Automatic assessment of developmental dysplasia of the hip

Abstract

Developmental dysplasia of the hip (DDH) refers to a spectrum of hip joint abnormalities that may lead to significant adverse consequences if not detected and treated early in infancy. Clinical exam with selective ultrasound (US) imaging is the current standard used for early detection and diagnosis of DDH in infants up to 6 months of age. However, current US systems require specialized training and have considerable between-center variability, which raises concerns of missed early diagnoses and subsequent complications from late treatment. We propose a novel, automatic, and near real-time image analysis approach that extracts bone contours in US images of the neonatal hip and calculates image based dysplasia metrics (alpha and beta angles). We present quantitative validation on 54 US scans of femoral head flexion collected from 10 infants under age of 3 months.