IBI fosters informatics education across Penn’s campus with degree programs, workshops, and a monthly biomedical informatics podcast. See below for details about these initiatives.
BMIN 520: Fundamentals of Artificial Intelligence
Course Director: Ryan Urbanowicz, PhD
As a subfield of computer science, artificial intelligence is often used interchangeably with the term ‘machine learning’, which itself is more accurately a subfield of AI dealing with the broader concept of inductive reasoning. However, a wealth of key prerequisite topics that focus on deductive reasoning align with the bulk of biomedical informatics applications being actively utilized today. These founding principles of AI and their intersection with biomedical informatics are the focus of this first course on artificial intelligence. The course is divided into modules that cover (1) introductory/background materials, (2) logic, (3) other knowledge representation, (4) essentials of expert systems, (5) search, and (6) advanced/auxiliary topics. These topics offer a global foundation for branches of AI application and research, including concepts that will later support a deeper understanding of inductive reasoning and machine learning. In a practical sense, this course focuses on how biomedical data can be organized, represented, interpreted, searched, and applied in order to derive knowledge, make decisions, and ultimately make predictions.
BMIN 521: Machine Learning
Course Director: Li Shen, PhD
Machine learning studies how computers learn from data and has enormous potential to impact biomedical research and applications. This course will cover fundamental topics in machine learning with an application focus on biomedical informatics. Specifically, the course will cover
- Supervised learning methods such as linear regression, logistic regression, nearest neighbors, support vector machines, decision trees and random forests
- Unsupervised learning topics such as clustering, dimensionality reduction and association rules
- Neural networks and deep learning methods for supervised or unsupervised learning tasks
- The applications of these machine learning techniques to various biomedical informatics problems via analyzing imaging, biomarker, electronic health record, clinical and/or other biomedical data.
The precise topics may vary from year to year based on student interest and developments in the field.
BMIN 522: Natural Language Processing
Course Director: Graciela Gonzalez Hernandez, PhD
The growing volume of unstructured health-related data presents unparalleled challenges and opportunities for informaticians that seek to mine the rich information “hidden” in plain sight – clinical records, social media, published literature, all sources designed for human eyes, but not necessarily for automatic processing. In this class, we will survey the most recent natural language processing methods used for information extraction, taking a “hands on” approach at how they are currently applied in the biomedical domain. Emphasis will be placed on lexical and syntactic methods, covering different approaches to classifcation and extraction for content discovery – including deep learning and unsupervised approaches.
- To provide students with a comprehensive view of natural language processing methods at the word (lexical) and sentence (syntactic)
levels, applicable to information extraction
- To introduce the fundamental concepts and methods used for contextual (semantic) analysis of biomedical text that aid information
- To prepare students for the development and application of information extraction methods in the realm of biomedical text mining
- To provide students with an optimal environment for guided research in order to seek innovative solutions to real biomedical NLP
problems in the collegiality of the classroom.
The Master of Biomedical Informatics (MBMI) program seeks to provide state-of-the-art graduate-level educational and training opportunities in biomedical informatics (BMI), adhering to the best practices as established by national competency standards, to create the next generation of biomedical informatics and practitioners.
Our program’s objectives are drawn from the key competencies identified by the American Medical Informatics Association’s 2017 guidelines, which identify the skills and knowledge that informatics practitioners need to set themselves apart in a rapidly developing field. By graduation, students should be able to:
- Identify the applicable information science and technology concepts, methods, and tools, which may be dependent upon the application area of the training program, to solve health informatics problems.
- Identify and draw on the social, behavioral, legal, psychological, management, cognitive, and economic theories, methods, and models applicable to health informatics to design, implement, and evaluate health informatics solutions.
- Identify possible biomedical and health information science and technology methods and tools for solving a specific biomedical and health information problem. Design a solution to a biomedical or health information problem by applying computational and systems thinking, information science, and technology.
- Define and discuss the scope of practice and roles of different health professionals and stakeholders including patients, as well as the principles of team science and team dynamics to solve complex health and health information problems.
The IBI Certificate in Biomedical Informatics is a four-course sequence for non-informatics professionals designed to build the informatics community at Penn and to train informatics-literate clinicians and researchers who will have a broad understanding of the field of biomedical informatics.
This certificate is unique among Philadelphia-area programs in that its curriculum covers general biomedical informatics, clinical informatics, and clinical research informatics.
Students in the Certificate in Biomedical Informatics Program can expect to obtain a working knowledge of biomedical informatics, its history, the current landscape, and future directions of the field. The four courses in this program also form the core of the Master of Biomedical Informatics (MBMI) program, so certificate students can expect to interact with a variety of students with diverse interests in informatics.
Certificate students who wish to expand their biomedical informatics skills in a degree program may apply to the MBMI program and, if accepted, may transfer credit from the certificate program to meet the requirements of the Master’s program. Contact Meg Tanjutco for details about this process.
The Biomedical Informatics Roundtable Podcast hosted by Drs. Jason H. Moore and Marylyn D. Ritchie. Our goal is to bring you discussion of hot topics, recent papers, news, conferences, open data, open-source software, and advice for trainees as well as interviews and spotlights with our biomedical informatics colleagues from around the world.
Held annually, Informatics Day brings together faculty, staff, and students from across the university for a full day of knowledge sharing.
Informatics Day 2020 will be held virtually. Details are coming soon.
Natural Language Processing Working Group
Faculty Organizer: Graciela Gonzalez-Hernandez, PhD
With interest in natural language processing research and applications evident in all areas of clinical and translational research, we organize the NLP for Health (HLP) Workgroup Seminar on a monthly basis, where we hope to network and strengthen each other’s efforts, learn about our work, and find new collaborations around our common interest in HLP.
Ontologies Working Group
Faculty Organizer: Chris Stoeckert, PhD
The Ontology Affinity Group provides a monthly meet-up for interested faculty, staff, and students at Penn and neighboring institutions to discuss the development and application of biomedical ontologies. Presentations include description of projects involving ontologies and discussion of papers or topics of interest on ontologies.
Statistical Issues in EHRs Working Group
Faculty Organizer: Jinbo Chen, PhD
This workgroup focuses on the many statistical issues that arise when working with EHR data. These include missing data, quality control, biases, study design, privacy, and security. The group discusses statistical solutions to these problems and meets regularly to foster collaborations.
AI in Healthcare Student Working Group
Faculty Organizer: Jason Moore
Student Organizer: Christy Hong
This workgroup is led by a group of graduate and medical students from across Penn Medicine, Wharton, and SEAS who are interested in AI and its impact in healthcare. The students regularly meet with Penn Medicine faculty to identify ways in which they can promote AI for the good of patients. An example project they are working on is the release of a large de-identified data set that can be used for a machine learning analysis competition.