Jump to topic
The goal of the strategic plan focuses on developing and applying biomedical artificial intelligence, neural engineering, machine learning and informatics capabilities to make rapid advances in biomedical research, education, health outcomes and value-based patient care.
Goal 2 will facilitate collaboration with researchers across the University, as well as clinicians from Penn State Health, to build unique tools and data sets that can answer complex questions needed to advance biomedical research and health outcomes. Penn State has tremendous resources in artificial intelligence (AI) and computational sciences that can be applied to health and health care in exciting and groundbreaking ways. To do so, we will create and implement a biomedical informatics infrastructure (algorithms, faculty/staff, computation power, custom tools and partnerships) to address the complex research problems present in individualized patient care and populations of care. Long-term goals include eliminating health disparities and transforming access to care by leveraging Penn State-wide strengths in digital innovation, computational and data sciences, population health, translational research and clinical care networks.
Artificial intelligence (AI) and bioinformatics (BI) approaches are playing an ever-important role in biomedical research. In parallel, big datasets have also become an integral part of many areas of research. It is more urgent than ever to equip biomedical researchers with AI and informatics expertise to better design the experiment, more effectively analyze the generated datasets, and enhance the rigor and reproducibility of research.
In response to these new challenges, Penn State College of Medicine included as part of its strategic plan a goal that focuses on AI and informatics. One critical mission for the strategic plan is to democratize AI. This workshop is a key step toward that mission. These two half-day workshops will introduce basic concepts of AI and machine learning, provide hands-on sessions, and illustrate areas of applications. It is also an opportunity to learn about the highlights of Penn State College of Medicine AI research projects in structural biology, multi-omics, electronic health records and imaging analysis.
Wednesday October 25, 2023
1–1:15 p.m.: Welcome
1:15–1:45 p.m.: Introduction to basic machine learning and deep learning
1:45–2:30 p.m.: Introduction to neural networks and Convolutional neural networks
2:45–3:30 p.m.: Advanced deep learning models – autoencoder, VAE, GAN
3:30–4:15 p.m.: Classical machine learning (Bias variance tradeoff/training/validation)
4:30–5 p.m.: Short seminars on focused areas
- Imaging application: Matthew Swulius
Thursday, October 26, 2023
1–1:30 p.m.: Hands on I: basic programming
1:30–2:15p.m.: Hands on II: deep learning. Example CNN, transfer learning (Sen)
2:30–3 p.m.: All of Us pipelines and applications
3:15–4:45 p.m.: Focused areas and application in the College of Medicine
- Omics application: Dajiang Liu
- Protein Structure: Jian Wang
- Electronic health records: Vida Abedi
Dajiang Liu, PhD
Professor Liu is a professor and Vice Chair of Research in Penn State College of Medicine’s Department of Public Health Sciences. He also leads the Penn State College of Medicine Strategic Plan Goal 2 on Artificial Intelligence and Biomedical Informatics Professor Liu and his laboratory use advanced informatics and machine learning methods to understand how genes and environment influence human disease risk. The Liu lab has developed the widely-used software packages for genetic data analysis. His lab also leads multiple large-scale genetic studies on substance use and addiction, autoimmune diseases, lipids and cardiovascular diseases.
Xingyan Wang, PhD, MS
Dr. Wang is an Assistant Professor. His research is devoted to the development and application of machine learning and informatics approaches to analyze large genomic and multi-omics data to understand risk factors underlying human diseases. Via the tools that he developed, he also leads several consortium studies on smoking and drinking addictions. He published his work as first-authors in a number of leading journals, including Nature and Nature Genetics.
Vida Abedi, PhD, MS
Dr. Vida Abedi, an Associate Professor at Penn State College of Medicine’s Department of Public Health Sciences, specializes in interdisciplinary research, encompassing machine learning and artificial intelligence, with a primary focus on electronic health record (EHR) mining. Over the past eight years, her team has worked on integrating machine learning techniques into clinical processes, particularly in cardio and cerebrovascular, infection, and immune-mediated diseases. Among recent work, her team has developed a strategy to improve data quality from EHR for machine learning applications, aiming to reduce algorithmic bias.
Matthew Swulius, PhD
Matt Swulius is an Assistant Professor in the Biochemistry and Molecular Biology department. He is also a Scientific Director of the Cryo-EM Core Facility at Penn State College of Medicine. His research focuses on mechanisms of neuronal development, where he uses a multimodal imaging approach to correlate cellular dynamics with nanoscale changes in cytoskeletal architecture. His group also uses deep learning approaches to denoise and segment large quantities of cryo-electron tomographic data, and they are one of the pioneers of using synthetic data for training neural networks to parse real microscopic data.
Jian Wang, PhD
Dr. Jian Wang is an Assistant Professor in the Department of Pharmacology at Penn State College of Medicine. He is an expert in computational biology with 10 years of experience. He published over 40 research articles and received several awards for his contributions to the field, including the Bridges to Translation Pilot Grant of Penn State Clinical and Translational Science Institute and an NIH R01 project.
Sen Yang, PhD
Dr. Yang is an (upcoming) Assistant Professor. His research uses deep learning methods to predict the occurrence and progression of human diseases. He has worked on deep learning models for microbiome data analysis, imaging analysis, and genomics.