- Description
- Learning Outcomes
- Logistics
- Schedule
- Prerequisites
- Expectations
- Folder Structure
- Acknowledgements
In this course, Topics in Deep Learning: Healthcare, participants will familiarize themselves with the key concepts and challenges of applying AI/ML technologies in the healthcare (HC) space. Particular emphasis will be placed on understanding why it is challenging to translate in-silico tools to clinical and biomedical practice. By the end of this course, participants will understand the organizational and institutional requirements for successful AI/ML adoption in HC, the current "cutting edge" applications, and specific technical and statistical tools that are needed in the HC practice area.
By the end of the course, students will be able to:
- Learning outcome 1: Be able to articulate and problem solve the key challenges of ML/AI adoption in healthcare
- Learning outcome 2: Identify and understand areas of value creation in the healthcare domain using ML/AI in a commercial and business context
- Learning outcome 3: Use Python-based tools for healthcare focused ML/AI problems
- Instructor: Erik Drysdale (MSc, MA, BA). Emails to the instructor can be sent to [email protected]
- TA: Jenny Du (BSc). Emails to the teaching assistant can be sent to [email protected]
- Email etiquette: Remember to put [DSI] in email subject line
The course runs synchronously over Zoom. The course consists of ten classes over three weeks. Classes are 6:00 PM - 8:30 PM EDT on weekdays. To mitigate online fatigue, each class will include one or two breaks, during which students are encouraged to stretch, grab a drink and snacks, or ask additional questions.
Tutorial sessions with a TA will also be offered over Zoom. These will take place from 5:30 PM - 6:00 PM EDT and 8:30 PM - 9:00 PM on weekdays.
The schedule is tentative and may be modified as needed. Learners will be notified of schedule changes.
- Day 1 (Tuesday, February 20, 6:00 PM - 8:30 PM): Overview: the practice of healthcare, medicine, and life sciences
- Day 2 (Wednesday, February 21, 6:00 PM - 8:30 PM): Implementing AI in healthcare #1
- Day 3 (Thursday, February 22, 6:00 PM - 8:30 PM): Implementing AI in healthcare #2
- Day 4 (Monday, February 26, 6:00 PM - 8:30 PM): Implementing AI in healthcare #2 (cont'd)
- Day 5 (Tuesday, February 27, 6:00 PM - 8:30 PM): Protein folding, drug discovery, and medical imaging, and ‘Omics
- Day 6 (Wednesday, February 28, 6:00 PM - 8:30 PM): Commercial applications
- Day 7 (Thursday, February 29, 6:00 PM - 8:30 PM): "Black box” model explainability
- Day 8 (Tuesday, March 5, 6:00 PM - 8:30 PM): "Black box” model explainability (cont'd)
- Day 9 (Wednesday, March 6, 6:00 PM - 8:30 PM): Guest lecture (AI-based drug discovery)
- Day 10 (Thursday, March 7, 6:00 PM - 8:30 PM): Business thinking and organization structure
Learners are expected to have completed the DSI foundational courses (i.e., Unix Shell/Git and GitHub, Python, R, Building Research Software/SQL, Estimation/Machine Learning/Testing, Production) as well as the Deep Learning Foundations course.
Learners should be active participants and are encouraged to ask questions throughout.
- Participants require a computer and internet connection to engage in online activities.
- lessons: Course slides as pdf files
- slide-resources: Course slides as md files
- homework: Optional homework to practice concepts covered in class
- assignments: Graded assignments
- README: This file!
- LICENSE: Copyright information for these materials
- .gitignore: Files to exclude from this folder, specified by the instructor
- Overview: the practice of healthcare, medicine, and life sciences
- Implementing AI in healthcare #1
- Implementing AI in healthcare #2
Survival modeling- Protein folding, drug discovery, and medical imaging, and ‘Omics
- Commercial applications
“White box” model explainability- "Black box” model explainability
Prediction intervals- (Under construction) Business thinking and organization structure
- Class attendance: To ensure everyone actively participates in class activities, attendance is mandatory and will be monitored. If you are unable to attend class, it is your responsibility to make up the work that was covered.
- Assignment 1
- Assignment 2
| Format | Details | Submission Instructions |
|---|---|---|
| Assignment 1 | Due on Feb 28th (Wed) by 5:59pm | Submit assignment via email |
| Assignment 2 | Due on Mar 8th (Friday) by 11:59pm | Submit assignment via email (zip folder) |
Course materials were developed by Julia Gallucci, Kaylie Lau, and Santiago Arciniegas under the supervision of Rohan Alexander and finalized by Erik Drysdale.