The project aims to explore available AI tools and resources to meet the needs of stakeholders within the AIM-AHEAD Coordinating Center (ACC). It focuses on promoting innovation and collaboration among AIM-AHEAD members through open-source AI/ML toolkits. Additionally, the project seeks to nurture a community of researchers by providing support and training on M-POWER open-source tools, addressing unmet needs, and facilitating workforce development.
The project is supported by a team of UTSA faculty and MATRIX thrust leads, including Amina Qutub, Ambika Mathur, Kevin Desai, Panagiotis Markopoulos, Anandi Dutta, and Erica Sosa along with UT Health San Antonio professor Mark P. Goldberg, MD.
MATRIX: The UTSA AI Consortium for Human Well-Being received funding from the National Institutes of Health (NIH)’s AIM AHEAD program, which seeks to advance health equity and research diversity through technology.
EVENTS
Past/Ongoing Workshop/Tutorial/Seminar/Bootcamp on Crossesction of AI and Healthcare (in-person and virtual)
1) Matrix AI Seminar Series: Mapping the Stroke “RE-Connectome” by Mark Goldberg, MD, Professor of Neurology, Associate Vice President for Strategic Research Initiatives, UT Health San Antonio
Seminar Date: 09/30/2022
Seminar Description: How can the brain restore its function when parts are irreversibly injured? Stroke, or “brain attack” occurs when an artery supplying one part of the brain becomes blocked by atherosclerosis or blood clot. The blockage may resolve spontaneously, or hospital stroke specialists may use drugs or catheters to remove the clot. But if the artery remains blocked for several hours the tissue will be permanently damaged, and the stroke patient will have deficits that reflect the brain region, such as loss of strength, sensation, vision, speech or memory. Surprisingly, most stroke patients have at least partial recovery of function over the next 90 days, even though the injured neurons and other brain cells do not regenerate. This occurs because of neural plasticity, the ability of the nervous system to change its activity in response to intrinsic or extrinsic stimuli by reorganizing its structure, functions, or connections.
Our lab studies a unique form of post-stroke plasticity in a mouse model. After a targeted injury of the primary motor cortex on one hemisphere, the axons from cortex to spinal cord (corticospinal tract) degenerate and muscle function is impaired. Within 2-4 weeks, axons from the opposite – uninjured – corticospinal tract send new collateral sprouts across the spinal cord, where they form synapses that replace the ones lost after stroke. This is a remarkable form of plasticity because the neurons responsible for repair are far from the location of injury. Neurons and supporting glial cells activate molecular programs of axon growth that do not normally occur in the adult nervous system.
We’ve developed a microscopy and image analysis pipeline to generate large scale 3D data of the full set of brain and spinal connections that change after injury (which we term the “RE-Connectome”). Viral vectors and transgenic mice express fluorescent proteins in axons and presynaptic terminals. Mouse brains and spinal cords are removed and fixed after stroke and imaged using a serial two-photon tomography microscope. The pipeline for whole brain image analysis includes supervised machine learning (pixel-wise random forest models via the “ilastik” software package) followed by registration to a standardized 3-D atlas of the adult mouse brain or spinal cord. In this talk, we’ll present some conclusions from our reconnectome data, and review challenges with image segmentation. This model of nervous system plasticity may provide approaches to promoting resilience in human stroke or in neuromorphic networks.
2) Matrix AI Seminar Series: Eye-AI: A New Approach to Quality, Labeling, and Vetting in Radiology by Dr. Kal Clark, University of Texas Health San Antonio
Seminar Date: 10/21/2022
Seminar Description: The predominant source of medical errors in radiology is surprisingly not deficient medical knowledge. Rather, errors primarily stem from the methods radiologists use to visually inspect the image, referred to as “perceptual errors.” Perception-related errors are estimated to reach 33% on abnormal chest radiology exams. To mitigate this problem, radiologists employ personalized and high-dimensional visual search strategies, otherwise known as search patterns. Quantitative descriptions of these search patterns via direct observation or self-report are not only unreliable, but also interfere with quality improvement interventions, which therefore negatively impact patient care. Through the use of eye-tracking technology and feature extraction we can employ novel techniques to improve health care quality, passively label datasets, and vet the next cohort of medical practitioners.
