Biomedical Informatics

Required fall of your 1st year, the course provides an overview of the field of biomedical informatics, combining perspectives from medicine, computer science, and social science. Use of computers and information in health care and the biomedical sciences, covering specific applications and general methods, current issues, capabilities and limitations of biomedical informatics. Biomedical Informatics studies the organization of medical information, the effective management of information using computer technology, and the impact of such technology on medical research, education, and patient care. The field explores techniques for assessing current information practices, determining the information needs of health care providers and patients, developing interventions using computer technology, and evaluating the impact of those intervention. Course Director: Gamze Gursoy.

Survey of the machine learning underlying the field of medical informatics. Explores techniques in mathematics, logic, decision science, computer science, engineering, cognitive science, management science and epidemiology, and demonstrates the application to health care and biomedicine. Course Director: Matthew McDermott.

Survey of foundational symbolic AI for modeling health information systems and for making those models explicit and sharable. The topics cover clinical terminologies (e.g., ICD-9, SNOMED-CT, MeSH, UMLS), biomedical ontologies (e.g., GO, Disease Ontology, PharmGKB), knowledge representation, computerized practice guidelines, semantic interoperability, and text processing. Course Director: Chunhua Weng.

Provides an overview of research methods relevant to biomedical informatics. The overall goal of the course is to prepare the student to participate in and perform scientific research. Competencies of the course include learning to design a study of a biomedical informatics resource; perform quantitative and qualitative analysis relating to a biomedical informatics resource; and write a biomedical informatics-related research proposal. By the end of the course, all trainees must be able to write a biomedical informatics-related research summary and complete certification in responsible conduct of research. Course Director: Lena Mamykina.

Accordion content...Biological Sequence Analysis introduces the basics of sequential, structural, and functional genomics. The course is both a lecture and lab course, in which students learn the basic bioinformatic principles and apply these principles through laboratory exercises. The course accommodates both students with a computational background with little previous biology, and students from a primarily biological background, with little previous computation. Topics include basic Unix, biological databases, sequence comparison, database searching, multiple sequence alignment, biological regular expressions, profile methods (including hidden Markov models), protein and RNA structure prediction, mapping, primer design, genomic analysis, molecular phylogetics, and functional genomics including microarray analysis and pathway analysis. Course Director: Richard Friedman.

Technology for obtaining DNA sequences have been consistently improving faster than Moore's law. This has opened a wealth of computational challenges in weaving the heaps of straw of DNA sequence data into gold of biological insight. The class serves as an introduction to computational genomics, explaining the basic challenges and teaching the general computer-science tools to tackle them. This course is intended to introduce students of both computational and bio-medical skill sets to current quantitative understanding of genomics and prepare them to computational research or industrial development in the field. Course Director: Itsik Pe’er.

This course offers an introduction to the practice and culture of medicine for informaticians using a mix of lectures, case-based discussions, and critical review of scientific journal articles. The goal is to “acculturate” students without clinical backgrounds to the practice of medicine to inform the study and design of clinical information systems. Each class session will be structured to touch upon items from one or more of 3 key competency areas: biomedical science, clinical workflow and culture, and clinical informatics. Students will learn about medical language and terminology; basic anatomy and physiology; introductory pathology and pathophysiology; the process of medical decision-making; patient safety, medication safety, and health IT; telemedicine; artificial intelligence; and the flow of information in the practice of medicine.

This course delves into the intersection of epidemiology and computational methods, equipping students with the tools to conduct rigorous epidemiological studies from big clinical data repositories. Students will explore techniques from informatics, computer science, machine learning, and statistics to clean, analyze, and interpret data from electronic health records (EHRs) and other large-scale datasets. Through hands-on projects and case students, students will gain practical experience in applying epidemiologic study designs to uncover patterns, identify risk factors, model disease transmission dynamics, and evaluate interventions. This interdisciplinary approach prepares students to address real-world public health challenges by leveraging the power of data-driven insights. The course is broken up into modules, each of which covers an epidemiologic study design or principle. Modules will range from 1-3 classes, and will include (i) a lecture and (ii) accompanying lab work. Students are expected to read technical texts carefully, participate actively in lecture discussion, and develop hands-on skills in labs involving real-world biomedical and health datasets. Students will curate their own analytic datasets from Observational Health Data Sciences and Informatics (OHDSI)-formatted electronic health record (SynPUF) data. 

