The field of physiological signal processing and analysis is moving towards the development of more robust and accurate methods for detecting and classifying various physiological signals. Researchers are exploring the use of multi-modal approaches, combining different signal types to improve classification performance and robustness. Additionally, there is a growing interest in explainable machine learning models, which provide transparent and interpretable results, allowing for better understanding and decision-making. Another trend is the development of lightweight and energy-efficient algorithms for real-time processing and analysis of physiological signals, enabling their use in wearable devices and mobile applications. Noteworthy papers in this area include:

• A study on automated fatigue detection using physiological signals, which demonstrated the effectiveness of feature-level fusion of multiple signal types.
• A paper on outlier detection in plantar pressure data, which highlighted the complementary potential of statistical parametric mapping and explainable machine learning approaches.
• A work on real-time noise detection and classification in single-channel EEG, which proposed a lightweight machine learning approach that outperformed advanced deep learning models.