3) Matrix AI Seminar Series: Computational Methods for Analysis of Functional Brain Images by Dr. Ananth Grama, Purdue University
Seminar Date: 11/4/2022
Seminar Description: Rapid advances in neuroimaging have resulted in large repositories of images with high temporal and spatial resolutions. This has motivated complex connectomic analyses aimed at understanding representation and processing of stimuli . Such analyses require novel computational tools that significantly extend the state-of-the-art in machine learning and data science. These studies have far-reaching implications for the fields of precision psychiatry, behavioural analysis, and neurodegeneration, in addition to applications in AR/VR and advanced human interfaces.
In this talk, I will discuss recent work in our group on understanding neuronal response to naturalistic visual stimulus — I present two methods based on archetypal analysis and deep learning to: (a) find interpretable representations of fMRI response, (b) predict objects in visual frames, and (c) reconstruct visual inputs. I will also briefly describe computational methods to characterize individual-level uniqueness of functional connectomes. I will present three methods based on matrix sampling, graph alignment, and linear algebraic techniques to accurately identify the identity of individuals, and the nature of the cognitive task being performed. I will conclude my talk by highlighting the immense significance and challenges posed by problems in the field of neuroimage analyses.
4) AI Bootcamp on Artificial Intelligence for MD/MSAI (Doctor of Medicine and Master of Science in Artificial Intelligence) students
Workshop Date: July 10, 2023 – July 21, 2023 (In-person and Virtual)
Workshop Description: This BootCamp is designed to prepare students for the next generation of healthcare advances by providing comprehensive training in applied artificial intelligence
5) Matrix AI Seminar Series: Workforce Diversity, Education Informatics, and AI: Challenges, Innovation, and Future Directions by Toufeeq Ahmed Syed, PhD, MS, Associate Professor, Assistant Dean of Education Informatics, McWilliams School of Biomedical Informatics, McGovern Medical School, The University of Texas Health Science Center at Houston
Seminar Date: 9/22/2023
Seminar Description: In this talk, Dr. Toufeeq Syed will discuss the role AI and Informatics can play in supporting life-long self-directed learners, increasing researcher diversity, creating national mentoring networks, and reducing health disparities. Dr. Ahmed co-leads National Research Mentoring Network (NRMN). The goal of NRMN is to increase the diversity of the biomedical workforce necessary to meet the projected needs by providing mentorship and networking activities, as well as serve as a national resource for best practices in mentoring and networking. For NRMN, Dr. Ahmed designed and developed MyNRMN (https://my.nrmnet.net), a powerful mentoring and networking platform custom-built to support NRMN mentors, mentees, and the community. With over 7,600 mentors and 14,600 mentees participating on the platform, MyNRMN enables mentees and mentors to connect professionally (11,300 mentoring connections made), support mentoring relationships, and build their professional networks. Our members represent more than 3,700 institutions from all 50 states. Dr. Ahmed serves as (multiple) Principal Investigator of the NIH-funded AIM-AHEAD program (Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity). With other national leaders in health equity and AI/ML, he is building a national consortium with goals to enhance the participation and representation of underrepresented minority students, faculty, and researchers in developing AI/ML models and reducing health disparities and health inequities. For this project, he has developed an online national platform, AIM-AHEAD Connect (https://connect.aim-ahead.net), to support participant registrations, the proposal application process, fellows/students, and AI experts to network/mentor, online collaborations, projects, helpdesk, and online courses. Previously, he served as Executive Director in Health IT at Vanderbilt University Medical Center and Assistant Dean of Education Informatics at Vanderbilt School of Medicine, and developed VSTAR (https://vstar.app.vanderbilt.edu), an integrated learning management platform to deliver competency-based education (Curriculum 2.0), and designed and developed innovative learning and assessment applications including QuizTime (https://quiztime.app.vumc.org), Compass, Portfolio, TurnOut (https://turnout.app.vumc.org), and Spark. He provides technical leadership in national and international medical education technology and standards committees.
