In clinical practice, algorithmic predictions may seriously jeopardise patients’ health and thus are required to be validated by medical experts before a final clinical decision is met. Towards that aim, there is need to incorporate explainable artificial intelligence techniques into medical research. In the specific field of epileptic seizure detection there are several machine learning algorithms but less methods on explaining them in an interpretable way. Therefore, we introduce XAI4EEG: an application-aware approach for an explainable and hybrid deep learning-based detection of seizures in multivariate EEG time series. In XAI4EEG, we combine deep learning models and domain knowledge on seizure detection, namely (a) frequency bands, (b) location of EEG leads and (c) temporal characteristics. XAI4EEG encompasses EEG data preparation, two deep learning models and our proposed explanation module visualizing feature contributions that are obtained by two SHAP explainers, each explaining the predictions of one of the two models. The resulting visual explanations provide an intuitive identification of decision-relevant regions in the spectral, spatial and temporal EEG dimensions. To evaluate XAI4EEG, we conducted a user study, where users were asked to assess the outputs of XAI4EEG, while working under time constraints, in order to emulate the fact that clinical diagnosis is done - more often than not - under time pressure. We found that the visualizations of our explanation module (1) lead to a substantially lower time for validating the predictions and (2) leverage an increase in interpretability, trust and confidence compared to selected SHAP feature contribution plots.

Background: The mobile phone app, TrackYourStress (TYS), is a new crowdsensing mobile health platform for ecological momentary assessments of perceived stress levels. Objective: In this pilot study, we aimed to investigate the time trend of stress levels while using TYS for the entire population being studied and whether the individuals’ perceived stress reactivity moderates stress level changes while using TYS. Methods: Using TYS, stress levels were measured repeatedly with the 4-item version of the Perceived Stress Scale (PSS-4), and perceived stress reactivity was measured once with the Perceived Stress Reactivity Scale (PSRS). A total of 78 nonclinical participants, who provided 1 PSRS assessment and at least 4 repeated PSS-4 measurements, were included in this pilot study. Linear multilevel models were used to analyze the time trend of stress levels and interactions with perceived stress reactivity. Results: Across the whole sample, stress levels did not change while using TYS (P=.83). Except for one subscale of the PSRS, interindividual differences in perceived stress reactivity did not influence the trajectories of stress levels. However, participants with higher scores on the PSRS subscale reactivity to failure showed a stronger increase of stress levels while using TYS than participants with lower scores (P=.04). Conclusions: TYS tracks the stress levels in daily life, and most of the results showed that stress levels do not change while using TYS. Controlled trials are necessary to evaluate whether it is specifically TYS or any other influence that worsens the stress levels of participants with higher reactivity to failure.