VALIDATION OF A MACHINE LEARNING DERIVED CLINICAL METRIC TO QUANTIFY OUTCOMES AFTER TSA
Christopher Roche, MSE, MBA, Vikas Kumar, PhD, Steven Overman, MD, MPH, Ryan Simovitch, MD, Pierre-Henri Flurin, MD, Thomas Wright, MD, Howard Routman, DO, Ankur Teredesai, PhD, Joseph Zuckerman, MD
WHAT IS THE ACCURACY OF THREE DIFFERENT MACHINE LEARNING TECHNIQUES TO PREDICT CLINICAL OUTCOMES AFTER SHOULDER ARTHROPLASTY?
Kumar V, Roche C, Overman S, Simovitch R, Flurin PH, Wright T, Zuckerman J, Routman H, Teredesai A. Clin Orthop Relat Res. 2020 Apr 20.
Request From Author Shoulder Paper on Machine Learning Techniques to Predict Clinical Outcomes
USING MACHINE LEARNING TO PREDICT CLINICAL OUTCOMES AFTER SHOULDER ARTHROPLASTY WITH A MINIMAL FEATURE SET
Kumar V, Roche C, Overman S, Simovitch R, Flurin PH, Wright T, Zuckerman J, Routman H, Teredesai A. J Shoulder Elbow Surg. In Press. 2020.
Quantifying Success after Total Shoulder Arthroplasty
USE OF MACHINE LEARNING TO ASSESS THE PREDICTIVE VALUE OF 3 COMMONLY USED CLINICAL MEASURES TO QUANTIFY OUTCOMES AFTER TOTAL SHOULDER ARTHROPLASTY
Vikas Kumar, PhD, Christopher Roche, MSE, MBA,*, Steven Overman, MD, MPH, Ryan Simovitch, MD, Pierre-Henri Flurin, MD, Thomas Wright, MD, Joseph Zuckerman, MD, Howard Routman, DO, and Ankur Teredesai, PhD
COMPARISON OF COMPLICATION TYPES AND RATES ASSOCIATED WITH ANATOMIC AND REVERSE TOTAL SHOULDER ARTHROPLASTY.
Parada SA, Flurin PH, Wright TW, Zuckerman JD, Elwell JA, Roche CP, Friedman RJ. J Shoulder Elbow Surg. In Press. 2020.
QUANTIFYING SUCCESS AFTER TOTAL SHOULDER ARTHROPLASTY: THE MINIMAL CLINICALLY IMPORTANT DIFFERENCE
Simovitch R, Flurin PH, Wright T, Zuckerman JD, Roche CP. J Shoulder Elbow Surg. 2018 Feb;27(2):298-305.
QUANTIFYING SUCCESS AFTER TOTAL SHOULDER ARTHROPLASTY: THE SUBSTANTIAL CLINICAL BENEFIT
Simovitch R, Flurin PH, Wright T, Zuckerman JD, Roche CP. J Shoulder Elbow Surg. 2018 May;27(5):903-911.
Predict+ is designed for general educational purposes only and is not intended in any way to substitute for professional medical advice, consultation, diagnosis, or treatment. Any information contained in or produced by this tool is intended to serve as a supplement to and not a substitute for the knowledge, expertise, skill and judgment of healthcare professionals.
Predict+ utilizes machine learning algorithms that learned patterns from aggregate clinical outcomes data collected from >30 different sites. This clinical data inevitably contains bias inherited from the unique circumstances of surgeons, hospitals, patients, and data collection procedures. As a result, model predictions in some cases may not be representative of the outcomes achieved by patients of different demographics, regions, or ethnicity/race than comprise the clinical outcomes database from which the algorithms are derived, and model predictions may be biased against patients too sick to safely undergo the procedure or patients whose condition is not sufficiently degenerative to have the procedure. Each patient’s needs are unique and different, and patient-specific requirements for pain relief and functional improvement may not align with established thresholds for improvement. As predictions indicate a range of anticipated outcomes, Predict+ should be used to inform treatment decision-making and shall never be misused to deny treatment.
Predict+ was developed from a dataset of primary anatomic and reverse total shoulder arthroplasty patients using the Exactech Equinoxe platform shoulder prosthesis where patients with revisions, humeral fractures, or hemiarthroplasty were excluded; therefore, model predictions may not be appropriate for those excluded indications or other prosthesis types or designs. Predict+ is intended for labeled indications for use of the Equinoxe platform total shoulder system. Please consult the instructions for use accompanying the Equinoxe implants.
Predict+ does not process or acquire medical images or signals. The locked predictive algorithms utilized in Predict+ and the factors utilized to develop methods are peer-reviewed and available online as linked above. The factors, also referred to as input parameters, that are found to be significant to the predictions are communicated within Predict+ in order to enable the healthcare professional to independently review the basis of the recommendation. In no event shall Predict+ be considered medical care, treatment, or therapy for patients or users of this tool. Predict+ and its services are provided ‘as is’. These services provide no warranties, express or implied and Predict+ shall not be liable for any direct, consequential, lost profits, or other damages incurred by the user of this information tool.