Self-training improves heart vessel images across different hospitals
Self-training for cross-centre coronary vessel segmentation in X-ray angiography using pre-contrast frame subtraction
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Coronary vessel images from X-ray machines differ between hospitals, making it hard for AI models to work well everywhere. The authors tested a technique called self-training that uses the model's own guesses alongside enhancing images by subtracting an early frame before contrast appears. This approach improved vessel detection on new hospital data without needing extra human help. However, the initial quality of the model’s guesses is key to success. Trying the process again didn’t further improve results.
What this means in practice
- •For medical imaging software teams: Improve coronary vessel segmentation models to work well on X-ray images from multiple hospitals without needing new annotations.
- •For radiology ai developers: Boost performance of vessel segmentation tools on external X-ray angiography datasets by combining model predictions with image subtraction techniques.
Authors
Danial Moafi, Aida Moafi, Mostafa Aryafar, Ashkan Sami, Maryam Aryafar
Abstract
Deep learning models for coronary vessel segmentation in X-ray angiography often perform poorly on data from centres that were not represented in training. Apart from differences in equipment, the labels of the largest public dataset (ARCADE) are incomplete, as only the segments defined in the SYNTAX score are annotated. In this study, we evaluated whether self-training with pseudo-labels can improve performance on external data without additional manual annotation. Pseudo-labels were generated from two sources, namely the predictions of an nnU-Net model trained on ARCADE and a contrast map obtained by subtracting a pre-contrast frame from the same sequence. The nnU-Net predictions were extended along the contrast map, and pixels detected only in the subtraction image were assigned an ignore label and excluded from the loss. Six models were trained and tested on ARCADE and on two external datasets (XCAD and XACV). All first-round models initialised with the baseline predictions performed better than both the baseline and a control model fine-tuned for the same number of epochs. The model trained on the XCAD pool improved the macro-averaged Dice score by 2.7 points (95% CI 2.3-3.2), while the control decreased it by 0.6 points. A second round of self-training did not lead to further improvement. When an external model trained on 50 fully annotated images was used to generate the initial labels, the XCAD Dice score increased from 0.623 to 0.745 without any XCAD labels, although performance on XACV did not improve because large main vessels were missed. These results suggest that the benefit of self-training depends largely on the sensitivity of the model used to generate the initial pseudo-labels.