17 Sep Message from Stan Huff: Sharing Interoperable Clinical Decision Support Applications Can Prevent Deaths
“The call to action is to invest the time to create interoperability so that we can take better care of patients. We need to do something about the 250,000 patients who die every year from preventable medical errors.” -Dr. Stanley Huff
Dr. Stanley Huff, Chief Medical Information Officer at Intermountain Healthcare and PMWC 2020 AI/ML Data Science Track Chair shared with us his thoughts on why we need interoperability to advance patient care and precision medicine. Find the full interview on our blog post. Here are some excerpts from the interview.
“New technologies such as digital health, AI/ML, and Data Sciences will help us discover new knowledge.”
New technologies help us create knowledge that helps us understand how to better treat patients, and help them attain better health. Our goal is to prevent illness and improve the quality of healthcare that we provide to them. We will see increasing impact as people continue to implement and use new knowledge discovery tools. This is a journey, and on this journey we are learning about new associations between a person’s genetic makeup and their health. For example, knowing how their body metabolizes a specific drug will allow us to determine whether the drug will be effective in this specific patient and whether we should prescribe the drug or not.
The expectations we have for these technologies:
1. Acceleration of knowledge extraction: The algorithms are fairly tolerant when it comes to missing data or inaccurate data and they can largely overcome the problem of dirty data if they have data from a large number of people to review. Large data sets allow us to overcome the errors that are part of all real world data. At the same time, it is critical that the data is as clean as possible to start with.
2. Access to the raw/clinical data: We have access to many kinds of data, but there are lots of valuable data that is either not noted in clinical systems, or it is hidden in narrative text somewhere in the record. We could answer many more questions if we had more complete structured and coded data from source systems. We need higher consistency in data representation and data collection, which overall would lead to faster learning. I would expect new technologies to address this component more efficiently in the near future.
“The single biggest motivation to create FHIR (Fast Healthcare Interoperability Resources) is to make sharing standard data easier.”
FHIR was designed specifically to work with the same infrastructure and tools that people use when developing web applications. FHIR uses the same kind of RESTful services for accessing health data that are used for accessing data in any other kind of web application. People have access to a huge set of tools and knowledge of how to use RESTful services. Therefore, the motivation behind FHIR was to make it easy and simple for people who already know how to develop web applications to share and create interoperability using the FHIR standards for accessing and using health data and information via services.
“The call is to invest the time to create interoperability so that we can take better care of patients.”
We need to do something about the 250,000 patients who die every year from preventable medical errors. I am not sure why more people aren’t talking about that epidemic. That number of preventable deaths should be a real motivator for everyone. The ability to share highly interoperable clinical decision support applications could lead to prevention of many of those deaths and allow us to take better care of patients, and at a lower cost. We need to invest the time to create interoperability as a way of sharing executable knowledge in the form of interoperable advanced decision support applications.
PMWC 2020 Silicon Valley has planned a three-day AI/ML Data Science track (Track 2) that focuses on various aspects of data science in healthcare, including Data Science in Drug Discovery, Clinical Trial Design and Patient Selection, Data Mining and Visualization, and Driving Towards Value-Based Care, see more details in the links below. In addition, PMWC 2020 will have four more tracks with a focus on the following themes:
- “Emerging Therapeutics” (Track 1),
- “Diagnostics in Clinical Practice” (Track 3),
- “Molecular Profiling – From Research to the Clinic” (Track 4),
- “Health Data, Microbial Profiling & Patient’s Education” (Track 5).
See the 400+ Speaker January 21-24th PMWC Program, taking place at the Santa Clara Convention Center.
Driving Towards Value-Based Care – A session chaired by AI/ML Data Science Track Chair Stan Huff
There is a massive amount of data that needs to be shared and made accessible to improve survival outcomes and quality of life for patients. There are many dimensions to this besides just providing the best care for an individual patient. The other areas that are of great interest are public health, observational research data, clinical trial data, and data from various devices. Specific challenges aside, the core issue is how interoperable we can make the healthcare system. This session will cover various aspects of interoperability in pursuit of value-based care. Read Stan Huff’s full bio.
See 17 sessions in the AI/Data Sciences Track with over 60 speakers:
AI Ethics & Privacy– Chair: Camille Nebeker, UCSD
AI/ML-Driven Clinical Trial Design and Patient Selection– Chair: Shweta Singh Maniar, Google Cloud
Computational Models That Expedite Clinical Diagnostics– Chair: Suchi Sara, John Hopkins
Data Mining and Visualization– Chair: Aalpen Patel, Geisinger
Data Science Drives Healthcare– Chair: Christopher P Boone, Pfizer, Inc.
Data Science For Payers– Chair: Viet Nguyen, HL7 Da Vinci Project
Data Science In Drug Discovery– Chair: Ron Alfa, Recursion Pharmaceuticals
Data Science In Hospitals & Health Systems– Chair: John Mattison, Kaiser Permanente
Driving Toward Value-Based Care– Chair: Stanley Huff, Intermountain Healthcare
FDA Perspective– Chair: Sean Khozin, FDA
Fireside chat- with Atul Butte, UCSF and Cora Han, Chief Health Data Officer, UC Health
Fireside chat- with Lars Steinmetz, SOPHiA GENETICS and Phil Febbo, Illumina
Healthcare Data Monetization Models-Chair: Eric Marton, Wavemaker 360 Health
Machine Learning Enriched Antibody Discovery – Chair: Matthew P Greving, RubrYc Therapeutics
Overcoming Data Analytics Challenges- Chair: Milan Radovich, Indiana U. School of Medicine
Semantic AI In Drug Discovery And Clinical Trials– Chair: Janusz Dutkowski, Data4Cure
VC Perspective And Fund-Raising Strategies– Chair: Greg Yap, Menlo Ventures