AI News, Explainable AI in medical informatics and decision support artificial intelligence

Biomedical Informatics and Medical Education

We are moving toward our vision with a number of activities across our various programs.

The work of our fellows in the clinical informatics fellowship program has received plaudits from clinical administrators and faculty, and we are currently recruiting a new faculty member in our department to assist with this program (view position description).

Building on this effort, and to further support the sharing of data within, the CSER CC requested and received a multi-year supplemental award in late 2019 to manage the harmonized outcomes and measures, and sequence data, and to have the CC serve as a data coordinating center (DCC).  The proposed ecosystem is one of the first National Human Genome Research Institute (NHGRI) programs to leverage the Genomic Analysis, Visualization, and Informatics Lab-space (AnVIL) resource for data deposition, storage, and retrieval.

key component of the new Hans Rosling Center for Population Health – opening this summer – is telling the story of the major advances, achievements, and contributors to improvements in population health (i.e., human health, environmental resilience, social and economic equity) that have occurred both domestically and internationally.

Our collective goal is to advance healthcare systems driven by data, evidence, and best practice with the explicit purpose of creating better health information technologies.

December 9-13, 2019 Jimmy Phuong was invited to give an e-lightning talk with an electronic poster at the American Geophysical Union (AGU) 2019 conference entitled, “Information needs for supporting population health researchers in hurricane and flood disasters”

December 2-6, 2019 Gang Luo’s lab obtained a subcontract from the VA Puget Sound Healthcare System to support a PhD student in Dr. Luo’s lab for the next few quarters to help the VA PI develop a deep learning model to predict no shows and walk-ins at VA.

Health Science Building, Room I-132 Title: Improving Design and Usability of Interactive Vulnerability Mapping for Global Health Preparedness Abstract: Global health preparedness –the ability of organizations and governments to anticipate and respond to health events and emergencies– presents an imperative, yet challenging, opportunity for public health informatics interventions.

In my dissertation research, I will be introducing a new type of SSDS, interactive vulnerability mapping tools, which can help decision makers in global health preparedness identify spatial areas that are at risk for disease outbreak.

Decision makers include epidemiologist, public health planners, vector control specialists, and directors, who might use this information to allocate resources or plan outreach activities to high risk regions.

Given gaps in prior research on data visualization and usability in global health informatics, the objective of this dissertation project is to develop visualization tools that are evaluated for usability, propose design standards for data visualizations in global health preparedness, and contribute visualization principles that can enhance interactive vulnerability mapping tools.

This work will contribute: 1) usable SDSS tools designed for public health decision makers in Peru and Kenya settings, 2) empirical data on the design, data visualization preferences, usability and acceptance of SSDS for disease vulnerability in global health settings, 3) demonstration of methodological approach for group usability in global public health settings, and 4) theoretical insights into the fit and translation of sociotechnical systems factors for global health informatics tools.

In disastrous times, spatiotemporally-relevant information escalate in importance as health systems strive to address emergent concerns, pre-existing needs, population migration, while experiencing disruption in available resources and infrastructure.

With their adoption by hospitals and health systems, Electronic Health Records (EHRs) contain a richness and diversity of information about patients that could inform where and how to prepare for population-scale patient needs in future disaster scenarios;

The aims are to: 1) assess information needs and priority use-cases for population health research in hydrologic disaster preparedness, 2) design spatiotemporal use-case workflows to survey trends and anomalies for regional areas using gridded hydrometeorological data products, a surrogate for structured multivariate datasets, and 3) develop an approach for spatiotemporal inferential statistics of EHR patient diagnosis information.

Title: Explainable AI in Healthcare: Prospects and Limitations Abstract: As AI and machine learning are being increasingly integrated into healthcare, challenges regarding creating responsible AI systems that are interpretable, fair, transparent, unbiased, robust and reliable are coming under increasing scrutiny.

Drawing on insights from the academia and experience from industry the talk will delve into why is explainability in healthcare different from other domains and how deployed systems may have counterintuitive implications e.g., explainability leading to distrust, explainable systems leading to more opaque models etc.

In this role, Dr. Eckert leads and works with doctors, data scientists, and developers to identify patterns in patient data to predict risk that can cost-effectively improve care outcomes.