In this course, students will learn about complexities that make health and medicine data unique and how it opens up opportunities for advanced AI. Students will learn advanced Machine Learning methods useful in health and medicine applications, for example, time-series modeling, reinforcement learning, probabilistic modeling, causal inference, foundation models, unsupervised learning, and self-supervised learning. The course will provide an overview of challenges such as fairness, interpretability, generalization, robustness, safety, and policy implications of ML in health and medicine. The course will train students to map real-world challenges of working with health and medical data to statistical challenges that require new and advanced ML methods.

In August, new trainees receive orientation on policies and procedures and meet individually with the graduate program manager for personalized advising. They are given a tour of the department where most department courses are taught. Students are assigned academic advisors from core faculty and meet with them at least once a term for assistance with ongoing course selection and support over the course of their PhD. Incoming students also meet individually with the graduate program director for an orientation to our program. The Office of Graduate Affairs (OGA) organizes activities for all trainees in the VIBRE PhD Programs at the end of August to acclimate the students to Columbia and New York City. Students also receive support from elected student representatives. To orient students to the field of biomedical informatics, they are registered to attend the annual American Medical Informatics Association (AMIA) symposium held in the fall.

PhD students are required to rotate among faculty research labs during their first year, and select a permanent research advisor by end of their first year. Students are allowed to do two research rotations, and an occasional third, if approved by the graduate program director and OGA.

Core Courses

1. BINF G4001 is required in the fall term of the first year. The other core courses (BINF G4002,  BINF G4003, BINF G6002,  BINF G4013, CBMF W4761) can be taken in any order, but should be completed by winter intercession of the 2nd year. 
2. Mandatory Ethics Course: CMBS G4010 Responsible Conduct of Research & Related Policy Issues, to be taken in the spring term, 1st year
3. Electives: five elective courses to be selected from the following
   1. Domain (specialization courses) - 2 courses from areas of specialization relevant to BMI, such as data science, clinical informatics, translational informatics, bioinformatics, and public health informatics
   2. Educational Objectives Courses - 3 courses selected within the specified educational objectives (Qualitative, Quantitative, Information Technology). The specific configuration of the Educational Objective Courses should be tailored to each student’ research needs. 
4. BINF G4099 Research Seminar - every fall and spring semester
5. BINF G6001 Projects-Biomedical Informatics or BINF G9001 Doctoral Research. BINF G6001 is for those students who have not yet achieved PhD Candidacy (MPhil degree conferral)
6. BINF G9999 Doctoral Research - during the last term. 
7. Provide teaching assistance for two courses anytime between the 2nd year-and final year (BINF G8010 MPhil Teaching Experience)

The Breadth Exam is generally administered during the winter intersection of their second year. The Breadth Exam is designed to assess the student’s breadth of knowledge of the field, the ability to express this knowledge in writing and verbally, reason with it, and to synthesize concepts from different areas. The Breadth exam includes a written portion and a follow-up oral examination. In the written part, students answer a set of scenario-based multi-component questions, which require them to integrate their knowledge from different courses and areas, reason over them, and present it in an informative manner. To ensure consistency, the exam is graded and carried out by members of the training committee for all students. Students are eligible to apply for MA degree conferral after successful passage of the Breadth exam. 

The Depth Exam is usually taken 12 months after the breadth exam, and no later than six months before the dissertation proposal. The goal of this exam is to assess the ability of the trainee to survey the literature in their area of research, synthesize available knowledge, identify gaps, and propose research questions to address these gaps. Before an exam is scheduled, the thesis committee reviews the student’s work to date and assess the student’s readiness to undertake the exam. The depth exam is also an opportunity for trainees to present in-depth research topics to a general informatics audience during the public session of the exam. After accruing 6 residence units, students are eligible to apply for MPhil degree conferral following successful passage of the Depth Exam.

Pre-doctoral students have a thesis advisory committee that consists of the primary faculty advisor, and two additional faculty members, at least one of whom must be a member of the core DBMI faculty. Timing and frequency of these meetings are at the discretion of a student's advisory committee with the approval from the Training Committee.

The department organizes an annual one-day retreat for DBMI faculty, students, and senior staff. This is held at the beginning of the new academic year so that incoming trainees can get to know others, and to learn about our department before the onslaught of course work is upon them. Current students have an opportunity to present their work in poster sessions. In addition, several selected former trainees in biomedical informatics serve on a panel called “Life after DBMI,” where they discuss their experience in DBMI, their career and research paths, and provide advice to students.