BMI 702 | Biomedical Artificial Intelligence

Harvard - Foundations of Biomedical Informatics II, Spring 2024

Artificial intelligence is poised to enable breakthroughs in science and reshape medicine. This course provides a survey of artificial intelligence for biomedical informatics, covering methods for key data modalities: clinical data, networks, language, and images. It introduces machine learning problems from a practical perspective, focusing on tasks that drive the adoption of machine learning in biology and medicine.

The curriculum delves into foundational algorithms and highlights the nuances of handling biomedical data. It places a strong emphasis on strategies for evaluating and seamlessly integrating machine learning methods into biomedical research and clinical practice. A key aspect of this course is its focus on the broader implications of artificial intelligence. This includes critical discussions on topics such as trustworthiness, interpretability, evaluation, and the ethical and legal challenges associated with the implementation of artificial intelligence in healthcare.

Marinka Zitnik

marinka@hms.harvard.edu

Office Hours: Thu, 4pm - 5pm

Week 1

Course overview and introduction to biomedical AI

Jan 25

LectureWhat makes biomedical problems unique: Slides, Reading List

Jan 26

QuizWeek 2 pre-class quiz (due Feb 1): Canvas

Week 2

Machine learning using electronic health records, subtype discovery, disease diagnosis and prognosis prediction

Feb 1

Module 1LectureClinical AI Part I: Slides, Reading List

Feb 2

QuizWeek 3 pre-class quiz (due Feb 8): Canvas

Week 3

Machine learning on clinico-genetic datasets, diagnostic odyssey and therapy selection

Feb 8

Module 1LectureClinical AI Part II: Slides, Reading List

Feb 9

QuizWeek 4 pre-class quiz (due Feb 15): Canvas

Week 4

Interpretability and explainability

Feb 15

Module 2LectureTrustworthy & Efficient AI Part I: Slides, Reading List

Feb 16

QuizWeek 5 pre-class quiz (due Feb 22): Canvas

Week 5

Fairness, bias, distribution shifts, and robustness

Feb 22

Module 2LectureTrustworthy & Efficient AI Part II: Slides, Reading List

Feb 23

QuizWeek 6 pre-class quiz (due Feb 29): Canvas

Feb 23

PSet releasedPSet 1: Bias, trustworthiness, and fairness: Canvas

Week 6

Foundations of graph representation learning, link prediction, node classification, graph clustering, graph classification, semi-supervised learning, label propagation, network medicine, and disease modules.

Feb 29

Module 3LectureBiomedical graph learning Part I: Slides, Reading List

Mar 1

QuizWeek 7 pre-class quiz (due Mar 7): Canvas

Week 7

Machine learning with heterogeneous graphs, graph neural networks, knowledge graph embeddings, prediction and reasoning using knowledge graphs

Mar 7

Module 3LectureBiomedical graph learning Part II: Slides, Reading List

Mar 8

QuizWeek 8 pre-class quiz (due Mar 21): Canvas

Mar 8

PSet duePSet 1: Bias, trustworthiness, and fairness: Canvas

Week 8

Large language models (Guest lecture by Dr. Jonathan Richard Schwarz)

Mar 21

Module 4LectureMedical language modeling Part I: Slides, Reading List

Mar 22

QuizWeek 9 pre-class quiz (due Mar 28): Canvas

Mar 22

PSet releasedPSet 2: Knowledge graphs and geometric deep learning: Canvas

Week 9

Natural language processing across clinical domains and tasks

Mar 28

Module 4LectureMedical language modeling Part II: Slides, Reading List

Apr 29

QuizWeek 10 pre-class quiz (due Apr 4): Canvas

Week 10

Vision and vision-language pre-training, towards generic vision interface, multimodal LLMs

Apr 4

Module 5LectureBiomedical imaging Part I: Slides, Reading List

Apr 5

QuizWeek 11 pre-class quiz (due Apr 11): Canvas

Apr 5

PSet duePSet 2: Knowledge graphs and geometric deep learning: Canvas

Week 11

Multimodal learning, computational pathology with applications to oncology, analysis of histopathology slides

Apr 11

Module 5LectureBiomedical imaging Part II: Slides, Reading List

Apr 12

QuizWeek 12 pre-class quiz (due Apr 18): Canvas

Apr 12

PSet releasedPSet 3: Biomedical imaging methods and applications: Canvas

Week 12

Introduction to ethical frameworks, data privacy, regulation and liability aspects of AI (Guest lecture by Dr. Sara Gerke)

April 18

LectureEthical and legal considerations for biomedical AI: Slides, Reading List

Apr 19

QuizWeek 13 pre-class quiz (due Apr 25): Canvas

Week 13

AI-guided drug design, small-molecule generation, molecule optimization, identification and characterization of therapeutic targets, design of chemical and genetic perturbations

Apr 25

Module 6LectureGenerative AI Part I: Slides, Reading List

Apr 26

QuizWeek 14 pre-class quiz (due May 2): Canvas

Apr 26

PSet duePSet 3: Biomedical imaging methods and applications: Canvas

Week 14

Protein generation and optimization, other emerging uses of generative AI

May 2

Module 6LectureGenerative AI Part II: Slides, Reading List