Lectures (Fall 2022)
Module 1: Introduction to Medicine
Lecture 1: Introduction to the course
Slides: Optional Reading:"Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications" Sorlie et al, Proc Natl Acad Sci U S A. 2001; 98:10869-74
"Molecular Portraits of human breast tumors" Perou et al, Nature 2000, 406: 747-752
"Gene expression profiling predicts clinical outcome of breast cancer" van't Veer et al Nature. 2002; 415:530-6
"Gene expression profiling in breast cancer: classification, prognostication, and prediction" Reis-Filho and Pusztai, Lancet 2011; 378:1812-23.
Lecture 2: Medicine
Slides: Optional Reading:“Preparing for precision medicine”, Mirnezami, R, et al., N Engl J Med 2012, 366 (6), 489-91
“Personalized medicine: been there, done that, always needs work!”, Murray, Am J Respir Crit Care Med 2012, 185 (12), 1251-2
“Toward precision medicine and health: Opportunities and challenges in allergic diseases”, Galli, J Allergy Clin Immunol 2016, 137 (5), 1289-300
"Personalized Medicine at FDA: The Scope & Significance of Progress in 2021", Personalized Medicine Coalition
Lecture 3: The healthcare industry
Slides: Optional Reading:U.S. Health Care from a Global Perspective, 2019: Higher Spending, Worse Outcomes?
“Comparing Health Outcomes of Privileged US Citizens With Those of Average Residents of Other Developed Countries”, Emanuel et al, JAMA Intern Med. 2021, 181 (3), 339–344
“Predicting the Future - Big Data, Machine Learning, and Clinical Medicine”, Obermeyer et al, N Engl J Med 2016, 375 (13), 1216-9
Lecture 4: Evidence Based Medicine
Slides:
Required Reading:
“Effect of Renin-Angiotensin-Aldosterone System Inhibitors in Patients with COVID-19: a Systematic Review and Meta-analysis of 28,872 Patients”, Baral et al, Curr. Atherosclerosis Reports (2020) 22:61
Optional Reading:
Evidence-based practice for individuals or groups: let’s make a difference, de Groot et al Perspect Med Educ (2013) 2:216–221
Osteoporosis therapy: an example of putting evidence-based medicine into clinical practice, Hosking, Geusens, Rizzoli; QJM 98(6), pp. 403–413,
"Why Most Published Research Findings Are False", Ioannidis, 2005, DOI: 10.1371/journal.pmed.0020124
“Healthcare outcomes assessed with observational study designs compared with those assessed in randomized trials”, Anglemyer et al, Cochrane Database Syst Rev 2014, (4), MR000034
"Clinical Versus Mechanical Prediction: A Meta Analysis" Grove et al Psychological Assessment, 2000 12(1) 19-30
A Randomized Trial of Hydroxychloroquine as Postexposure Prophylaxis for Covid-19, Boulware et al, N Engl J Med. 2020 doi: 10.1056/NEJMoa2016638
Module 2: Disease phenotyping
Lecture 5: Asthma Phenotyping
Slides:
Required Reading:
An overview paper: “A guide to machine learning for biologists”, Greener et al, Nat Rev Mol Cell Biol 2022, 23 (1), 40-55
“Unsupervised phenotyping of Severe Asthma Research Program participants using expanded lung data.” Wu et al, J Allergy Clin Immunol. 133(5):1280-8
Optional Reading:
"Asthma phenotypes: the evolution from clinical to molecular approaches." Wenzel, Nat Med. 2012;18(5):716-25.
Lecture 6: Comorbidity Phenotyping
Slides:
Required Reading:
"Comorbidity Clusters in Autism Spectrum Disorders: An Electronic Health Record Time-Series Analysis." Doshi-Velez et al Pediatrics. 2014, 133(1):e54-63
Lectures 7-8: Drug Response Phenotyping
Slides:
pdf (Lecture 7: Background - Multiple kernel clustering)
pdf (Lecture 8: Corticosteroid response phenotyping in asthma)
Required Reading:
“Multiview Cluster Analysis Identifies Variable Corticosteroid Response Phenotypes in Severe Asthma”, Wu et al, Am J Respir Crit Care Med 2019, 199 (11), 1358-1367
Optional Reading:
“Predicting Response to Triamcinolone in Severe Asthma by Machine Learning. Solving the Enigma”, Chung, Am J Respir Crit Care Med 2019, 199 (11), 1299-1300
“The prevention and treatment of missing data in clinical trials”, Little et al, N Engl J Med 2012, 367 (14), 1355-60
Lecture 9: Phenotyping via Tensor Factorization
Slides:
Required Reading:
"Rubik: Knowledge guided tensor factorization and completion for health data analytics", Wang et al. KDD 2015, pp. 1265–1274
Module 3: Biomarker discovery
Lecture 10: Feature Selection, and Gene Set Enrichment Analysis
Slides:
Required Reading:
Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles, Subramanian et al PNAS, 2005, 102 (43) 15545-15550
Lecture 11: Feature Selection for microarray data
Slides:
Required Reading:
Xing, Jordan, Karp, Feature selection for high-dimensional genomic microarray data. In Proceedings of the Eighteenth International Conference on Machine Learning (ICML), 2001.
