Skip to main content

Table 1 List of features that were used to building the machine learning algorithms

From: Development of machine learning models for detection of vision threatening Behçet’s disease (BD) using Egyptian College of Rheumatology (ECR)–BD cohort

Sex (male/female)

Age, years

Age at disease onset, years

Disease duration, years

Smoking status (never vs current/former)

Comorbidities (DM, HTN)

Oral ulcer (Y/N)

Genital ulcer (Y/N)

Mucocutaneous involvement (Y/N)

Musculoskeletal involvement (Y/N)

Neurological involvement (Y/N)

Vascular involvement (Y/N)

Gastrointestinal involvement (Y/N)

Disease activity (BDCAF)

Current treatment use (colchicine, MTX, AZA, CYC, CsA, chlorambucil, anticoagulants, biologics) (Y/N)

Current glucocorticoid dose, mg/day

C-reactive protein (CRP), mg/L

Erythrocyte sedimentation rate (ESR), mm/1st hour

Thrombocytosis (Y/N)

  1. Define variables: mucocutaneous involvement includes presence of any erythema, papulopustular, pseudo folliculitis or positive Pathergy test; musculoskeletal involvement includes presence of arthralgia or arthritis, neurological involvement includes stroke, transient ischemic attacks, convulsion, ataxia, cranial or peripheral neuropathy, or psychosis, vascular involvement includes presence of vasculitis, arterial or venous thrombosis, thrombophlebitis, or aneurysm, and gastrointestinal involvement includes presence of diarrhea, bloody diarrhea, or bleeding per rectum
  2. BDCAF, Behçet’s disease current activity form; MTX, methotrexate; AZA, azathioprine; CYC, cyclophosphamide; CsA, cyclosporine A
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us