Health Sciences Building, Room T478 Facilitators: Will Kearns, Aakash Sur (BHI Graduate Students), Trevor Cohen, BHI Faculty Andrea Hartzler will be giving a talk in the iSchool Research Symposium on Dec 2nd: UNBIASED: UNDERSTANDING PATIENT-PROVIDER INTERACTION AND SUPPORTING ENHANCED DISCOURSE Health-care bias — based on patients’ race, gender, socioeconomic status, sexual orientation, and other factors — leads to health disparities, such as lack of appropriate treatment and inadequate pain support.

Dr. Hartzler will discuss a new project funded by NIH, in which researchers combine social signal processing — a computational approach that detects subtle forms of bias in nonverbal communication — with reflective feedback designed in collaboration with patients and providers.

Health Sciences Building, Room E216 Title: Assessing the utility of digital health technology to improve our capacity to assess and intervene in depression Abstract: When it comes to mental health, no country is considered developed.  In the last decade, the burden of mental health disorders(MHD) has risen in all countries due to disparities in timely diagnosis and access to evidence-based treatments.

Additionally, scientists, are still conducting research to understand the underlying mechanisms behind MHD.  Part of the problem is that measures of symptom severity are all based on self-reports by patients and clinician observation often resulting in an imprecise measurement of MHD.

Additionally, in-clinic assessments tend to be episodic and often miss capturing the lived experience of disease over time including the potential impact of social and environmental factors that are suspected to be linked to neurodevelopmental and psychological processes.

To improve long term outcomes in MHD, there is a critical need to develop new ways to objectively assess specific underlying constructs of behavior patterns linked with neuropsychiatric conditions.  The pervasive network of smartphones offers researchers a unique opportunity to study MH at a population scale and at a fraction of the cost of traditional clinical research.

The high-frequency daily usage of smartphones also provides new ways to capture the individualized momentary experience of living with mental health issues based on “real-world data”(RWD) in an objective, momentary and nonreactive way.

The principal findings of this dissertation research show the feasibility of utilizing smartphones to reach, enroll and engage a diverse and nationally representative population as well as the potential of using RWD in predicting mental health outcomes.

The RWD collected from more than 2000 participants showed notable inter-/intra-person heterogeneity highlighting the challenges of developing a robust cohort level machine learning model to predict depression.

However, personalized N-of-1 models show the promise of “precision digital psychiatry” by assessing an individual’s drifts from their own average “digital behavior” as a more reliable predictor of a person’s daily mood.

Cross study analysis using data from >100,000 participants showed significant underlying biases in technology access and utilization based on participants’ demographics that could impact the generalizability of the statistical inference drawn.

In addition, the results from a survey-based study on a large and diverse sample show growing concerns among the general public about the security and privacy of their digital data which if left unaddressed can negatively influence people’s decision to participate and share data in digital health research.

These findings are contemporary and extend the on-going efforts to objectively evaluate the potential fit of technology in psychiatry in engaging the general population to monitor their mental health in the real world outside the clinic.

100,000) that are powered to detect the association between RWD and behavioral anomalies, the ability to integrate RWD across similar studies, improve equitable utilization of technology across a diverse and representative population and address people’s concerns about data security and privacy.

https://doi.org/10.1371/journal.pone.0225058 Andrea Hartzler will be giving a talk in the iSchool Research Symposium on Dec 2nd: UNBIASED: UNDERSTANDING PATIENT-PROVIDER INTERACTION AND SUPPORTING ENHANCED DISCOURSE Health-care bias — based on patients’ race, gender, socioeconomic status, sexual orientation, and other factors — leads to health disparities, such as lack of appropriate treatment and inadequate pain support.

Dr. Hartzler will discuss a new project funded by NIH, in which researchers combine social signal processing — a computational approach that detects subtle forms of bias in nonverbal communication — with reflective feedback designed in collaboration with patients and providers.

Please direct any questions to Jill Fulmore and John Gennari Eric Tham MD, Adjunct BIME faculty, was elected a Fellow of AMIA which “signals to patients, employers, and colleagues that you are an expert in evidence-based informatics practice and engaged with a community of life-long learners who apply the latest advances in informatics to improve health and health care”.

