UNet Deep Learning Architecture for Segmentation of Vascular and Non-Vascular Images: A Microscopic Look at UNet Components Buffered With Pruning, Explainable Artificial Intelligence, and Bias

Biomedical image segmentation (BIS) task is challenging due to the variations in organ types, position, shape, size, scale, orientation, and image contrast. Conventional methods lack accurate and automated designs. Artificial intelligence (AI)-based UNet has recently dominated BIS. This is the first review of its kind that microscopically addressed UNet types by complexity, stratification of UNet by its components, addressing UNet in vascular vs. non-vascular framework, the key to segmentation challenge vs. UNet-based architecture, and finally interfacing the three facets of AI, the pruning, the explainable AI (XAI), and the AI-bias. PRISMA was used to select 267 UNet-based studies. Five classes were identified and labeled as conventional UNet, superior UNet, attention-channel UNet, hybrid UNet, and ensemble UNet. We discovered 81 variations of UNet by considering six kinds of components, namely encoder, decoder, skip connection, bridge network, loss function, and their combination. Vascular vs. non-vascular UNet architecture was compared. AP(ai)Bias 2.0-UNet was identified in these UNet classes based on (i) attributes of UNet architecture and its performance, (ii) explainable AI (XAI), and, (iii) pruning (compression). Five bias methods such as (i) ranking, (ii) radial, (iii) regional area, (iv) PROBAST, and (v) ROBINS-I were applied and compared using a Venn diagram. Vascular and non-vascular UNet systems dominated with sUNet classes with attention. Most of the studies suffered from a low interest in XAI and pruning strategies. None of the UNet models qualified to be bias-free. There is a need to move from paper-to-practice paradigms for clinical evaluation and settings.

Differentiation of COVID-19 Conditions using Mediastinum Shape in Chest X-ray Images

In this work, an attempt has been made to analyze the shape variations in mediastinum for differentiation of Coronavirus Disease-2019 (COVID-19) and normal conditions in chest X-ray images. For this, the images are obtained from a publicly available dataset. Segmentation of mediastinum from the raw images is performed using Reaction Diffusion Level Set (RDLS) method. Shape-based features are extracted from the delineated mediastinum masks and are statistically analyzed. Further, the features are fed to two classifiers, namely, multi-layer perceptron and support vector machine for differentiation of normal and COVID-19 images. From the results, it is observed that the employed RDLS method is able to delineate mediastinum from the raw chest X-ray images. Eight shape features are observed to be statistically significant. The mean values of these features are found to be distinctly higher for COVID-19 images as compared to normal images. Area under the curve of greater than 76.9% is achieved for both the classifiers. It appears that mediastinum could be used as a region of interest for computerized detection and mass screening of the disease. © 2022 The Author(s), published by De Gruyter.