This course provides an introduction to data science and machine learning for applications in biomedical engineering. The lectures cover topics in biomedical signal processing (1D convolution, denoising, filtering), biomedical image processing (2D convolution, denoising, edge detection, template matching), biomedical data reduction (feature extraction, principal component analysis), and biomedical data regression, classication (including deep learning), and clustering. Prior exposure to topics in signals and systems (convolution, Fourier transform), calculus III (constrained optimization), linear algebra (eigenvalue decomposition), and probability and statistics (expectation and variance) is required. The lab complements methods learnt in lectures by providing a hands-on experience in biomedical applications such as denoising, reconstruction and classication of action potentials, and reconstruction of biomedical images. Lab sessions will include writing Python scripts to analyze both synthetic and real data, so some familiarity with basic Python programming and Python notebooks (see this Python Tutorial) is advised.

Course Topics

• Data Processing: Fundamentals of digital signal and image processing, including 1D and 2D convolution and its application to signal denoising, edge detection, and template matching.

• Data Reduction: Feature extraction from signals and images (e.g., histogram of gradients). Representation of signals in terms of a fixed basis (e.g., an orthonormal Fourier basis) and a basis learned from the data via Principal Component Analysis (PCA).

• Data Regression: Linear regressions, least-squares and maximum likelihood. The problem of overfitting, ill-posedeness, and regularization.

• Data Classification: Parameter estimations and mean squared error: the bias-variance trade-off. Classifcation using Nearest-Neighbors. Linear classifiers and Support Vector Machines. Introduction to Convolutional Neural Networks. Training CNNs using backpropagation.

• Data Clustering: Unsupervised learning, K-Means clustering and bag of words.

Course Objectives

• Master the right tools to tackle biomedical signal and data processing problems.

• Have an intuitive understanding of data science methods through practical examples.

• Learn about topics that are at the forefront of biomedical data science research.

Course Material
All related course material will be posted on Blackboard and on the class webpage:

Reference Textbooks (relevant chapters in parenthesis)

• R. C. Gonzalez, and R. E. Woods. "Digital Image Processing", Pearson, 2018 (3.4,4.4,12.7,13.5).

• R. Vidal, Y. Ma, and S. Sastry, "Generalized Principal Component Analysis", Springer 2016 (2).

• C. M. Bishop, "Pattern Recognition and Machine Learning", Springer-Verlag, 2006 (3,4,5,12).

• R. O. Duda, P. E. Hart, and D. G. Stork, "Pattern Classication". Wiley-Interscience, 2000.

• T. Hastie, R. Tibshirani, and J. H. Friedman, "Elements of Statistical Learning", Springer, 2009.

1. Biomedical Data Processing
• 8/30: 1D Convolution and Signal Denoising
• 8/31: Review of Probability
• 9/3: (Labor day - no class)
• 9/5: 1D Edge Detection and Template Matching
• 9/7: Review of Statistics
• 9/10: 2D Convolution and Image Denoising
• 9/12: 2D Edge Detection and Template Matching
• Lab I Signal denoising 9/13: Heart beat rate estimation
• Lab I Signal denoising 9/14: Mobile health
2. Biomedical Data Reduction
• 9/17: Feature Extraction, Basis Representation
• 9/19: Principal Component Analysis
3. Biomedical Data Regreesion
• 9/24: Linear Regression, Overfitting
• 9/26: Regularized Linear Regression
• Lab II Data Reduction 9/27, 9/28 Regression Temporal segmentation of MRI sequence
4. Biomedical Data Classification
• 10/01: Introduction to classification, Nearest-Neighbor
• 10/03: Linear Classifiers, Support Vector Machine (SVM)
• 10/08: (Convolutional) Neural Networks
• 10/10: Training CNNs, Backpropagation
• Lab III NN Classifiers 10/11, 10/12 Linear Classifiers Action Potential Classification

Ability to apply mathematics, science and engineering principles (a).

Ability to design and conduct experiments, analyze and interpret data (b).

Ability to function on multidisciplinary teams (d).

Ability to identify, formulate and solve engineering problems (e).

Ability to communicate eectively (g).

Ability to use the techniques, skills and modern engineering tools necessary for engineering practice (k).

1. Lecture Policy: Homework assignments will be out each week after Wednesday's lecture and solutions must be submitted in PDF through Blackboard the next Wednesday by 11:59pm. Late submissions will not be accepted. There will be a total of 5 assignments, which will be worth 30% of your grade (6% per assignment).
2. Lab Policy: Lab reports are due for each of the three labs for 25% of the grade A final project will cover the other 25% of the grade. Lab reports must be submitted through Blackboard no later than 9:00 AM on Thursday morning the week after the lab. Late lab reports will lose 50 points until 5 PM on Thursday. Lab reports will not be accepted after 5 PM..

The strength of the university depends on academic and personal integrity. In this course, you must be honest and truthful. Ethical violations include cheating on exams, plagiarism, reuse of assignments, improper use of the Internet and electronic devices, unauthorized collaboration, alteration of graded assignments, forgery and falsification, lying, facilitating academic dishonesty, and unfair competition.

1. All assignments must be entirely done on your own. We suggest that when you work together you do not take pictures or copy anything down on paper. Discuss the problems, work through the approach on the board, and then do the assignment from scratch on your own.
2. During lab sessions students are encouraged to work together (in pairs) and discuss methods for performing the lab assignments. However, all other assignments outside the lab session such as pre-quizzes (if any) must be entirely your own.
3. Copies of all code must be included with your lab report.

Report any violations you witness to the instructor. You can find more information aboaut university misconduct policies on the web at these sites:

For undergraduates:

For graduate students:

Students with Disabilities
Students with Disabilities Any student with a disability who may need accommodations in this class must obtain an accommodation letter from Student Disability Services, 385 Garland, (410) 516-4720.