Human-Computer Interaction Lab,
University of Maryland, College Park

Organizer

Sponsor

Intel Corporation

Initial Schedule
(Some changes are anticipated)

Related Links

University of Maryland Class Projects
Medical Device Resources
iHealth: Taking Control of Personal Healthcare
 

Articles

The Interaction Between Safety Culture and Uncertainty Over Device
Behaviour: The Limitations and Hazards of Telemedicine (PDF)
Chris Johnson, Univ. of Glasgow, Scotland

Attendees

Back row: Peter Castensen, Chris Gough, Richard Mead, Matthew Kim, Corey Angst, Geoffrey Clapp, Bill Hughes, Ron Kay, Chase Ighani, Stephanie Fishel, Piet de Groen Front row: Adam Darkins, Dena Puskin, Michael Ackerman, Wendy Angst, Dina Demner, Mara Hemminger, Ben Shneiderman, Chunyuan Liao Not Shown: Sandy Pentland, Catherine Plaisant

Abstracts

Introduction: Interface Design Requirements for Personal Medical Devices
Ben Shneiderman (ben@cs.umd.edu), University of Maryland

The strong interest in medical informatics is shifting its focus from physician-centered to patient-centered approaches. The emergence of inexpensive, yet powerful computing technologies, accompanied by growing user techno-fluency has led to a wide array of desktop, mobile, and web-based applications. The increasing user participation in healthcare and wellness monitoring has led to some turbulence because of increased patient knowledge, assertiveness, and responsibility. Unfortunately, the numerous early systems have inconsistent interfaces, incompatible file formats, and are often poorly designed. We hope to identify opportunities for human-computer interaction research and intervention to improve the potential for user success. I will show prototype designs (in part from student projects: http://www.cs.umd.edu/class/spring2004/cmsc434/index.html)
and suggest where improvements to commercial products could be made.

Healthwear: Medical Technology Becomes Wearable (PDF)
Alex Pentland, MIT Media Lab

Until recently, researchers have had little success in extending healthcare into the home environment, yet there clearly is a huge demand for this service. Americans currently spend $27 billion on healthcare outside the formal medical establishment, which they find difficult, expensive, and painful to access. A dramatic shift in the composition of the US population makes it absolutely necessary to develop such distributed systems.

To address these demands, a research group at the MIT Media Lab has been developing healthwear, wearable systems with sensors that can continuously monitor the user's vital signs, motor activity, social interactions, sleep patterns, and other health indicators. The system's software can use the data from these sensors to build a personalized profile of the user's physical performance and nervous system activation throughout the entire day-providing a truly personal medical record that could revolutionize healthcare.

The Impact and Opportunity of Ubiquitous and Secure Portable Personal Health Records
Wendy Angst, CapMed-Personal HealthKey

The Electronic Health Record and Personal Health Record are fast emerging markets. However, with the fragmented health care system and mobile patients, many challenges still remain with the interoperability of these systems. CapMed believes the patient is an untapped resource for aiding in the interoperability of the patient's health information across multiple domains.

This discussion will provide an overview on the issues surrounding interoperability, and provide a review of patient-held solutions available to immediately address this problem. Emphasis will be placed on USB-Flash Disk Technology and CapMed's participation in the Continuity of Care Record (CCR) demonstration project at the Towards Electronic Patient Records conference in Fort Lauderdale May 17-22.

Learning objectives of the presentation are as follows:
- Review of interoperability challenges
- CCR Demonstration Project review
- Evaluation results

Mayo Clinic/IBM Computational Biology Collaboration: UI Challenges
Piet C. de Groen, Mayo Clinic/IBM Computational Biology Collaboration

The computer was introduced in healthcare facilities to store and process demographic and financial data. At present, the content of the medical record is being transferred from paper to electronic media, with some facilities being paperless and others still partially paper-based. At the Mayo Clinic we have nearly completed the transfer from paper to electronic media. As the initial focus was patient care, the electronic format of the medical record was created and optimized to allow access to the record of one patient at a time. However, there is an increasing need to have access to cumulative data for research, administrative, quality assurance and educational purposes.

