ANN uncertainty estimates in assessing fatty liver content from ultrasound data

• Published in: Computational and structural biotechnology journal Vol. 24; pp. 603 - 610
• Main Authors: Del Corso, G., Pascali, M.A., Caudai, C., De Rosa, L., Salvati, A., Mancini, M., Ghiadoni, L., Bonino, F., Brunetto, M.R., Colantonio, S., Faita, F.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.12.2024; Research Network of Computational and Structural Biotechnology; Elsevier
• Subjects: Artificial neural networks; Bayesian uncertainty; Fatty liver content; Ultrasound imaging; Uncertainty quantification; Uncertainty quantification; Ultrasound imaging; Fatty liver content; Artificial neural networks; Bayesian uncertainty
• ISSN: 2001-0370; 2001-0370
• DOI: 10.1016/j.csbj.2024.09.021

This article uses three different probabilistic convolutional architectures applied to ultrasound image analysis for grading Fatty Liver Content (FLC) in Metabolic Dysfunction Associated Steatotic Liver Disease (MASLD) patients. Steatosis is a new silent epidemic and its accurate measurement is an impelling clinical need, not only for hepatologists, but also for experts in metabolic and cardiovascular diseases. This paper aims to provide a robust comparison between different uncertainty quantification strategies to identify advantages and drawbacks in a real clinical setting. We used a classical Convolutional Neural Network, a Monte Carlo Dropout, and a Bayesian Convolutional Neural Network with the goal of not only comparing the goodness of the predictions, but also to have access to an evaluation of the uncertainty associated with the outputs. We found that even if the prediction based on a single ultrasound view is reliable (relative RMSE [5.93%-12.04%]), networks based on two ultrasound views outperform them (relative RMSE [5.35%-5.87%]). In addition, the results show that the introduction of a “not confident” category contributes to increase the percentage of correctly predicted cases and to decrease the percentage of mispredicted cases, especially for semi-intrusive methods. The possibility of having access to information about the confidence with which the network produces its outputs is a great advantage, both from the point of view of physicians who want to use neural networks as computer-aided diagnosis, and for developers who want to limit overfitting and obtain information about dataset problems in terms of out-of-distribution detection.