Overview

Description

CALENDAR DESCRIPTION:

The principles involved in using computers for data acquisition, real-time processing, pattern recognition and experimental control in biology and medicine will be developed. The use of large data bases and simulation will be explored.

COURSE DETAILS:

Overview

Biomedical computing is a rapidly emerging area that is revolutionizing medicine and biology. It focuses on the use of computational and mathematical techniques to accelerate scientific discovery and to improve diagnosis, treatment, and understanding of diseases. The general objectives of the course are: To give the student breadth in the topics related to the use of computers in biology, medicine, and healthcare, in both theory and application. To enable the student to converse knowledgeably with radiologists, physicians, and MDs about these topics and related technologies. To learn the technical basis of many aspects of the use of technology in biomedicine (e.g. data acquisition, biosignal/image processing and visualization, feature representation and extraction, data interpretation and machine learning), current and future trends, and their benefits and limitations.

Topics

• Introduction and motivation to biomedical computing
• Biodata (signals, images, point-clouds) modalities and representations
• Fourier transform and convolution
• Biosignal digitization (sampling theorem and quantization)
• Biosignal enhancement  (temporal/spatial and frequency domain filtering)
• Feature extraction from biodata (1D, 2D, 3D, shapes, networks)
• Decision support (basic classification and multiple features/multiple classes)
• Dimensionality reduction using principal component analysis and linear discriminant analysis (PCA and LDA)
• Bayes' rule.
• Machine learning techniques (decision forests, artificial/convolutional neural networks), time permitting
• Project work: application of computational techniques to biomedical applications

 

COURSE-LEVEL EDUCATIONAL GOALS:

Learning Incomes

Upon entry, every student is expected to be able to:

• write code using some programming or scientific computing language (e.g. C/C++, Java, Python, MATLAB)
• basic familiarity with version control and markup languages (so you are ready to use or learn to use GitHub and LaTex on Overleaf)
• understand documentation for using and calling functions from external libraries 
• be able to follow instructions for installing software and performing other basic IT tasks
• understand high-school level concept in physics, biology, and chemistry
• understand high-school and first-year undergraduate level concepts in mathematics (e.g. calculus, probability, trigonometry, vectors and matrices)

Learning Outcomes

At the end of the course, students are expected to be able to:

• define different biomedical data acquisition devices and biosignals
• explain the different mathematical and computer representation of biodata modalities
• give several examples of biomedical applications of computing
• extract various types of features from multi-modal bio-data
• appreciate the meaning and utility of some mathematical and computational methods and concepts and their application to biomedical data (e.g. Fourier analysis, convolution, correlation, n-D histogram, high-dimensional data, probability distributions, entropy, covariance matrix, eigen-decomposition)
• understand the meaning of and perform basic feature selection and dimensionality reduction
• appreciate how computing plays a role in decision support
• represent some biomedical tasks (e.g. diagnosis) as classification problems
• calculate basic classification evaluation measures (e.g. true positive, confusion matrix, area under ROC)
• state and utilize Bayes’ theorem for estimating answers to health-related problems
• appreciate the strengths and limitations of using computers for biomedical applications

Grading

Materials

RECOMMENDED READING:

REQUIRED READING NOTES: