Imagine a team of detectives trying to solve a mysterious case using different types of evidence – fingerprints, footprints, and DNA samples. In the world of electrophysiology studies, researchers face a similar challenge when classifying data using multiple modalities. However, the lack of explainability in current classification methods hinders their progress. That’s why this study introduces novel methods to shed light on the importance of different modalities in multimodal electrophysiology classifiers.

To tackle this issue, the researchers trained a convolutional neural network with electroencephalogram (EEG), electrooculogram, and electromyogram data to automatically classify sleep stages. They then developed global and local explainability approaches specific to electrophysiology analysis. The global approach revealed that EEG is generally the most important modality for most sleep stages. However, the local explanations uncovered subject-level differences not captured by global methods.

Moreover, the study explored how clinical and demographic variables affect the classifier’s patterns. Interestingly, sex was found to have a significant impact, followed by medication and age. These findings provide valuable insights into the intersection of biology and machine learning.

By improving explainability in multimodal electrophysiology classification, these novel methods can advance personalized medicine and pave the way for implementing clinical classifiers in this field. Curious about the details? Dive into the research!

IntroductionMultimodal classification is increasingly common in electrophysiology studies. Many studies use deep learning classifiers with raw time-series data, which makes explainability difficult, and has resulted in relatively few studies applying explainability methods. This is concerning because explainability is vital to the development and implementation of clinical classifiers. As such, new multimodal explainability methods are needed.MethodsIn this study, we train a convolutional neural network for automated sleep stage classification with electroencephalogram (EEG), electrooculogram, and electromyogram data. We then present a global explainability approach that is uniquely adapted for electrophysiology analysis and compare it to an existing approach. We present the first two local multimodal explainability approaches. We look for subject-level differences in the local explanations that are obscured by global methods and look for relationships between the explanations and clinical and demographic variables in a novel analysis.ResultsWe find a high level of agreement between methods. We find that EEG is globally the most important modality for most sleep stages and that subject-level differences in importance arise in local explanations that are not captured in global explanations. We further show that sex, followed by medication and age, had significant effects upon the patterns learned by the classifier.DiscussionOur novel methods enhance explainability for the growing field of multimodal electrophysiology classification, provide avenues for the advancement of personalized medicine, yield unique insights into the effects of demographic and clinical variables upon classifiers, and help pave the way for the implementation of multimodal electrophysiology clinical classifiers.

Read Full Article (External Site)