Integrating Machine Learning with Biomedical Signal Processing and Systems Analysis: An Applications-based Course.

Area of Science:

• Biomedical Engineering Education
• Data Analytics in Medicine
• Physiological Signal Processing

Background:

• Traditional biomedical engineering curricula often separate machine learning from core engineering subjects like signal and systems analysis.
• There is a growing need to integrate data analytics and machine learning into biomedical engineering education to address complex physiological data.

Purpose of the Study:

• To propose and describe an innovative course that systematically integrates signal processing and systems analysis with machine learning techniques.
• To provide students with hands-on experience in applying machine learning algorithms to practical biomedical engineering problems.
• To enhance the analysis and interpretation of physiological signals within a unified educational framework.

Main Methods:

• The course is structured around four application-based modules: human activity recognition, epileptic seizure detection, respiratory-cardiac coupling quantification, and sleep apnea detection.
• Students engage in "ground up" data work within each module, applying learned algorithms to real physiological data.
• The curriculum emphasizes practical application and skill development in physiological signal analysis.

Main Results:

• Students gain practical experience in conditioning, analyzing, and interpreting diverse physiological signals using integrated machine learning and signal processing methods.
• The course facilitates a deeper understanding of applying advanced analytical techniques to complex biomedical datasets.
• The modular, applications-based approach allows for tailored learning experiences in physiological signal analysis.

Conclusions:

• The proposed integrative approach effectively bridges traditional engineering principles with modern machine learning for biomedical data analysis.
• This curriculum equips senior undergraduate and graduate biomedical engineering students, as well as clinician-scientists, with essential skills for analyzing physiological signals.
• The course offers a valuable model for enhancing biomedical engineering education through the synergistic application of signal processing and machine learning.