Course Details
Subject {L-T-P / C} : CS6606 : AI in Healthcare { 3-0-0 / 3}
Subject Nature : Theory
Coordinator : Puneet Kumar Jain
Syllabus
UNIT – I: Fundamentals of AI in Healthcare
Role of AI in Healthcare, Data acquisition concepts: Sampling, quantization and interpolation, Data filtering: FIR and IIR filters, Image filtering techniques, Data analysis: Time-domain, frequency-domain and time-frequency-domain analysis,
UNIT – II: AI for Cardiovascular signals
Electrocardiogram (ECG), Cardiac physiology, Electrophysiology, ECG introduction, Cardiovascular diseases and ECG, Processing and feature extraction of ECG, AI/ML models for ECG, Phonocardiogram(PCG), PCG introduction and overview, Analysis(Acquisition, Filtering, and feature extraction) of PCG, AI/ML models for PCG
UNIT – III: AI for Electroencephalogram
EEG: signal of the brain, Brain and its function, EEG introduction and diagnostic purpose, Processing and Feature extraction of EEG
AI/ML models for EEG
UNIT – IV: AI for Biomedical images
Computed Tomography, X-ray imaging and computed tomography, Introduction and Principle of X-ray, AI/ML for classifying diseases on x-rays, Magnetic resonance imaging, Physics and Physiological principles of MRI, AI/ML to process MRI images, Ultrasound imaging, Physical and physiological principles of ultrasound
Course Objectives
- Understand the foundations of the physiological system and then study cutting-edge AI/ML models in the context of various healthcare data, including cardiovascular signals, brain signals, and medical imaging modalities.
- Exploring the potential of AI methods, emphasizing machine learning and applying them to specific areas in medicine and healthcare.
- Understand the challenges of regulation of AI applications and which model components can be regulated.
- As a project-based course, applying cutting-edge machine learning techniques to concrete problems in modern medicine.
Course Outcomes
On completion of this course, the students will have the ability to: CO1: Understand the principle of various biomedical signals and images. CO2: Understand the physiological characteristics provided by the biomedical signals and images, such as the reading of an ECG signal. CO3: Analyze the performance of specific AI/ML models as applied to biomedical problems and justify their use and limitations. CO4: Identify and apply appropriate AI/ML models and computational tools to specific problems in biomedicine and healthcare. CO5: Comprehend a collection of AI/ML models and their applications in medicine and healthcare.
Essential Reading
- Kayvan Najarian, Robert Splinter, Biomedical Signal and Image Processing, CRC Press, Taylor and Francis Group , 2nd Edition, 2012
- Saravanan Krishnan, Ramesh Kesavan, B. Surendiran, Handbook of Artificial Intelligence in Biomedical Engineering, CRC Press, Taylor and Francis Group , 1st edition, 2021
Supplementary Reading
- DC Reddy, Biomedical Signal Processing – Principles and Techniques, Tata McGraw Hill Publishing company Ltd. , 2005
- Donna L. Hudson, Maurice E. Cohen, 4. Neural Networks and Artificial Intelligence for Biomedical Engineering, IEEE Press Series on Biomedical Engineering , 1999
Journal and Conferences
- IEEE transactions on Biomedical Engineering
- Biomedical Signal Processing and Control