CS372 Artificial
Intelligence for
Disease Diagnosis
and Information
Recommendations
Spring 2020
--- SARS-CoV-2 (COVID-19) Edition
3 units/20
sessions, including lectures and project discussion/presentation.
Edward Chang
(echang@cs.stanford.edu)
Adjunct Professor, Computer Science, Stanford University
President, DeepQ Healthcare
(areas: AI and healthcare)
Technical Advisor, SmartNews (areas: NLP and NLU)
TA: Saahil Jain (sj2675@cs.stanford.edu)
Announcements
(3/18/20) Due to the COVID-19 pandemic, the 2020
edition of this course will focus on addressing COVID-19 prediction,
containment, and mitigation
(in addition to covering cancer, depression,
and common diseases), using statistical models and AI methods.
(3/29/20) For
COVID-19 advice and information, please visit CDC & NIH Websites.
o Get the
latest public health information from CDC: https://www.coronavirus.gov
o Get the latest research information from NIH: https://www.nih.gov/coronavirus.
About
Artificial
intelligence, specifically deep learning, stands out as one of the most
transformative technologies of the past decade. AI can already outperform humans in several
computer vision and natural language processing tasks. However, we still face
some of the same limitations and obstacles that led to the
demise of the first AI boom phase five decades ago. This research-oriented course will first review and reveal the limitations (e.g., iid
assumption on training and testing data, voluminous training data requirement,
and lacking interpretability) of some widely used AI
algorithms, including convolutional neural networks (CNNs), transformers,
reinforcement learning, and generative adversarial networks (GANs). To address these limitations, we will then explore topics including transfer
learning for remedying data
scarcity, knowledge-guided multimodal learning for improving
data diversity, out of distribution generalization, subspace learning for
enabling interpretability, and privacy-preserving data management.
The course will be taught through
a combination of lecture and project sessions. Lectures on specialized AI
applications (e.g., cancer/depression diagnosis and treatment) will feature
guest speakers from academia and industry. Students will be assigned to work on
an extensive project that is relevant to their fields of study (e.g., CS,
Medicine, and Data Science). Projects may involve conducting literature
surveys, formulating ideas, and implementing these ideas. Example project
topics are but not limited to 1) knowledge guided GANs for improving training
data diversity, 2) disease diagnosis via multimodal symptom checking, and 3)
methods for COVID-19 path prediction and immunity analysis. Feel free to bring
your ideas.
Level: CS senior/graduate,
proficient in programming
Perquisite: Introductory
course in AI, Statistics, or Machine Learning
Time
and Location
·
Time: Tuesdays and Thursdays
9:00am to 10:20am, starting on April 7th.
·
Location: On Zoom (link is
provided at Stanford AXESS, public access is prohibited)
Assignments
·
No exams.
·
Two assignments in the first half of the
quarter.
·
Reading: A list of papers/articles/podcasts
grouped by topics. Each student selects
two topics of interest to conduct in-depth surveys and presentation.
·
Term project: An extensive design and coding
project. Some example projects are:
a.
COVID-19 symptom checker.
b.
COVID-19 outbreak tracking and prediction
using data driven approach.
c.
Thoracic disease diagnosis with limited
training data.
d.
A Virtual Assistant for news and/or COVID-19
information recommendations.
e.
Methods for fusing knowledge and perception
with deep learning.
Grading
·
Assignments 10%
·
Participation/Discussion 10%
·
Term Project 80% broken down into five
factors
o Proposal
(functional specifications) 10%
o Literature
survey and algorithm design 10%
o Implementation
& Demo quality 40%
o Presentation/documentation
10%
o Effort
10% (ranked within a team)
Syllabus
|
Date |
Description |
Course
Materials |
Assignments |
|
4/7 |
Deep
Learning in Healthcare and
Course Logistics |
[1] Schwartz, W. B., R. S. Patil, and
P. Szolovits. "AI in Medicine: where do we stand." New England Journal of
Medicine 316
(1987): 685-688 [link]. [2]
Edward Y. Chang, et al., "Context-Aware Symptom Checking for Disease
Diagnosis Using Hierarchical Reinforcement Learning." AAAI (2018). [3]
Edward Y. Chang, et al., "Artificial Intelligence in XPRIZE DeepQ Tricorder. " ACM MM Workshop for Personal
Health and Health Care (2017). |
|
|
4/9 |
[1]
Small training data challenge; [2]
COVIS-19, What are we facing?
Contrarian opinions? What can AI help? |
[1]
The next outbreak? We’re not ready | Bill Gates [link], April 2015. [2]
What is a Virus, Vincent Racaniello,Columbia
University [link].
[3]
What is COVID-19 [link].
[4]
Virus Classification, the Baltimore System [link].
