X-Ray Bone Age Prediction

Bone age is a predictor of the skeletal maturity of a child. Early detection of abnormal bone age allows early intervention. It is usually assessed by pediatricians using an X-ray of the fingers and wrist with an atlas of X-rays to determine bone age. The goal of this algorithm is to determine the bone age from X-ray images accurately.

Inspired by the clinical workflow (Tanner-Whitehouse) of bone age assessment, this algorithm is based on a unified network for robust bone age assessment. We used two subnets:

  1. A segmentation sub-net to segment hands from the image to remove the distractions from backgrounds;
  2. A residual attention sub-net to generate the final prediction and visual evidence. Promising results were published.
  • Accurately bone age prediction
  • Highlights the suspected image region

Residual Attention Based Network for Hand Bone Age Assessment
Wu E, Kong B, Wang X, Bai J, Lu Y, et al.,
IEEE International Symposium on Biomedical Imaging (ISBI), 2019.