Applied Medical Image Processing and Analysis

BME59500

Credit Hours:

3

Learning Objective:

Upon successful completion of this course, students will be able to:

  1. Demonstrate advanced knowledge of medical imaging modalities and image processing techniques, including the physical principles behind MRI, CT, ultrasound, PET, and microscopy.
  2. Apply engineering and computational principles to enhance, segment, register, and visualize medical imaging data across preclinical and clinical contexts.
  3. Develop and implement reproducible workflows using tools such as MATLAB, Python, ImageJ, and 3D Slicer to extract quantitative metrics from biomedical images.
  4. Formulate and solve real-world biomedical problems by designing image analysis strategies tailored to clinical or experimental objectives.
  5. Critically evaluate the strengths and limitations of different image processing techniques, selecting appropriate methods based on imaging modality, data characteristics, and research goals.
  6. Communicate complex technical content effectively through written reports, oral presentations, and data visualizations suitable for both engineering and interdisciplinary audiences.
  7. Recognize the broader impact of medical image analysis in clinical decision-making, research innovation, and healthcare equity, demonstrating awareness of professional and societal responsibilities. 

Description:

Imaging science is rapidly advancing, driving innovation across healthcare and biomedical research. Biomedical engineering has been ranked among the fastest-growing career fields, with medical imaging identified as a major contributor. As a core component of biomedical imaging, medical image processing and analysis are critical for understanding, visualizing, and quantifying anatomical and physiological features in both clinical and research settings. Automated and quantitative image analysis techniques are making disease diagnosis faster, more accurate, and more reproducible—accelerating progress in medical research and patient care.


Advances in imaging systems such as magnetic resonance imaging (MRI), computed tomography (CT), ultrasound, and microscopy have led to the generation of vast, high-resolution datasets. Extracting clinically relevant information from these images presents a major challenge—and opportunity—for engineers and scientists. This course equips students with the computational tools and analytical frameworks needed to transform raw image data into meaningful biomedical insights.
BME595: Applied Medical Image Processing and Analysis is a graduate-level course that meets this need by developing practical skills in image processing, quantitative analysis, and data visualization. The course begins with an overview of imaging physics and progresses to hands-on implementation of algorithms for enhancement, segmentation, registration, and visualization using tools such as MATLAB, Python, 3D Slicer, and ImageJ. It concludes with an introduction to deep learning for medical image segmentation, providing exposure to current computational methods in the field.

 

Topics Covered:

  • Foundations of Medical Imaging and Data
  • Image Preprocessing and Enhancement
  • Image Analysis and Quantification
  • Image Registration and Visualization
  • Modern Computational Approaches in Medical Imaging

 

Prerequisites:

  • Basic knowledge of programming (MATLAB and Python will be used)
  • Basic knowledge of linear algebra and calculus

 

Web Address:

https://purdue.brightspace.com

 

Homework:

Weekly/biweekly assignments (20%) Hands-on coding exercises in MATLAB, Python, ImageJ, or 3D Slicer to reinforce core concepts such as enhancement, segmentation, and registration

 

Projects:

Mid-semester project (25%) Applied analysis using a selected imaging modality. Includes image processing pipeline, quantitative interpretation, and a brief written report.

Technical presentation (10%) Oral presentation of project results with visualizations; emphasizes communication and interpretation of results to interdisciplinary audiences.

Final project (35%) Capstone project applying a full workflow (enhancement, segmentation, registration, visualization), including an introductory deep learning module. Students will submit code, analysis, and a final report.

 

Exams:

Concept quizzes (10%) Short, low-stakes quizzes to assess understanding of imaging modality physics, image formats, and algorithm basics.

 

Textbooks:

  • Course lecture notes/powerpoints and various book chapters (supplied as a pdf and posted in Brightspace) will be the main source for the course.
  • The following books are optional. These are great books to fuel your interest in image processing!
    • Deep Learning, Goodfellow Ian, Bengio Yoshua, and Courville Aaron, 2016, Freely available, MIT Press.
    • Applied Medical Image Processing – A Basic Course, Wolfgang Birkfellner, 2024. ISBN 9781032127675
    • Learn Python with Jupyter, a Free and Excellent E-book developed by my colleague and friend, Dr. Serena Bonaretti here

 

Computer Requirements: