Course

Signals and systems: Review, Medical Imaging Modalities and the need for different modalities (MRI, CT, OCT for Retinal Images, PET, X-Ray, Ultra Sound, Microscopy, Flow Cytometry, Imaging Flow Cytometry, etc. Pre-processing – Image Enhancement – Focus Analysis – Noise reduction

(Additive and Speckle Noise) – Image Quality Measures – Domain Transformation: Fourier Domain and Wavelet Domain- Thermal Imaging. Basic electrocardiography, ECG lead systems, ECG signal characteristics

Medical Image Segmentation – Deep Learning based Segmentation on 2D or 3D volume of Data Feature Extraction – Morphological Features – Textural Features –, Feature extraction for 1D Biomedical signals– Deep Features. Image Registration and Fusion — Key Point Matching – Geometric transformations. ECG data acquisition, ECG lead system, ECG signal characteristics (parameters and their estimation), Analog filters, ECG amplifier, and QRS detector, Power spectrum of the ECG, Band pass filtering techniques, Differentiation techniques, Template matching techniques, A QRS detection algorithm

Classification and Clustering– Examples of image classification for diagnostic/assistive technologies –Deep learning-based classifiers. 3D volume reconstruction – Reconstruction techniques for CT, MRI, Reconstruction of cell structure from focus stack of images – CT and MRI volume reconstruction – Wavelet based Volume Rendering, Applications of EEG.

Preamble

The development of Medical Imaging over the past four decades have been truly revolutionary. It is therefore essential to develop knowledge in Medical Signal Processing, stepping out of the conventional notion of extending the art of instrumentation in biomedicine. The course aims to provide the graduate students with a detailed background of state-of-the-art electrical engineering practices used in biomedical engineering. The course aims to provide an understanding of various image modalities captured based on various signal processing techniques. This course will also cover the application of various Deep learning techniques, for segmentation as well as classification problems.

Course Objectives

• To understand various signals and the image modalities in the field of Biomedical.
• To study origins and characteristics of some of the most commonly used biomedical signals like ECG.
• To explore research domain in biomedical signal processing.
• To understand various reconstruction techniques for CT and MRI.

Course Outcomes

Prerequisites

• None.

CO-PO Mapping

Evaluation Pattern – 50:50

• Midterm Exam – 30%
• Continuous Evaluation – 20%
• End Semester Exam – 50%

Text Book / References

1. Klaus D. Toennies, ”Guide to Medical Image Analysis – Methods and Algorithms”, Advances in Computer Vision and Pattern Recognition, 2nd Edition, Springer-Verlag London, 2017, DOI: 10.1007/978-1-4471-7320-5, ISBN 978-1-4471-7318-2
2. Geoff Dougherty, “Medical Image Processing Techniques and application”, Springer New York 2011
3. Mostafa Analoui, Joseph D. Bronzino, Donald R. Peterson, “Medical Imaging: Principles and Practices”, Taylor and Francis group, 2012
4. Analyzing Neural Time Series Data-Theory and Practice (MIT Press) 2014