[5]
Coronavirus: Why You Must Act Now? [link]. [6]
How to act? Cost and benefit of social distancing [link][link]. [7]
An infected ICU doctor’s frontline experience [link]. |
Assignment
#1: COVID-19
Analytical Models & Mitigation Strategies [Please
visit Canvas] Deadline
4/18 |
|
4/14 |
The
Science of COVID-19: virus, infection cycle, models, mitigation strategies,
and data sources |
[1]
Exponential growth and Epidemics [link]. [2]
Estimating actual COVID-19 cases [link]. [3]
COVID-19 public dataset program [link]. [4]
COVID-19: Predicting the Path and Analyzing Immunity [link]. [5]
IDM’s innovative software tools provide a quantitative and analytical means
to model infectious disease [link]. [6]
Impact of non-pharmaceutical interventions (NPIs) to reduce COVID19
mortality and healthcare demand, London Imperial College [link]. [7]
How much ‘normal’ risk does COVID-19 represent? [link] |
|
|
4/16 |
Multimodal
NLP: Word2vec, Transformers, BERT, GPT, Reformers, and beyond |
[1]
Attention is all you need, Ashish Vaswani, et al., [link]. [2]
BERT: Pre-training of Deep Bidirectional Transformers for
Language Understanding [link]. [3] Reformer: the
Efficient Transformer [link]. |
Assignment
#2: COVID-19
Vaccines and Treatments [Please
visit Canvas] Deadline
4/28 |
|
4/21 |
Tutorials on term projects,
datasets and resources |
Suggested Topics: [1] Classifying thoracic diseases
on small data (Jason Chou, DeepQ) [2] Document/news classification on
small training data w/ OOD challenge (Jim Deng, SmartNews) [3] COVID-19 virtual assistant (Stanford OVAL) [4] COVID-19 simulation
modeling [5] Your own project(s) |
Assignment #3: Term Project Proposals Due on 4/30 |
|
4/23 |
Project Q&As COVID-19 Vaccines and Antivirals |
[1]
COVID-19 Vaccines (link listed on assignment #2) [2] COVID-19 Antivirals (link
listed on assignment #2) [3] Milken Institute COVID-19
treatment and vaccine tracker [link]. [4] The science behind COVID-19,
Series II, Patrick Soon-Shiong [link]. [5] The first modern pandemic,
Gates Notes, published today (4/23/2020) [link]. |
|
|
4/28 |
Project
Proposal Presentation Healthcare
Challenges and Opportunities for a Paradigm Shift |
Each
includes 1) key objective, 2) related work, 3)
proposed methods, and 4) data sources. [1]
Ted Talk on ”Future of Healthcare” by Edward Y. Chang, Taiwan, December 2017
(in Mandarina) [Link] [2]
AI joins the fight against COVID-19, Op-Med, published today (4/28/2020) [Link]. |
|
|
4/30 |
Principles of Diagnosis, Models and Algorithms |
[1] Edward Y. Chang et al.,
"Context-Aware Symptom Checking for Disease Diagnosis Using
Hierarchical Reinforcement Learning." AAAI (2018). [2] Edward Y. Chang et al.,
"REFUEL: Exploring Sparse Features in Deep Reinforcement Learning for
Fast Disease Diagnosis." NIPS (2018). [3] Szolovits,
P., and S. G. Pauker. "Categorical and
Probabilistic Reasoning in Medical Diagnosis.” Artificial Intelligence 11(1-2),
1978: 115-144. [4] Wu, T. D. "Efficient
Diagnosis of Multiple Disorders Based on a Symptom Clustering
Approach." Proceedings of AAAI, 1990, pp. 357-364. [5] Patil, R. S., P. Szolovits, and W. B. Schwartz. "Causal
Understanding of Patient Illness in Medical Diagnosis."
