CS372 Artificial Intelligence for Disease Diagnosis a
CS372 Artificial Intelligence for Disease Diagnosis
and
Information Recommendations
Spring 2021 T Th 8:30am – 9:50am (Zoom), from March 30th
to June 3rd
3 units/20 sessions,
including lectures and project discussion/presentation.
Online zoom
link, registration required
Edward Chang
(echang@cs.stanford.edu)
Adjunct
Professor, Computer Science, Stanford University
AI/NLP
Advisor, SmartNews
President, DeepQ Healthcare (2012-2021)
Director of Research, Google (2006-2012)
TA: Saahil Jain
(saahil.jain@stanford.edu)
Announcements
§ (5/27) Lecture #18 slides and video online (only lecture made public).
§ (5/25) Lecture #17 slides and video online.
§ (5/18) Lecture #15 slides and video online
§ (5/13) Lecture #14 slides and video online.
§ (5/8) Lecture #12 slides and video online.
§ (5/4) Lecture #11 slides and video online.
§ (4/29) Lecture #10 slides and video online.
§ (4/22) Lecture #8 slides and video online. Project proposal due @11:59pm today.
§ (4/20) Lecture #7 by Rich online.
§ (4/15) Lecture #6 slides and video online. Project proposal due in a week on 4/22.
§ (4/13) Lecture #5 slides and video online.
§ (4/8) Lecture #4 slides and video online.
§ (4/6) Lecture #3 slides and video online. Please join Lecture #4 in real time to
brainstorm project topics.
§ (4/5) Assignment #1 posted, due at 10am on April 13th.
§ (4/1) Lecture #2 slides and video online (video can be
assessed only by Stanford’s students.)
§ (3/30) Lecture #1 slides and video online (video can be
assessed only by Stanford’s students.)
§ (3/29) Three guest speakers Dr. Melissa Ko (Cancer
Biologist, Stanford), Dr. Vikas Duvvuri (PHD/MD, Stanford), and Rich
Jaroslovsky (CJO of SmartNews) will cover subjects of cancer,
depression/anxiety disorder, and information polarization.
§ (3/28) Received several audit requests. Audit is not permitted as we would like all
to involve in hands-on problem solving.
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, privacy-preserving data management, and other
topics. (The 2021 edition
plans to cover quantum ML models.)
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. The information recommendation part of this
course in 2021 will address the problem of global
political polarization. Students will be assigned to work on a term
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.
Students are welcome to formulate a project that leverages their own
graduate research.
Level:
Senior/graduate of EE/CS, Medicine, Medical Informatics, Engineering;
proficient in programming
Perquisite:
Introductory course in AI, Statistics, or Machine Learning
Time and
Location
·
Time: Tuesdays and Thursdays
8:30am to 9:50am, starting on March 30th.
·
Location: On Zoom (Link is
provided at Stanford AXESS. Enrolled students
only.)
Assignments
·
No exams.
·
Three written assignments in the
first half of the quarter.
·
Reading: A list of papers/articles/podcasts grouped by
topics. Each student selects one topic
of interest to conduct in-depth surveys and presentation.
·
Term project: A group project of two to three
involving algorithm design and coding.
Some past example projects are:
§
COVID-19 demographic analysis
§
COVID-19 outbreak tracking and prediction
§
Thoracic disease diagnosis with limited X-ray
images
§
News article summarization (extractive and
abstractive)
§
Methods for fusing knowledge and perception
(text and image)
Grading
·
Assignments 30%
·
Literature survey and project
proposal 20%
·
Project implementation and demo
50%
Syllabus
|
Date |
Description |
Course
Materials |
Notes |
|
Week #1 3/30/2021 |
Course Aims and Syllabus |
Deep Learning
Review [link], Afshine Amidi and Shervine Amidi (Stanford
CS230) |
Prof. Edward Chang |
|
4/1/2021 |
P4 Medicine Part I: Healthcare Paradigm Shift |
[1] Edward Y.
Chang, et al., "Artificial Intelligence in XPRIZE DeepQ Tricorder.
" ACM MM Workshop for Personal Health and Health Care, 2017. [2] FC Chang, JJ
Chang, CN Chou, EY Chang, “Toward Fusing Domain Knowledge with Generative
Adversarial Networks to Improve Supervised Learning for Medical Diagnoses”,
FC Chang, JJ Chang, CN Chou, EY Chang, IEEE MIPR Conference, 2019. [3] Edward Y.