Health Informatics Lecture Series Thursday, 11:00am-11:50am, UW Medicine South Lake Union, Building C, C123 A/B November 21: Sean Mooney, PhD Title: Predictive analytics and machine learning in healthcare and the UW: Vision, anecdotes and comments on readiness Abstract: It is an opportune time to be engaged in the research and application of informatics in biomedicine.  The increased use of electronic and personal health records and personal mobile devices is creating many opportunities at research academic medical centers.  At the University of Washington, I believe we are laying the ground work to build the informatics and information technology infrastructure to support research on personalized approaches and the use of data science to enable them.

In this presentation, I will discuss our support of data for research use within UW Medicine, our efforts to build new machine learning and data science approaches using clinical datasets, and our efforts to develop new machine learning methods.  Further, we are leveraging the crowd by organizing and participating in community challenges (critical assessments) to build a better understanding of the types of approaches that perform well and in what context.  I will describe our involvement in the Institute of Translational Health Sciences, UW Medicine Information Technology Services, the Department of Biomedical Informatics and Medical Education and the National Center for Data To Health and how this collaboration is building upon a vision to enable research translation to clinical care with a particular emphasis on data science methodology.

Additionally, he has a strong interests in leveraging the community of scientists to collaboratively solve difficult problems in biomedical research through open challenges and has participated in several as an organizer, assessor, predictor and advisor.

Throughout his career, he has been an author on more than 100 publications, given more than 120 invited seminars and helped write the proposals of well over $100 million in funding.  He has won several awards and accolades for his work.

DSI - Unit -bdi

This research falls under the general umbrella of discovery informatics that has recently emerged as a field that explores the potential of applying various computer science technologies like Semantic Web, big data analytics, AI or machine learning to interdisciplinary challenges in turning data and information into actual knowledge.

This is reflected by the strategic vision of the unit that is to deliver novel scientific results with substantial impact potential in life sciences and healthcare.The core research topics of the unit are as follows: Representation learning for biomedical data (including networked data, protein/gene sequences, etc.);

Knowledge graph embeddings (statistical relational learning models, focused primarily on link prediction and knowledge base completion, but also on more advanced issues like discovery of causal relationships);

Healthcare AI: Explainable AI as a Service for Community Healthcare

Healthcare AI: Explainable AI as a Service for Community Healthcare Presented by SGInnovate & NUS-SSI Health technologies are changing fast, posing these ...

From Explainable AI to Human-centered AI | Andreas Holzinger | TEDxMedUniGraz

From Explainable AI to Human-centered AI Andreas Holzinger is currently Visiting Professor for explainable Artificial Intelligence at the Alberta Machine ...

Using Artificial Intelligence (AI) in medical coding Teaser

Artificial Intelligence is a hot topic, but how are hospital really using it today? Watch our 2-minutes video to know what our sympo is about.

From Explainable AI to Human‐Centered AI by Andreas Holzinger @UAlberta AI Seminar on 2019-7-26

Abstract: In the medical domain the expectations to automatic AI systems are high, particularly in disciplines requiring prognostic models (oncology) and/or ...

CogX 2018 - Pre-Primary Care: How AI Can Unlock Self-Care, The Biggest Opportunity in Healthcare

CognitionX: The AI Advice Platform Connecting organisations with a global, on-demand...

Why do We Need Artificial Intelligence In Medicine? | Kurt Zatloukal | TEDxMedUniGraz

Why do We Need Artificial Intelligence In Medicine? Kurt Zatloukal, M.D. is professor of pathology at the Medical University of Graz, Austria and head of the ...

Artificial Intelligence in Primary Care

Hear experts speak about strategies to ensure emerging technologies strengthen relationships and team-based primary care, while also reducing administrative ...

DRF 8: Interpretable Machine Learning to Deconstruct the Neural Basis of Psychiatric Disorders

David Carlson, PhD Assistant Professor Civil and Environmental Engineering Biostatistics and Bioninformatics Duke/DCRI.

Data ethics in healthcare, medicine, and public health

Health care and medicine are highly regulated professions and it shouldn't be too surprising that some of this applies to data science as well. Now, when it ...

Exploring AI and Machine Learning Approaches to Precision Medicine and Healthcare Overview

Dr. Frank Hollinger, Vice President of Corporate Operations for the Americas (Canada, USA, South America) at the InnovationWell Interaction meeting that took ...