Browsing by Keyword "explainability"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Explainable COVID-19 Infections Identification and Delineation Using Calibrated Pseudo Labels(2023-02-01) Li, Ming; Fang, Yingying; Tang, Zeyu; Onuorah, Chibudom; Xia, Jun; Ser, Javier Del; Walsh, Simon; Yang, Guang; IAThe upheaval brought by the arrival of the COVID-19 pandemic has continued to bring fresh challenges over the past two years. During this COVID-19 pandemic, there has been a need for rapid identification of infected patients and specific delineation of infection areas in computed tomography (CT) images. Although deep supervised learning methods have been established quickly, the scarcity of both image-level and pixel-level labels as well as the lack of explainable transparency still hinder the applicability of AI. Can we identify infected patients and delineate the infections with extreme minimal supervision? Semi-supervised learning has demonstrated promising performance under limited labelled data and sufficient unlabelled data. Inspired by semi-supervised learning, we propose a model-agnostic calibrated pseudo-labelling strategy and apply it under a consistency regularization framework to generate explainable identification and delineation results. We demonstrate the effectiveness of our model with the combination of limited labelled data and sufficient unlabelled data or weakly-labelled data. Extensive experiments have shown that our model can efficiently utilize limited labelled data and provide explainable classification and segmentation results for decision-making in clinical routine.Item Prediction of Metabolic Syndrome Based on Machine Learning Techniques with Emphasis on Feature Relevances and Explainability Analysis(Institute of Electrical and Electronics Engineers Inc., 2023) Ispizua, Begoña; Manjarrés, Diana; Niño-Adan, Iratxe; Jiang, Xingpeng; Wang, Haiying; Alhajj, Reda; Hu, Xiaohua; Engel, Felix; Mahmud, Mufti; Pisanti, Nadia; Cui, Xuefeng; Song, Hong; IAMetabolic syndrome (MetS) is considered to be a major public health problem worldwide leading to a high risk of diabetes and cardiovascular diseases. In this paper, data collected by the Precision Medicine Initiative of the Basque Country, named the AKRIBEA project, is employed to infer via Machine Learning (ML) techniques the features that have the most influence on predicting MetS in the general case and also separately by gender. Different Feature Normalization (FN) and Feature Weighting (FW) methods are applied and an exhaustive analysis of explainability by means of Shapley Additive Explanations (SHAP) and feature relevance methods is performed. Validation results show that the Extreme Gradient Boosting (XGB) with Min-Max FN and Mutual Information FW achieves the best trade-off between precision and recall performance metrics.