A machine learning and nomogram-based study: effect of applying biologically formulated platelet-rich plasma (PRP) on the degree of pain relief after rotator cuff injury and prediction modeling: integrating biomedicine and artificial intelligence

Jianguo Zhang^1*Jian Gao²Hao-Yu Feng^3*Wei Liu^3*

• ¹Department of Orthopedics, Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Tongji Shanxi Hospital, Third Hospital of Shanxi Medical University, Taiyuan, China, Taiyuan, China
• ²Shanxi Bethune Hospital（ Shanxi Academy of Medical Sciences） Third Hospital of Shanxi Medical University； Tongji Shanxi Hospital，, taiyuan, China
• ³Shanxi Bethune Hospital（ Shanxi Academy of Medical Sciences） Third Hospital of Shanxi Medical University； Tongji Shanxi Hospital, taiyuan, China

Background A popular orthopedic operation that frequently results in severe postoperative discomfort, rotator cuff repair can impair functional recovery and hamper rehabilitation. An increasingly popular biologically active autologous treatment for improving tendon repair and lowering postoperative inflammation is platelet-rich plasma (PRP). Clinical reactions to PRP are still uneven, though, and individual results for pain reduction differ greatly. According to new research, a patient's age, body mass index (BMI), and preexisting inflammatory condition may all affect how well PRP works as an analgesic. Such multifactorial interactions are frequently not adequately captured by conventional linear models. As a result, the demand for sophisticated, interpretable technologies to forecast unique pain alleviation trajectories is rising. While nomograms give an easy way to visualize individualized risk, machine learning offers improved power in modeling complicated, nonlinear interactions. In addition to developing a machine learning-based nomogram model for personalized prediction by including important clinical and demographic factors, the study intends to assess the impact of PRP on postoperative pain reduction after rotator cuff repair.