6) Matrix AI Seminar Series: Data Science for Translational Research by Dr. Debashis Sahoo, PhD Associate Professor Pediatrics and Computer Science and Engineering University of California San Diego.
Seminar Date: 11/17/2023
Seminar Description: Translational research aims to swiftly apply basic science discoveries in practical healthcare. Productivity hurdles can impede the quick translation of research into healthcare breakthroughs, with issues like reproducibility and generalizability of biological results in clinical settings posing significant challenges. Concerns about irreproducible results are on the rise, with over 70% of researchers reportedly facing challenges reproducing others’ experiments, and more than half struggling with their own. Drug development, a crucial facet of translational research, maintains a low success rate despite past efforts.
Part of the problem lies in the tools of data science and statistics used. I’ll introduce you to a new approach—Boolean analysis of large and diverse biological datasets. In my research group, we’ve pioneered this method and developed computational tools to uncover invariant logical relationships between genes. Join me as we explore how leveraging these invariant relationships can enhance the reproducibility of current approaches. We will discuss how Boolean analysis can transform the power of AI/ML approaches.
7) Introduction to Large Language Models: Focus Area: Healthcare Use Cases
Workshop Date: 01/24/2024 (In-person and Virtual)
Workshop Description: Join our workshop for an exploration of the transformative applications of large language models (LLMs) in the healthcare sector. Discover how LLMs like GPT-3 and BERT are revolutionizing clinical documentation, medical literature analysis, and patient communication. We’ll present open-source tools such as Hugging Face Transformers and spaCy, exploring their role in facilitating LLM adoption. Engage with recent contributions in the field, unlocking the potential of these models to enhance diagnostics, streamline administrative tasks, and ultimately improve patient care.
8) Matrix AI Seminar Series: Through the Lens of Innovation: AI-powered Healthcare Solutions for Enhanced Outcomes and Health Equity by Nawar Shara, Ph.D. Chief, Research Data Science, Co-Director of the Center of Biostatistics, Informatics and Data Science (CBIDS), MedStar Health Research Institute, Washington D.C.
Seminar Date: 03/22/2024
Seminar Description: My talk will focus on the power of innovative solutions such as AI-powered chatbots and ML-based clinical decision support systems in healthcare and the impactful role they can play. The power of AI coupled with the catalytic role of data in redefining patient experiences and in improving healthcare utilization and patient outcomes. I will demonstrate how the integration of AI and data from different sources is reshaping healthcare paradigms, driving inclusivity, and paving the way for a patient-centric future that hopefully will leave no one behind.
9) Introduction to Computer Vision: Focus Area: Healthcare Use Cases
Workshop Date: TBD
Workshop Description: In this workshop, we delve into the fundamentals of computer vision with a special emphasis on its application in the medical domain. This workshop will involve – 1) Understanding the basics of image processing and computer vision algorithms, 2) Introduction to convolutional neural networks (CNNs) and their applications in medical image classification, segmentation, and detection, 3) Real-world examples showcasing how computer vision is transforming medical diagnosis.
Throughout the workshop, participants will gain hands-on experience with essential computer vision techniques and tools tailored specifically for medical use cases. This workshop is designed for healthcare professionals, medical researchers, data scientists, and anyone interested in leveraging computer vision to address challenges in the medical field. No prior experience in computer vision is required, although familiarity with basic programming concepts will be beneficial, especially with Python programming using Google Colab or Jupyter Notebooks.
10) Introduction to Federated Learning: Focus Area: Healthcare Use Cases
Workshop Date: TBD
Workshop Description: TBD