<p>The growing reliance on video-based imaging for diagnosis and surgical guidance necessitates the development of robust Video Quality Assessment (VQA) methods in the medical domain to ensure accurate evaluation of visual information. Such methods are essential for assessing the applicability, limitations, and potential of imaging systems to enhance perceptual quality. In this context, subjective VQA depend on human observers to rate the perceived visual quality of medical videos, while objective VQA employ mathematical models for automated quality prediction. This paper provides a systematic review of available datasets, subjective evaluation protocols and objective assessment techniques for medical video quality analysis. Relevant studies were collected from IEEE, Springer, ScienceDirect, SPIE and PLOS digital repositories, covering the period from 2001 to 2025. The surveyed literature highlights the need for more publicly available medical video datasets for supporting research advancement in this field. It also indicates that subjective assessments typically use four reference videos with a temporal resolution of 25 fps and an average duration of 10 seconds. Expert participation, the Double Stimulus Continuous Quality Scale (DSCQS) and Mean Opinion Score (MOS) ratings are found to be the most preferred choices for subjective evaluation. Various statistical approaches, along with machine learning and deep learning models, have been employed to estimate objective video quality scores, with performance commonly validated using Pearson’s Linear Correlation Coefficient (PLCC) and Spearman’s Rank Order Correlation Coefficient (SROCC). The key findings from this review consolidates recurring methodological trends and design preferences, providing a structured foundation for future research in perceptual quality assessment of medical videos.</p>