Personalized treatment enhances healthcare by tailoring optimal decisions to each patient based on their specific characteristics and treatment history. Reinforcement learning (RL) methods are powerful tools for estimating optimal, data-driven, dynamic treatment decision rules. This article presents a tutorial on Tree-based RL and Multi-Objective Tree-based RL for advancing the estimation of optimal dynamic treatment regimes. Data from the Silicone Arthroplasty in Rheumatoid Arthritis study demonstrate their application in optimizing joint arthroplasty decisions. These methods support personalized, data-driven strategies while balancing competing clinical priorities, aiding clinicians in making informed, patient-centered decisions within ethical and practical constraints.