Nurses and Data Science
Download as PPTX, PDF•0 likes•31 views
The document summarizes the author's journey working with healthcare data and machine learning to improve clinical outcomes. It discusses challenges in translating data insights into action, and the need for collaboration between data scientists and medical professionals like nurses. Nurses understand healthcare interventions and how to improve patient outcomes, while data scientists create models but lack clinical expertise. The author provides their toolbox for finding opportunities and communicating skills to facilitate collaborations that leverage both data science and clinical knowledge.
![High Level Game Plane
1. Literature Review [pubmed, ebscho, uptodate]– find evidenced based
interventions to support the implementation of some type of
intervention (most likely nurse, PT, social worker led)
2. Human Capital - hospital/client resources (e.g., social workers, nurse
specialists) that are already in place
3. Understand output from ML/Ai specialists – what it can and can’t do
4. Pairing of human capital resources to EBP’s](https://image.slidesharecdn.com/hwilliamspresentationnortheasternboston-200123032612/85/Nurses-and-Data-Science-19-320.jpg)
Nurses and Data Science
- 1. Translating Data into Clinical Change Hants Williams, PhD, RN Founder & Chief Innovation Officer - BioVirtua The Black Box of Data
- 2. Part 1 - Problem Part 2 – Solutions and Toolkit Part 3 – Taking action and Executing
- 3. My Journey Data & Blackbox Nurses
- 4. Birds eye view of my journey thus far… Home Education Consulting Founder
- 5. Duke University, PhD Nonpharmacological pain management Development of remote MBSR protocols UMMC, Postdoc Pain Genetics RNA sequencing multidimensional pain Parsley Health Data Scientist Customer insights, health outcomes pulseData Health Informaticist Data translation, clinical recommendations Robincare Intervention consultant Mindfulness-based interventions for cancer
- 8. Data ML/Ai Outcome “Data Scientist Stuff”
- 10. “Statistics is the science of learning from data...” - American Statistical Association
- 12. Data ML/Ai ? ⇧ Outcome Translate into action
- 14. Mathematical Models Translation Improved Health Basic Science Translational Science Clinical Science
- 15. Data Problem ML/Ai ModelModel Translation Intervention with Metric
- 16. Real world example: Background: pulseData, preventable chronic diseases with machine learning – chronic kidney disease Problem: exploring partnership with Oscar health, how could they integrate, provide additional value Solution:
- 19. High Level Game Plane 1. Literature Review [pubmed, ebscho, uptodate]– find evidenced based interventions to support the implementation of some type of intervention (most likely nurse, PT, social worker led) 2. Human Capital - hospital/client resources (e.g., social workers, nurse specialists) that are already in place 3. Understand output from ML/Ai specialists – what it can and can’t do 4. Pairing of human capital resources to EBP’s
- 20. Do nurses need to become experts in ML/Ai? No Do nurses need to understand basics of ML/Ai? Yes
- 21. Data scientists are not trained in: • Anatomy • Disease/pathology • Pharmacogenetics • Acute Health • Hospital Operations/Management
- 22. Data scientists’ are mathematicians/visualization experts that should be agnostic to the context or scenario Can create ML/Ai models, but not healthcare professionals, and are not trained to translate their findings into improved patient outcomes Nurses’ are focused on improving patient reported outcomes/metrics through delivery of evidenced-based care Nurses are trained and knowledgeable of healthcare interventions that could be delivered, and what may be most effective
- 23. Famous Failures and Problems IBM Watson – University Texas “The partnership between IBM and one of the world’s top cancer research institutions is falling apart. The project is on hold, MD Anderson confirms, and has been since late last year. MD Anderson is actively requesting bids from other contractors who might replace IBM in future efforts. And a scathing report from auditors at the University of Texas says the project cost MD Anderson more than $62 million and yet did not meet its goals. The report, however, states: ‘Results stated herein should not be interpreted as an opinion on the scientific basis or functional capabilities of the system in its current state’….”
- 26. • pulseData --> They create risk scores for preventable illnesses, but what do they do with those scores? • Parsley Health --> They provide adjunct doctor visits and health coaching interactions, but lack evidenced-based protocols for coaches • Robincare Using non-pharmacological approaches for managing cancer in young adults, but with approaches that are not matched with the condition (e.g., generic use of mindfulness) Omada Care (prediabetes) Vita Health (weight management) Flatiron (cancer)
- 27. The best part of this ‘blackbox’ - you don’t need to be a data scientist, you need to be a nurse with only a basic understanding of what is under the hood
- 28. “…provide input to the overall NIH vision and actions undertaken by each of the 27 Institutes and Centers in support of biomedical research as a digital enterprise. Among other duties, the office oversees the Big Data to Knowledge (BD2K) initiative, stimulating the best developments in the data science community.” Patricia Flatley Brennan, RN, PhD NIH Interim Associate Director for Data Science National Library of Medicine Director
- 30. My Toolbox (“stack”) How I Utilize Free & Open Source Tools to… 1. Find opportunities (e.g., clients, projects) 2. Learning basic ‘data science’ – a.k.a how to make pretty graphs and do statistics 3. Attract and sell my skillset Discloser: I am receiving no monetary or other forms of compensations from any of the services I am about to discuss
- 31. Findings Opportunities and Clients
- 32. Data Specific Tools to Solve The Problems Interact w/data Example code Visualize & Report
- 33. Getting The Word Out – my skills
- 34. Wordpress and the Divi Theme – Getting the Word Out
- 35. Wordpress and the Divi Theme – Getting the Word Out
- 44. Alternatives to VC money-
- 45. My contact: Personal email: [email protected] 650-218-3789 Schedule some time to chat with me @ https://calendly.com/hants
Editor's Notes
- #3: Part 1: My journey The data translation problem Solving it with nurses Part 2: How you can do the same…. A few modern day tools you will not learn in school Part 3: How I applied the tools to get where I’m at now
- #5: San Francisco Educartion: DUKE, SFSU, SJSU, UMB.UMMC > Received multiple individual awards from the NIH, NINR - APS / Board Member, Consulting: PulseData/Parsley Health, RobinCare, Founder: BioVirtua Public Health Nurse, CA PhD – Non-pharm pain management; Duke University Postdoc - Pain genetics, University of Maryland Medical Center Private/individual consultant Formal: Parsley Health & pulseData Informal: Robincare BioVirtua – Founder, Chief Innovation Officer; venture capital backed
- #7: [bring in some of the slides, information from old PPT that you gave after going to the NIH data science boot camp] BRING IN A LIST OF ALL THE POTENTIAL COMPANIES THAT ARE TRYING TO DO THINGS WITH ML/AI Then do a quick search of which of those companies have succeeded, failed - ? I think there is a website that also talks about the failed companies, type of angellist version, maybe see if can find out Then bring up the black box – ML/Ai is cool, but what the hell is it? What does it do?
- #8: Historically, this is how things have looked:
- #9: But what is a data scientist? What do they do?
- #12: The real problem isn’t the black box of what the data scientist does, it’s the black box of what to then do after the data scientist finishes
- #15: Save this slide for later on
- #24: What famous failures our there that I can cite? IBM watson? What other startups? IBM watson: After four years it had not produced a tool for use with patients that was ready to go beyond pilot tests
- #34: Wordpress.com –> You need to PAY Wordpress.org IT IS FREE Elegant themes -> have very nice themes for wordpress check out the DIVI theme Flaticons is where I got all the icons for this presentation, as well as my own websites
- #45: Look Out for “OLD” Money, stayed focused on healthcare knowledgeable ventures that understand you can’t expect the same return 1-yr out as you would a pure tech company