BMI 7320 Topics for Artificial Intelligence in Cancer Discovery

3 semester credit hours
Lecture contact hours: 2; Lab contact hours: 3
Web-based and classroom instruction
Prerequisite: Consent of instructor
Enrollment only open to PhD students

This course introduces a few common deep learning architectures (e.g., convolution neural network, graph neural network, recurrent neural network and autoencoder) to the students who are new to AI. The primary aim of this course is to flatten the learning curve in AI and to provide students with a basis for further implementation of more complex models using enormous real-world data, especially in cancer research.

This course will have a combination of lectures and demos to guarantee the students will have adequate first-hand experience with course concepts and with the opportunity to explore AI methods implemented in cancer research. We also include one tutorial of basic programming skills with Python and its machine learning libraries.

Upon successfully finishing this course, you will:

• Describe common artificial intelligence (AI) models, and their structure, including input data, interpretation of the model and output, and how they work.
• Assess the pros and cons of certain AI design and measurement on a given task.
• Apply several deep learning models to harness diverse types of biomedical data, especially in cancer research
• Evaluate the issues and challenges of implementing current AI approaches in cancer research.