adam baker

To follow up on Dede Ainbinder recent post: If we use the current EHR data to build a AI clinical decision support tool it is unlikely to capture our clinical decision making process. EHR's are inherently designed to maximize billing and minimize risk of litigation. A clinical decision support tool based on this data will likely do the same. There is a large sunken cost fallacy here where we continue to hang on to garbage data. Just because the data gathering and storage process was capital and time intensive doesnt mean the data is valuable. To truly capture the clinical decision process we need to capture data which sits outside billing and litigation. Its hard to imagine a future where this data isnt used for billing and cannot be called upon during litigation. #EHR #healthdata #clinicaldecision

17

8 Comments

The important milestone of medical AI development was highlighted by Eric Topol, MD in his Ground Truth publication "When Medical A.I. is Lifesaving. Counterbalanced by some disappointments". At the end of April, research group from Taiwan published a paper describing results of a  multisite randomized controlled trial evaluating the ability of an AI-enabled electrocardiogram analysis tool to identify hospitalized patients with a high risk of mortality. The study was appeared in Nature Medicine and included 39 physicians and 15,965 patients. Electrocardiograms of patients in interventional arm were analyzed with AI and physicians received alert with report and warning messages, flagging patients predicted to be at high risk of mortality. The control arm included patients whose ECGs were analyzed by conventional method (by a physician with no AI involvement). The implementation of the AI-ECG alert was associated with a significant reduction (17%) of all-cause mortality within 90 days and saving 7 lives per 100 in some groups. As Dr. Topol pointed out, the results are comparable to most successful medical treatments, such as statins. Notably, it was the first-ever RCT to estimate the effect of AI-based intervention on patients' mortality. However, the results of four trials of real world scenarios of Generative AI use, as described in the same Ground Truth publication, were not so promising.  Large Language Models (including Chat GPT 4) were unsuccessful in answering oncology questions and assisting in medical coding. They also failed to reduce clinicians’ time spent writing reply messages to patients (though they did generate more empathetic replies) and posed a risk of severe harm in 7% of generated messages which is definitely not acceptable. Surely, as Dr. Topol stated “Progress in medical A.I. won’t occur in a straight line”. It has been repeatedly confirmed that AI is far better than humans at analyzing repeated patterns not generated by humans, such as MRI or ECG. However, when it comes to working with more abstract entities, such as diagnoses or clinical messages, its effectiveness is much more modest. In any case, I strongly encourage you to refer to the original Ground Truth publication for more details - https://lnkd.in/dXaEAq6m

21

2 Comments

You try your best to be "patient-centric." You gather patient input, and you work some of the feedback into your protocol. But how do you if it's enough or even appropriate? To help, Crohn's & Colitis Foundation's Carol Moss Chapman and Zachary Palazzotto provide helpful tips and the Foundation's Patient Centricity Program Review. #clinicaltrials #clinicalresearch

7

I like the idea of proactively mapping the bias that will infiltrate your AI product. As a designer I think about how this might be an early activity collaborator done by owners and stakeholder of AI effort. #Aiforall #designAIforhumans

8

1 Comment

The focus at Omada Health is keeping humans at the center of care. We see our members mention this special relationship repeatedly. The latest from Omada proves just that, as the company announced a pledge to Healthcare AI Commitments underscoring its goals of using AI to empower human-led care teams to drive behavior change at scale. This commitment means plans to continue pursuing cutting-edge technologies that enable more personalized and equitable care for members. With billions of actionable health data points to draw on Omada is utilizing AI and machine learning to help scale member/care team interactions. Learn more about this commitment in the latest release: https://lnkd.in/gSCiDgg2

17

AI for Accessibility: Cheese Wheel of Cheeses Based on a request from engineers and artists, I've been building (or have built) a complex alt text generator for large, information-rich visuals. While testing, I've processed thousands of vintage maps, schematics, infographics ... you name it. The absurdly high resolution images can be found at https://buff.ly/4bZiXvf. However, there are a lot of smaller graphics too that don't warrant their own major post. I'm sharing them free as Substack notes - but join the Substack (or Patreon) for a lot more!

