HAD7001H-S

Prerequisite

N/A

Objectives

• Develop an understanding of the fundamental concepts behind different health data modalities and
  their processing, and learn how AI methods are applied to medical images, clinical text, audio, and
  multimodal data in healthcare and medicine.
• Acquire the ability to select, design, and implement appropriate AI approaches for applied health
  research. This includes building, fitting, and evaluating models across different data modalities,
  interpreting results, and determining the suitability of AI methods for applications such as diagnosis,
  prognosis, and patient monitoring.
• Gain experience with Python programming for carrying out data preparation, model training, and
  performance evaluation using real-world healthcare datasets.
• Develop an understanding of ethical considerations and challenges in applying AI to healthcare,
  including issues of data quality, interpretability, bias, fairness, generalizability, transparency, privacy,
  deployment, and responsible use.

Description

Artificial Intelligence (AI) is rapidly transforming healthcare by enabling new approaches to
diagnosis, prognosis, and personalized treatment. This course provides a comprehensive
introduction to applied AI in health and medicine, with a focus on medical images, clinical text
(including large language models), audio (e.g., patient voice recordings), and multimodal data.
Generative AI concepts are also introduced in the context of health applications. Students will first
learn fundamental methods for processing and analyzing different health data modalities before
exploring how AI techniques are applied to each. The course emphasizes both theoretical
understanding and hands-on practice using Python. Weekly datathon exercises and a multi-phase
course project allow students to apply AI methods to realistic healthcare datasets, reinforcing key
concepts and practical skills. The course also addresses current challenges in applying AI within
healthcare and public health, including data quality, interpretability, bias, generalizability,
transparency, privacy, deployment, and ethics. By the end of the course, students will be able to
design, implement, and critically assess AI approaches across multiple health data modalities.

Instructors

Evaluation Breakdown

Competencies