Chang, et al., "Context-Aware Symptom Checking for Disease Diagnosis
Using Hierarchical Reinforcement Learning." AAAI, 2018. [4] Transfer
representation learning for medical image analysis, Chuen-Kai
Shie, Chung-Hisang
Chuang, Chun-Nan Chou, Meng-Hsi Wu, Edward Y
Chang, IEEE EMBC, 2015. |
Prof. Edward Chang |
|
Week #2 4/6 |
P4 Medicine Part II: w/ small data and Inductive
Bias |
[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]. [3] Universal
Equivariant Multilayer Perceptrons, Siamak Ravanbakhsh, ICML,
2020. [4] The problem of
Protein Folding,
Wikipedia. [5] On Bottleneck of
Graph Neural Networks and its Practical Implications, Uri Alon, Eran Yahav, ICRL 2021. [6] SIGN: Scalable
Inception Graph Neural Networks, Fabrizio Frasca, Emanuele
Rossi, Davide Eynard, Ben
Chamberlain, Michael Bronstein, Federico Monti, GraphSaint, 2020. [7] Panel: VR/AR for Surgery and Medical Education,
April 2018 [link]. [8] Stanford SNI Talk: Advancing Healthcare w/ AI
and VR, Edward Y. Chang [link] (Stanford ID
required). [9] Daphne
Koller: Biomedicine and Machine
Learning, AI Podcast #93 with Lex Fridman,
May 2020 [link] |
Prof. Edward Chang Assignment
#1: Deadline
4/13 |
|
4/8 |
NLP Part I: Hand-engineering, Similarity,
Collaborative Filtering, Latent Dirichlet Allocation [Video][Slides] Brainstorming on term
projects, datasets and resources |
[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]. [4] Perceiver:
General Perception with Iterative Attention [link]. |
Prof. Edward Chang |
|
Week #3 4/13 |
NLP Part II: Attention, Transformers, BERT,
and GPT [Video][Slides] |
[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]. [4] Perceiver:
General Perception with Iterative Attention [link]. |
Prof. Edward Chang Project
Proposal Deadline
4/22 |
|
4/15 |
NLP Part III: CLI: Conversational Language
Interface, Dialogue & Virtual Assistant (Alexa, Google Home, & Oval),
and Conversational Language Interface |
[1] GPT3 and
multimodal pre-trained models [2] OVAL [3] Chirpy Cardinal [4] Genie |
Prof. Edward Chang |
|
Week #4 4/20 |
Information Recommendations:
The Crisis of Global News and Information Polarization Project signed up, project
group formed, and project abstract due |
[1] Can a Divided
America Heal? Ted Talk [link], Jonathan Haidt, 2016 [2] … more will be
provided by Rich. Each abstract includes
1) key objective, 2) related work, 3) proposed
methods, and 4) data sources. |
Rich
Jaroslovsky, VP/Chief Journalist, SmartNews |
|
4/20 |
Proposal Discussion (by appointment) Session #1 10am – 11:30am Session #2 2pm – 3:30pm |
Each group can sign
up for a 30-min slot. Please sign up
@ Slack. |
Saahil Jain |
|
4/22 |
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 Project
Proposal due @
11:59pm |
|
Week #5 4/27 |
Project Proposal Presentation,
title, goals, methods, and datasets |
[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. [3] Perceiver:
General Perception with Iterative Attention [link]. [4] Stanford OVAL COVID-19
Appointment & QA app [link] |
Prof. Edward Chang |
|
4/29 |
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 |
|
Week #6 5/4 |
Intell. & Consciousness Part I: History of AI in Diagnosis and
its Promises and Limitations |
[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. |
Prof. Edward Chang Assignment
#3 Deadline
5/18 |
|
5/6 |
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 |
|
Week #7 5/11 |
Project Check-in: methods,
datasets, experiments, and preliminary results |
|
Prof. Edward Chang |
|
5/13 |
Intell. & Consciousness Part II: What and Where is
Consciousness? Rule of Nature, Principles of Similarity
and Classification. |
[1] What is Life,
Schrodinger, Cambridge University Press, 1992 [2] Discovery of
Discovery of a Perceptual Distance Function for Measuring Image Similarity Beitao Li, Edward Chang, and Yi Wu, Multim. Syst. 8(6): 512-522 (2003) [link] |
Prof. Edward Chang |
|
Week #8 5/18 |
An Uncertain Future: Coping
with COVID-19 anxiety with facts and right perspectives [video][Slides] |
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 |
|
5/20 |
AI in Healthcare Panel [video] |
Open discussion on
various topics including AI, healthcare, and career advice. |
Moderator: Saahil
Jain Panelists: Vikas Duvvuri,
Pranav Rajpurkar, Edward Chang |
|
Week #9 5/25 |
Data Privacy and Regulations
Compliance |
[1] GDPR, CCPA and
HIPAA. [2] Differential
Privacy [3] 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]. |
Prof. Edward Chang |
|
5/27 |
Intell. & Consciousness Part III: AI Future: Consciousness, Mind,
and Ethics [video] [slides] |
1] What is Life, Erwin Schrödinger, 1944 [2] The Emperor’s New Mind, Roger Penrose, 2016
edition [3] Divine Comedy, Dante Alighieri, 1320 |
Prof. Edward Chang |
|
6/1: 8:30am – 10:20am |
Project: final presentations Session #1 |
|
Final report due on 6/4 |
|
6/3: 8:30am – 10:20pm |
Project: final presentations Session #2 |
Final report due on 6/4 |
|
|
|
|||