Harshith.
Back to work

Run 05ML for Medical Applications · Johns Hopkins · Spring 2025

Pneumothorax Detection & Segmentation

Multi-task U-Net · SIIM-ACR

IoU
0.65
accuracy
80%

PyTorch · Albumentations · SIIM-ACR

Problem

Pneumothorax is a sparse, class-imbalanced segmentation problem on chest X-rays. A network that reports a global accuracy number without looking at collapsed lung boundaries is not a clinical model.

Approach

I trained a multi-task U-Net with residual blocks on the SIIM-ACR pneumothorax dataset.

  • Loss: Dice + BCE with positive-class up-weighting, because empty masks dominate the set.
  • Augmentation: elastic deform and random affine, tuned for chest X-ray morphology rather than natural-image defaults.
  • Metrics: Dice and IoU, plus a failure-mode pass across subgroups so confidence is something I measured, not assumed.

Result: IoU 0.65 and 80% accuracy on the held evaluation. The useful artifact is the imbalance handling and the subgroup analysis, not the headline number.