In this paper, we present the potential of Explainable Artificial Intelligence methods for decision support in medical image analysis scenarios. Using three types of explainable methods applied to the same medical image data set, we aimed to improve the comprehensibility of the decisions provided by the Convolutional Neural Network (CNN). In vivo gastral images obtained by a video capsule endoscopy (VCE) were the subject of visual explanations, with the goal of increasing health professionals’ trust in black-box predictions. We implemented two post hoc interpretable machine learning methods, called Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP), and an alternative explanation approach, the Contextual Importance and Utility (CIU) method. The produced explanations were assessed by human evaluation. We conducted three user studies based on explanations provided by LIME, SHAP and CIU. Users from different non-medical backgrounds carried out a series of tests in a web-based survey setting and stated their experience and understanding of the given explanations. Three user groups (n = 20, 20, 20) with three distinct forms of explanations were quantitatively analyzed. We found that, as hypothesized, the CIU-explainable method performed better than both LIME and SHAP methods in terms of improving support for human decision-making and being more transparent and thus understandable to users. Additionally, CIU outperformed LIME and SHAP by generating explanations more rapidly. Our findings suggest that there are notable differences in human decision-making between various explanation support settings. In line with that, we present three potential explainable methods that, with future improvements in implementation, can be generalized to different medical data sets and can provide effective decision support to medical experts.

Whole-brain computational modelling of emergent brain dynamics has become an increasingly important tool for understanding the systems-level mechanisms that govern large-scale brain activity in health and disease. Neuronal synchronization and brain criticality are key candidates of such mechanisms both for healthy cognitive functions and in constituting mechanistic biomarkers for several brain disorders. Despite significant advances, there is not an abundance of modeling approaches that yield both in-vivo-like critical synchronization dynamics and observables that directly match those obtainable with multi-modal neuroimaging. Here, we advance a framework for hierarchical Kuramoto models where the simplest two-layer model comprises a network of pairwise coupled nodes which each contain a large number of oscillators. Already at two levels of hierarchy, this enables explicit representation of local and inter-areal coupling and observations of emergent multi-scale synchronization dynamics. We show here that the model produces observables of meso- and macro-scale synchronization dynamics that are physiologically plausible when compared to, e.g., magneto- and electroencephalographic (M/EEG) data. We also present here a novel approach for fitting this model with individual experimental data that uses not only functional connectivity but also critical dynamics to converge model parameters that are personalized with individual M/EEG and structural-connectivity data. We posit that brain-dynamics matched personalized models yield a basis for ‘digital twins’ that may support both clinical applications and basic research on the systems-level mechanisms of cognitive functions. The model and model fitting approach advanced here support both new mechanistic understanding of experimental observations and derivation of predictions for future research.

Machine learning-based systems are rapidly gaining popularity and in-line with that there has been a huge research surge in the field of explainability to ensure that machine learning models are reliable, fair, and can be held liable for their decision-making process. Explainable Artificial Intelligence (XAI) methods are typically deployed to debug black-box machine learning models but in comparison to tabular, text, and image data, explainability in time series is still relatively unexplored. The aim of this study was to achieve and evaluate model agnostic explainability in a time series forecasting problem. This work focused on proving a solution for a digital consultancy company aiming to find a data-driven approach in order to understand the effect of their sales related activities on the sales deals closed. The solution involved framing the problem as a time series forecasting problem to predict the sales deals and the explainability was achieved using two novel model agnostic explainability techniques, Local explainable model-agnostic explanations (LIME) and Shapley additive explanations (SHAP) which were evaluated using human evaluation of explainability. The results clearly indicate that the explanations produced by LIME and SHAP greatly helped lay humans in understanding the predictions made by the machine learning model. The presented work can easily be extended to any time