In Proceedings of the Seventh International Joint Conference on
Artificial Intelligence. |
|
|
5/5 |
Cancer Causes and Diagnosis |
[1] Principles and methods of integrative genomic analyses in
cancer, Vessela N. Kristensen, Ole Christian Lingjærde, Hege G. Russnes,
Hans Kristian M. Vollan, et al., Nature Reviews,
2014. [2] Biomarker development in the precision medicine era: lung
cancer as a case study, Ashley J. Vargas, and Curtis C. Harris, Natural
Reviews, 2016. [optional] Artificial intelligence in radiology, Ahmed Hosny, Chintan Parmar, John Quackenbush, Lawrence H. Schwartz,and Hugo J. W. L. Aerts, Nature Reviews, 2018. |
Dr. Melissa Ko, PhD, Cancer Biology, Stanford |
|
5/7 |
AI for Psychiatry: Beyond Brainless and Mindless? |
[1] Students Face Mental Health
Challenges --- barriers to care (during COVID-19)”, Stanford Daily
(4/30/2020) [link]. [2]Meditation
Music: Sangchhen Dorji
Lhuendrup Nunnery, Bhutan, Himalayas,
Edward Y. Chang [link]. |
Dr. Vikas Duvvuri,
MD/PhD Stanford |
|
5/12 |
Cancer Treatment and How AI May
Help |
[1] Mass cytometry: blessed with the curse of dimensionality, Evan
W Newell & Yang Cheng, Nature Immunology, June 2016. [2] Next-Generation Machine Learning for Biological Networks, Diogo M. Camacho, Katherine M. Collins, Rani K. Powers,
James C. Costello, and James J. Collins, Leading Edge Review, June 2018. [optional] Personalized Cancer Models for Target, Discovery and
Precision Medicine, Carla Grandori1 and Christopher J. Kemp, Trends in
Cancer, CellPress Reviews, September 2018. |
Dr. Melissa Ko, PhD, Cancer Biology, Stanford |
|
5/14 |
Project: literature surveys and
algorithm reviews 9am: session #1 [video] 2pm: session #2 [on Axess] Lecture: Explainable/Interpretable
Deep Learning [Slides] |
[1] Zellers, R., Bisk, Y.,
Farhadi, A., & Choi, Y. (2019). From recognition to cognition: Visual
commonsense reasoning. In Proceedings of the IEEE Conference on Computer
Vision and
Pattern Recognition, pp. 6720–6731. [2]
Bahri, Y., Kadmon, J.,
Pennington, J., Schoenholz, S. S., Sohl-Dickstein, J., & Ganguli, S. (2020). Statistical
mechanics of deep learning. Annual Review of Condensed Matter Physics. |
|
|
5/19 |
AI
Assisted Coding and Debugging (Ask AI to code your project!) [video][Slides (NA)] Project
review 2pm:
session #3 [on Axess] |
[1] Code search
work [link]. [2] Code
recommendation work [link]. [3] Automatic
bug fixing [link]. [4] Predictive test selection [link]. |
Eric
Meijer, |
|
5/21 |
Interpretable and robust AI:
Successive Subspace Learning (SSL), Forward-Only Feature Learning |
[1] C.-C. Jay Kuo,
“Understanding convolutional neural networks with a mathematical model,” the
Journal of Visual Communications and Image Representation, Vol. 41, pp.
406-413, 2016. [2] C.-C. Jay Kuo,
Min Zhang, Siyang Li, Jiali
Duan and Yueru Chen,
“Interpretable Convolutional Neural Networks via Feedforward Design,” the
Journal of Visual Communications and Image Representation, Vol. 60, pp.
346-359, 2019. [3] M. Zhang, H. You, P. Kadam, S.
Liu, C.-C. J. Kuo, “PointHop:
an explainable machine learning method for point cloud classification,”
arXiv:1907.12766 (2019), IEEE Trans. on Multimedia. [4] Yueru
Chen and C.-C. Jay Kuo, “PixelHop:
a successive subspace learning (SSL) method for object classification,”
arXiv:1909.08190 (2019), Visual Communications and Image Representation. |
Prof. C.-C Jay Kuo, USC |
|
5/26 |
Data Privacy and Regulations Compliance |
[1] GDPR, CCPA and HIPAA. [2] Soteria: A Provably Compliant
User Right Manager Using a Novel Two-Layer Blockchain Technology, Stanford
University, DeepQ, NTU, Edward Y. Chang el al.,
2020 [link]. |
|
|
5/28 |
AI Future, novel topics in Deep Learning, NLP, and Healthcare |
[1] Panel: VR/AR for Surgery and Medical Education, April 2018 [link]. [2] Stanford SNI Talk: Advancing Healthcare w/ AI and VR, Edward Y.
Chang [link]
(Stanford ID required). [3] From system 1 to system 2, keynote, Yoshua
Bengio, NeurIps 2019 [link]. [4] Daphne Koller: Biomedicine and Machine Learning, AI
Podcast #93 with Lex Fridman,
May 2020 [link]. |
|
|
6/2 |
An Uncertain Future: Coping with
COVID-19 anxiety with facts and right perspectives |
This pandemic is not going away
soon. It has impact health, economy,
and living tremendously, perhaps the most drastic one in our lifetime this
far. How should we deal with it
mentally? Dr. Duvvuri lectures on
COVID-19 and AI. |
Dr. Vikas Duvvuri,
MD/PhD Stanford |
|
6/4 |
Extended
Office Hours |
Schedule
a time slot by appointment between 9am and 3pm |
Organized
by TA |
|
6/9: 9am –
10:20am |
Project: final presentations Session #1 [video] |
[1]
Project #1, Learning Metastatic Features of Tumor Microenvironment [2] Project #2, Thoracic Disease
Detection using Transfer Learning and Meta Learning |
Final report due on 6/11 |
|
6/9: 2pm –
3:20pm |
Project: final presentations Session #2 [Included in the video
above] |
[1] Project #3, COVID-19 modeling
with regression [2] Project #4, Document
classification/summarization with small data |
Final report due on 6/11 |