Chest X-ray lung and heart segmentation based on minimal training sets
This addresses the problem of reducing doctor workload and improving diagnostic pipelines in medical imaging, particularly for conditions like cardiomegaly, but is incremental as it builds on existing deep learning methods with specific enhancements.
The paper tackles lung and heart segmentation from chest X-ray images to aid medical diagnosis, achieving state-of-the-art results with 98.1% Dice score and 95.2% IoU score on a standard dataset, and demonstrates strong performance with minimal training sets of size 10 and 20, reaching up to 97.3% Dice score.
As the COVID-19 pandemic aggravated the excessive workload of doctors globally, the demand for computer aided methods in medical imaging analysis increased even further. Such tools can result in more robust diagnostic pipelines which are less prone to human errors. In our paper, we present a deep neural network to which we refer to as Attention BCDU-Net, and apply it to the task of lung and heart segmentation from chest X-ray (CXR) images, a basic but ardous step in the diagnostic pipeline, for instance for the detection of cardiomegaly. We show that the fine-tuned model exceeds previous state-of-the-art results, reaching $98.1\pm 0.1\%$ Dice score and $95.2\pm 0.1\%$ IoU score on the dataset of Japanese Society of Radiological Technology (JSRT). Besides that, we demonstrate the relative simplicity of the task by attaining surprisingly strong results with training sets of size 10 and 20: in terms of Dice score, $97.0\pm 0.8\%$ and $97.3\pm 0.5$, respectively, while in terms of IoU score, $92.2\pm 1.2\%$ and $93.3\pm 0.4\%$, respectively. To achieve these scores, we capitalize on the mixup augmentation technique, which yields a remarkable gain above $4\%$ IoU score in the size 10 setup.