The use of technology to assess balance and alignment during total knee surgery can provide an overload of numerical data to the surgeon. Meanwhile, this quantification holds the potential to clarify and guide the surgeon through the surgical decision process when selecting the appropriate bone recut or soft tissue adjustment when balancing a total knee. Therefore, this paper evaluates the potential of deploying supervised machine learning (ML) models to select a surgical correction based on patient-specific intra-operative assessments. Based on a clinical series of 479 primary total knees and 1,305 associated surgical decisions, various ML models were developed. These models identified the indicated surgical decision based on available, intra-operative alignment, and tibiofemoral load data.Aims
Methods
The aim of this study was to investigate the effect of laboratory-based simulator training on the ability of surgical trainees to perform diagnostic arthroscopy of the knee. A total of 20 junior orthopaedic trainees were randomised to receive either a fixed protocol of arthroscopic simulator training on a bench-top knee simulator or no additional training. Motion analysis was used to assess performance objectively. Each trainee then received traditional instruction and demonstrations of diagnostic arthroscopy of the knee in theatre before performing the procedure under the supervision of a blinded consultant trainer. Their performance was assessed using a procedure-based assessment from the Orthopaedic Competence Assessment Project and a five-point global rating assessment scale. In theatre the simulator-trained group performed significantly better than the untrained group using the Orthopaedic Competence Assessment Project score (p = 0.0007) and assessment by the global rating scale (p = 0.0011), demonstrating the transfer of psychomotor skills from simulator training to arthroscopy in the operating theatre. This has implications for the planning of future training curricula.