Optional Reading:
Xing et al, Distance Metric Learning, with application to Clustering with side-information. In Advances in Neural Information Processing Systems 16 (NeurIPS) MIT Press: 2002; pp 521-528.
Lecture 12: Biomarker discovery for indolent prostate cancer
Slides:
Required Reading:
"A Molecular Signature Predictive of Indolent Prostate Cancer", Irshad et al Sci Transl Med. 2013; 5(202): 202ra122
Lectures 13-14: Finding Genomic Markers
Slides:
pdf (Lecture 13: Background - DNA sequencing technologies, genomic variations)
pdf (Lecture 14: Finding genomic biomarkers for colorectal cancer)
Required Reading:
"Identification of novel mutations by exome sequencing in African American colorectal cancer patients", Ashktorab et al Cancer 2014, 121(1):34-42
Optional Reading:
“Hallmarks of cancer: the next generation”, Hanahan et al, Cell 2011, 144 (5), 646-74
“Why don't we get more cancer? A proposed role of the microenvironment in restraining cancer progression”, Bissell et al, Nature Medicine 2011, 17 (3), 320-9
Lectures 15-16: Finding Phylogenetic Markers
Slides:
pdf (Lecture 15: Background - Single cell analysis, phylogenetics)
pdf (Lecture 16: Finding phylogenetic markers for prostate cancer)
Required Reading:
"Single-cell genetic analysis reveals insights into clonal development of prostate cancers and indicates loss of PTEN as a marker of poor prognosis", Heselmeyer-Haddad et al Am J Pathol. 2014;184(10):2671-86
Module 4: Predictive modeling
Lecture 17: Detecting high-risk surgical patients
Slides:
Required Reading:
“Development and validation of machine learning models to identify high-risk surgical patients using automatically curated electronic health record data (Pythia): A retrospective, single-site study”, Corey et al, PLoS Med 2018, 15 (11), e1002701
Optional Reading:
"The elements of statistical learning: data mining, inference, and prediction." 2nd ed. Hastie, T., Tibshirani, R., & Friedman, J. H. (2009). New York, Springer. -- Chapters 4.4 (LR), 10 (boosting), 15 (RF).
Lecture 18: Detecting Autism from video
Slides:
Required Reading:
“Mobile detection of autism through machine learning on home video: A development and prospective validation study”, Tariq et al, PLoS Med 2018, 15 (11), e1002705
Optional Reading:
"The elements of statistical learning: data mining, inference, and prediction." 2nd ed. Hastie, T., Tibshirani, R., & Friedman, J. H. (2009). New York, Springer. -- Chapter 12 (SVM).
Lectures 19-20: Detecting onset of heart failure via deep learning
Slides:
pdf (Lecture 19: Background - modeling time-varying processes)
pdf (Lecture 20: Background - deep neural networks, recurrent neural networks; Paper - using RNN for early detection of heart failure onset)
Required Reading:
"Using recurrent neural network models for early detection of heart failure onset", Choi et al J Am Med Inform Assoc. 2017; 24(2): 361–370
Optional Reading:
"Deep Learning", Goodfellow, Bengio & Courville (2016), MIT Press. -- Chapter 10.2 (RNN), 10.7, 10.10 (LSTM, GRU)
Module 5: Medical imaging analysis
Lecture 21: Predicting cancer survival using deep learning
Slides:
Required Reading:
"Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study", Kathner, et al PLOS Medicine, 2019
Optional Reading:
“Deep learning”, LeCun, Bengio & Hinton, Nature 2015, 521 (7553), 436-44
"Deep Learning", Goodfellow, Bengio & Courville (2016), MIT Press. -- Chapter 9 (CNN)
Lecture 22: Classification of skin cancer using deep learning
Slides:
Required Reading:
“Dermatologist-level classification of skin cancer with deep neural networks”, Esteva et al, Nature 2017, 542 (7639), 115-118
“Adversarial attacks on medical machine learning”, Finlayson et al, Science 2019, 363 (6433), 1287-1289
Optional Reading:
“Validation of artificial intelligence prediction models for skin cancer diagnosis using dermoscopy images: the 2019 International Skin Imaging Collaboration Grand Challenge”, Combalia et al, Lancet Digit Health 2022, 4 (5), e330-e339
Module 6: Drug design and development
Lecture 23: Designing Therapeutic Proteins
Slides:
Required Reading:
"Optimization algorithms for functional deimmunization of therapeutic proteins", Parker et al, BMC Bioinformatics 2010, 11, 180
Optional Reading:
"Mapping the Pareto Optimal Design Space for a Functionally Deimmunized Biotherapeutic Candidate", Salvat et al, PLoS Comput Biol 2015, 11 (1), e1003988
Lecture 24: Predicting Drug Response; Interpretable AI
Slides:
Required Reading:
"Predicting Drug Response and Synergy Using a Deep Learning Model of Human Cancer Cells", Kuenzi et al, Cancer Cell 2020, 38 (5), 672-684.e6
Module 7: Advanced Topics
Lecture 25: Algorithmic Bias in AI and Machine Learning; Course Summary
Slides:
Required Reading:
"Dissecting racial bias in an algorithm used to manage the health of populations", Obermeyer et al, Science 2019, 366 (6464), 447-453