Skip to main content

Table 1 Input variables for ML models

From: Development and internal validation of machine learning algorithms to predict patient satisfaction after total hip arthroplasty

Demographics

Comorbidities

Preoperative PROMs

Age*

Sex

BMI (numerical)*

BMI (categorical)*

Number of comorbidities*

Diabetes

Hypertension

High cholesterol

Ischemic heart disease,

Stroke

Renal disease

Back pain

Depression

Previous hip surgery*

Previous knee surgery

Previous lumbar spine surgery*

SF-36 PCS*

SF-36 MCS*

WOMAC*

OHS*

  1. *10 candidate variables selected from recursive feature elimination