Teaching Strategies for Graduates with Different Specialty Background in Biomedical Signal Processing Course

Nini Rao, Liu Ding-yun, Wei Zeng2015 · DOI

Abstract

Biomedical signal processing is one of the key courses in the major of Biomedical Engineering. The teaching contents of this course are involved in multiple subject fields, such as biology, medicine, signal processing, computer science, and even mathematics/physics. Most graduates studying this course usually have a single subject background. Thus, it is a huge challenge to realize the teaching objectives of this course. This paper proposed five strategies to improve the teaching effects of this course. The teaching practice showed that the proposed methods are feasible and effective.

Introduction

Biomedical Signal Processing (BSP) serves as a cornerstone of the Biomedical Engineering curriculum, requiring students to synthesize knowledge across biology, medicine, mathematics, and computer science. However, graduate students typically enter the program with specialized backgrounds in only one of these domains. This creates a "silo" problem: engineering and computer science students often lack the physiological context to interpret biological signals, while biology and medicine students frequently struggle with the abstract mathematical foundations of signal theory, such as Fourier Transforms.

To address this disparity, a five-pillar pedagogical framework was developed to harmonize these diverse backgrounds and ensure all students meet the course objectives.

Methodology

The following five strategies were implemented to bridge the interdisciplinary gap:

  1. Foundational Leveling (The Math Bridge): Dedicated hours were allocated to establish basic signal concepts (digital systems, transforms) specifically for biology and medicine majors.
  2. Contextual Extension (The Bio Bridge): For engineering students, modules on anatomy and physiology were added. For instance, teaching ECG processing began with the electrophysiological origins of heart waves.
  3. Cross-Pollination Seminars: Students from biological backgrounds presented on medical problems, while engineering students presented on specific algorithms, facilitating peer-to-peer learning.
  4. Interdisciplinary Team Projects: Group assignments were structured to include a mix of biology, engineering, and CS students, requiring collaborative curriculum design.
  5. Bifurcated Final Examinations: Two distinct exam papers of equal difficulty were designed. One focused on application and medical interpretation (for bio-majors), while the other focused on algorithm development (for engineering majors).

Student Demographics

The strategies were evaluated using a cohort of 70 students at the University of Electronic Science and Technology of China in 2014. The diversity of the student body is illustrated in the major distribution below.

Major distribution of the 2014 student cohort showing a mix of BME, Biology/Medicine, EE, CS, and Math/Physics backgrounds.

Results and Discussion

The implementation of these strategies in 2014 resulted in significant quantitative improvements compared to the previous two years. The average examination score rose to 75, a notable increase from 69 in 2012 and 70 in 2013. Furthermore, the "Teaching Estimation" score—a composite metric from students and a teaching quality committee—reached 89, qualifying the course for an "Excellence" rating.

Comparison of average exam scores and teaching estimation scores from 2012 to 2014, showing a clear upward trend following the implementation of new strategies.

These results suggest that the dual-track assessment and peer-led knowledge transfer effectively reduced test anxiety and leveraged the classroom's diversity as an asset rather than a hurdle.

Conclusion

The proposed five-pillar framework successfully addressed the challenges of teaching an interdisciplinary BSP course to a diverse graduate student body. By tailoring both the instruction and the assessment to the students' primary fields of expertise, the course achieved higher academic performance and student satisfaction. These methods are feasible and effective for broader application in other interdisciplinary biomedical engineering courses.

Summary updated

Abstract

Biomedical signal processing is one of key courses in the major of Biomedical Engineering. The teaching contents of this course are involved in multiple subject fields, such as biology, medicine, signal processing, computer science and even mathematics/physics. Most graduates studying this course usually have a single of subject background. Thus, it is a huge of challenge to realize the teaching objectives of this course. This paper proposed five strategies to improve the teaching effects of this course. The teaching practice showed that the proposed methods are feasible and effective.
BibTeX
@misc{rao2015teaching,
  author = {Nini Rao and Liu Ding-yun and Wei Zeng},
  title = {{Teaching Strategies for Graduates with Different Specialty Background in Biomedical Signal Processing Course}},
  year = {2015},
  doi = {10.2991/icemit-15.2015.65},
}
Figure