One of the best things about being a HIMSS Social Media Ambassador (four years in a row!) is being asked to write about anything that has anything to do with healthcare workflow. It is both flattering and satisfying. So, when I was had an opportunity to write about the upcoming Big Data & Healthcare Analytics Forum (Boston, 10/23-24) I jumped at the chance! Also, please join the #PutData2Work Twitter Chat (today, 1PM EST, immediately after #KareoChat).
— Chuck Webster MDMSMS (@wareFLO) October 12, 2017
The relationship between data and workflow in healthcare is an interesting one. The Forum illustrates this. It emphasizes “action” based on data: “actionable information,” “actionable strategy,” and “actionable insights.” Action is part of the definition of workflow, since workflow is a series of actions, consuming resources, achieving goals. In fact, big data, data science, machine learning, and business intelligence platforms are helping to bring sophisticated process automation tools to healthcare.
In my three-series just before the HIMSS17 conference, I describe how workflow technology makes modern machine learning and data science initiatives possible. It is simply no longer practical, to manually download, transform, then put into a format causing useful action at the point of care. Data sets so large they cannot fit on puny desktop drives, and then so slow to upload and download and upload again, force us to, essentially, model data workflows and then execute these models, in the cloud, without continuous, direct, manual, human intervention. I discussed this at length in A guide to AI, machine learning and new workflow technologies at HIMSS17 Part 1: Machine learning and workflow.
In particular, I hope you’ll pay attention to the following three presentations at the Big Data & Healthcare Analytics Forum…
- Machine Learning: What Can It Do For Healthcare?
- What Healthcare Providers Need to Know
- Seven Ways We Are Screwing Up AI for Healthcare
…and ask yourself:
- How does workflow and process automation help make machine learning and smart systems practical?
- How does workflow and process automation make generated insights “actionable”?
I’ll close with a quote from Hal Wolf, President And Chief Executive Officer, HIMSS:
“We want to maximize the patient experience at each clinic, and thus it’s important that we not be too rigid about workflows and systems. The clinics have room for flexibility and innovation.”
Most folks think of workflows in terms of day-to-day tasks of clinicians and staff. Obviously these “flows” influence patient experience. However, data-flow is also a kind of workflow. These workflows exist as both models of data flow, and executions of these models by various kinds of engines (workflow, process, orchestration, and data pipeline engines). Big data, business intelligence, and machine learning platforms have, at their core, sophisticated models and engines necessary to strike the right balance, between efficiency through best practice standards, and flexibility for healthcare organizations to innovate.
I also hope to see you at the #PutData2Work Twitter Chat: Building a More Informed Healthcare System, at 1PM EST, on October 12,
Viva la workflow-powered data, and data-powered healthcare workflow!
If you are interested in the fascinating relationship between healthcare data and healthcare workflow, I hope you’ll follow me on Twitter at @wareFLO (for soft(ware) work(FLO)w).