At Mayo Clinic Rochester, in collaboration with IBM, we have created a data warehouse that allows us to store clinical as well as genomic data of millions of patients. In addition, we have created a search engine that can identify subsets of patients with specific features [1]. This search engine is able to handle a large variety of data types, including ICD-9 codes, DRG codes, various forms of laboratory data (numeric data, text data, dilution data, changes from baseline data), demographic data and free text data content (unmodified text, lexicon-based normalized text, ontology-mapped concepts). In addition, the search engine provides the opportunity to search using event profiles – a tool that allows the query of specific conditions within a condition (“conditional conditions”, for example colon cancers that were diagnosed at a specific year). The resulting User Interface is complex. Extensive usability studies of untrained users of components as well as the complete interface were done to evaluate whether such a complex User Interface could be introduced as a general warehouse search engine. The results will be discussed.

1. de Groen PC, Dettinger Richard, Johnson Pete. Mayo Clinic/IBM computational biology collaboration: A simple user interface for complex queries. In: Universal Access in HCI – Volume 4 of the Proceedings of Human - Computer Interaction (HCI) International, 2003, pages 1083-7.

Integration of Consumer Collected Information and Healthcare Provider Organizational Data to Facilitate Patient Outcomes
Bill Hughes, IDX Corp

The US healthcare delivery system is going through a major transformation from a ‘disease-centered model’ to a ‘patient-centered model’. One of the challenges for the Healthcare Information Technology (HIT) industry is to develop the technology infrastructure and tools to support the collaborative IT requirements of the provider and an increasingly IT empowered consumer. As consumers assume more responsibility for their health management they will demand better digital access to their existing clinical information and the ability to share their personal, via next-generation medical devices, biometric data with their provider(s). Providers, on the other hand, will need to make the transition from processing episodic snapshots of patient data to managing and processing a continuous stream of patient data.

Supporting a Collaborative Care model will introduce a new set of usability issues for both consumers and providers because information will be shared seamlessly between the patient and provider. This will be especially important in supporting chronic care management, e.g., diabetes. high-blood pressure, etc.

In this session IDX will share early prototypes that address the following problems:

• The consumer’s ability to access their clinical record from the provider via the Internet

• The ability for a diabetic patient to store their blood sugar data in their personal health record and share this data with their provider

• The introduction of a provider based Case Manager application that would be responsible for monitoring and managing a population of patients that are remotely submitting blood sugar levels.

• A provider’s ability to view and process both patient and organization captured biometric data.

• The ability for ongoing electronic collaboration between the provider and patient via electronic messaging.

• The ability for the ‘Clinical System’ to automatically escalate abnormal biometric data and generate alerts.

Telemedicine: Expanding the Opportunity for Personal Medical Devices
Dena Puskin, Director of the Office for the Advancement of Telehealth, HRSA

Dr. Puskin will review the lessons learned from the Office for the Advancement's 16 years of funding telemedicine activities, emphasizing the growing use of low cost technologies to reach patients in their homes. The Health Resources and Services Administration of the U.S. Department of Health and Human Services has established the Office for the Advancement of Telehealth (OAT) (http://telehealth.hrsa.gov/) to serve as a leader in telehealth. It is a focal point for HRSA's telehealth activities and as a catalyst for the wider adoption of advanced technologies in the provision of health care services and education.

Designs for Dialogue Systems for Telehealth Monitoring
Geoffrey Clapp, Health Hero Network, Inc.

The challenge of any technology deployed in the home for remote patient monitoring is building an engaging trusted interface between the technologies in the home and the patient. Mr. Clapp will discuss the key lessons learned by Health Hero Network over the last six (6) years of deploying the Health Buddy system into patient homes, and the successful collection of millions of data points from patients with varied clinical diagnosis, from common chronic illness such as diabetes and chronic heart failure, to schizophrenia, hepatitis C, and pediatric asthma.

Specifically, Mr. Clapp will address:
* building the patient trust with a device in the home
* eliciting factual, timely and appropriate information
* building a multi-disease interface
* cultural lessons learned from international deployment
* key limitations addressed over the deployment timeline
* the future roadmap of dialog systems for remote patient monitoring

PubMed on Tap: Access PubMed on Handheld, Wireless Devices
Dina Demner-Fushman, University of Maryland

Medline and other NLM databases contain a wealth of biomedical information that can successfully support the clinical decision process, as well as facilitate consumers' access to health information.

PubMed on Tap is a testbed system for delivery of this information when and where it is needed, which implies a mobile information delivery system, and small wireless devices such as handheld computers (PDAs) and tablet PCs.

In designing the interface for the small screens of these devices we strived to preserve the functionality available in PubMed, NLM's service that provides online access to over 14 million citations for biomedical articles. Here we present the latest version of the interface for the handheld computers, and the changes in its design based on a small usability study and users' feedback.