Consolidation of safe and quality software applications in medicine is crucial for the safety of patients, diagnosis, and better healthcare delivery. Testing medical software is more challenging due to strict regulatory requirements, high complexity, and criticality in application. The success of artificial intelligence (AI) tools as enablers in improving the efficiency, accuracy, and comprehensiveness of medical software testing speaks volumes. This literature review explores the current state of AI-based methodologies deployed in the QA domain of medical software. Key methods like learning-based test generation, automated defect detection, anomaly identification, and intelligent risk assessment have been discussed. The review establishes the potential AI tool has in alleviating some issues, for example, reduction of manual intervention, improvement of test coverage, and prediction of software failures ahead of deployment. Despite the promise, several challenges pose barriers to the adequate adoption of AI tools within this domain. These include the need for explainable AI models for transparency, tackling domain-specific data constraints, and robust validation frameworks for safety-critical systems. Their adoption requires addressing ethical, technical, and operational concerns in order to maximize benefits to software quality. The existing review concludes that AI tools can play a pivotal role in enhancing medical software testing if existing limitations are adequately addressed through interdisciplinary collaboration between AI experts and medical professionals. At best, this collaboration can make formal testing practices in the healthcare sector safer, more cost-effective, and simultaneously improve patient outcomes.