Ina Hanninger1, Eylem Levelt2,3, Jennifer J Rayner2, Christopher T Rodgers2,4, Stefan Neubauer2, Vicente Grau1, Oliver J Rider2, and Ladislav Valkovic2,5
1Oxford Institute of Biomedical Engineering, Oxford, United Kingdom, 2Radcliffe Department of Medicine, University of Oxford Centre for Clinical Magnetic Resonance Research, Oxford, United Kingdom, 3University of Leeds, Leeds, United Kingdom, 4Wolfson Brain Imaging Centre, Cambridge, United Kingdom, 5Slovak Academy of Sciences, Institute of Measurement Science, Bratislava, Slovakia
1Oxford Institute of Biomedical Engineering, Oxford, United Kingdom, 2Radcliffe Department of Medicine, University of Oxford Centre for Clinical Magnetic Resonance Research, Oxford, United Kingdom, 3University of Leeds, Leeds, United Kingdom, 4Wolfson Brain Imaging Centre, Cambridge, United Kingdom, 5Slovak Academy of Sciences, Institute of Measurement Science, Bratislava, Slovakia
Random Forest classifiers and Bayesian networks applied to MR spectroscopy measures suggests a high predictive impact of cardiac metabolism in classifying diabetic and obese patients, further implying a causal mechanism with visceral fat, concentric LV remodeling and energy impairment
Figure 2(a,b,c): Bayesian networks learned through NOTEARS structure learning algorithm for each subgroup pair of the data. Each node represents a feature variable, and each directed edge encodes causal influence in the form of conditional probability dependence.
Figure 1(a,b,c): SHAP value plots computed for each Random Forest classification, representing a rank of feature importances. The x-axis gives the SHAP value, i.e. the impact on the model output (for a positive classification). Red indicates higher feature values, while blue indicates lower values.
