical role in resource optimization within hospitals, automating the allocation of intensive care unit (ICU) beds, ventilators, and surgical schedules based on predictive analytics. In robotic-assisted surgery, DRL enhances precision and adaptability, enabling autonomous control of surgical instruments, improving accuracy, and reducing surgical complications. Similarly, RL-driven rehabilitation therapies personalize physiotherapy sessions, optimizing recovery plans for stroke and spinal cord injury patients by dynamically adjusting therapy intensity based on real-time patient performance. Despite its transformative potential, challenges such as model interpretability, ethical considerations, and data efficiency must be addressed for RL to be effectively deployed in real-world clinical settings. This paper provides a comprehensive review of RL applications in healthcare, emphasizing advancements, challenges, and future prospects in AI-driven medical decision-making.