Description: This course teaches novel and recent advancements in real-world evaluation of genetic, biological and clinical technologies by their integration with deep learning, bioinformatics and statistical systems to diagnose and cure chronic, acute and inherited diseases. Learning modules are structured with domain knowledge and current status of emerging technologies from textbooks and academic publications from our research group and peer-reviewed scientific literature. Tutorials , assignments and final project will provide hands on experience for developing and validating deep learning algorithms and statistical testing with genetic, medical and clinical data. Current status of real-world evaluations using clinical trials, and strategies for regulation and derisking of emerging technologies and their positive and negative impact on patients, physicians and providers will be one of the key learning outcomes of the course.
Goals and learning objectives:
- Students/participants will learn deep learning, bioinformatics, statistical and machine learning methods for analyzing and generating diverse observational and real world biomedical datasets.
- Hands-on experience in developing experimental strategies for prospective clinical validation and regulatory approvals for adoption of emerging technologies in healthcare systems.
- Special emphasis on understanding and addressing causality, bias, fairness and interpretability of digital algorithmic evidence for positive societal impact
Start date: September 1st, 2020
End date: December 15th, 2020
Office hours: By appointment
TA: Sam Ghosal, firstname.lastname@example.org
- Consent, explainability, risk, regulation, computational social science, bias, fairness of emerging technologies
- Genetic engineering, CRISPR-Cas9 gene editing, CAR T-cells immunotherapy
- Single-cell profiling, high throughput DNA, RNA, OMICS, Genome Wide Association Studies
- Medical and microscopic cellular images, Electronic Health Records, connected technologies for remote monitoring
- Deep learning, machine learning and applied mathematics
- Biostatistics, causal inference, observational data, interventions
- Digital clinical trials, computational medicine, real world data and evidence, personalized medicine
- Biological: Recombinant DNA mutation & complementation, cellular pathways, next-generation sequencing, mass spectrometry for proteomics, high performance liquid chromatography for metabolomics, sequence alignments, BLAST searches, KEGG pathway analyses, homology mapping
- Computational: Convolutional neural networks, Bayesian inference, deep learning, Hidden Markov Models, auto-encoders, recurrent neural networks, Reinforcement learning, Markov Decision Process, image processing, bioinformatics, clustering and classification, k-means, R, Perl
- Clinical: Human subject studies, biomarker assays and interpretations, surrogate endpoints, interventions and treatments, adverse events, randomized control trials
- Statistical: p-value, students’ t-test, Pearson’s correlation coefficient, Principal Component Analysis, causal inference, Mann-Whitney U Test
Speakers: Opportunity to engage with invited speakers and professional at the forefront of funding agencies, research and product development, government agencies (NIH, NSF, USPTO etc.), technology (IBM, Google, Apple etc.) and healthcare companies (biotech, medical devices, software) and startups and non-profit foundations.
Prerequisites: Students with prior knowledge and interests in biological or clinical research, data science, engineering and computational medicine and social science are welcome to enroll. Critical analysis of research publications and periodicals are a plus.