2

AI Voice Interactions: From Sci-Fi to Reality - The Future of UX? Don’t think AI is going to drastically change things? Take a quick read/listen to this article by Ethan Mollick and his experience with ChatGPT Voice. The ease of the voice conversation is something right out of Star Trek. Most of us have only experienced mediocre to terrible voice AI.  There doesn’t appear to be any reason something like this won’t replace all the super annoying call center voice prompts at the very least. But it will be exciting to see how it really revolutionizes human computer interaction. Seems likely we will be reimagining UX to include some form of voice interaction in the coming years (months?). #AIVoice #UXDesign #FutureOfTech #InnovationInTech #AIRevolution

5

1 Comment

Generative Ai and Diagnoses I marvel at these studies and the 'academic' views on leveraging Generative AI technologies. “Integration of AI into health care holds great promise as a tool to help medical professionals diagnose patients faster, allowing them to start treatment sooner,” said NLM Acting Director, Stephen Sherry, Ph.D." What's in Generative Ai supported diagnosis? Exploring generative AI tools is a valid pursuit - but alas - in the case of diagnosing a patient's disease the value starts in confirming a diagnosis. And it ends in democratizing the logic of a diagnosis - to a wider range of clinicians. The use of a computer aided diagnosis has been a fertile academic pursuit for over 50 years and has helped to mint many PhDs. In the last 50 years, the accuracy of real world medical diagnosis by clinicians has been studied many times. A doctor who conducts a proper history and a proper physical examination - plus a few standard medical tests - will on average render the correct diagnosis ~89% of the time. By proper - I mean what is taught to doctors in medical school and post graduate studies. But alas - doctors are wrong in their 'decisions' 20% of the time. (Another well documented statistic.) Sadly that is not a great batting average. Doctors are often wrong as - they did not do a proper history or physical exam. The most common reasons are - they don't listen to the patient and they rush through or don't do a proper physical. Alas, doctors as a rule do not fess up on other doctors' averages. So for a patient - the quality of care can be spotty and not transparent. Generative AI depends on accurate and complete data to run through the medical school logic of deducing diagnostic options in a 'closed book' setting. So that means some clinician needs to have entered the proper and complete data that they would have had to gather - had they done the end to end encounter themselves. So what are you left with when using 'Generative Ai' for diagnosis? (For the doctor) If a clinician enters all the data they need AFTER listening to and questioning a patient and enters it (maybe via dictation) along with what they found on a physical exam and lab tests - they should improve their diagnostic average and maybe cover for their 20% error rate. But it all starts and ends in doing what they learned in medical school and picked up 'over the years'. (For the patient) After being subjected to a complete and proper history and physical and having that information entered into a computer - then hopefully the patient will benefit from the delivery of 'proper' care, and not been given the wrong treatment or advice at 1 out of 5 doctor visits. IMHO - until we have robots to do a proper history and physical - conduct simple lab tests (with a drop of blood) we are a ways away from the Generative Ai diagnostic benefit to health care 😀

7

1 Comment

Psssst healthcare folks (and job seekers on the prowl)! AI in healthcare is a hot topic these days—so here’s a quick rundown of the major trends (credit to HLTH USA). ➡️ Imaging & Diagnostics: AI in medical imaging boosts diagnostic speed, accuracy, and efficiency, aiding clinicians in early disease detection and secure workflows. ➡️ Virtual Scribes & Assistance: AI-powered virtual scribes reduce admin tasks, letting providers focus on patients while improving workflow efficiency and documentation. ➡️ Patient Engagement & Care Management: AI enhances patient engagement through personalized care, streamlined communication, and coordinated care management. ➡️ Clinical Decision Support: AI supports data-driven decisions, improving patient care with real-time insights and evidence-based guidance. ➡️ Personalized Care & Workflow Automation: AI personalizes treatment with clinical/genetic insights, optimizing care for complex diseases and improving outcomes. ➡️ Predictive Analytics & RCM: AI-driven analytics enhance healthcare by predicting trends, improving outcomes, and streamlining billing for scalable, value-based care. ➡️ Healthcare Consulting & Administration: AI-powered consulting drives efficiency, innovation, and superior outcomes by transforming healthcare operations.

21

Big healthcare AI news yesterday as Tempus AI received the first-ever CPT code for an algorithm-only analysis of previously-sequenced, laboratory developed test transcriptomic data. TLDR: AI molecular phenotyping has a new business model. The PurISTSM test uses RNA profiling to identify molecular subtypes in patients with advanced pancreatic ductal adenocarcinoma, which may inform first-line therapy selection and management. While I'm skeptical that payer balance sheets are going to solve the "who pays for healthcare AI" problem, the reimbursement pipes for clinical AI are maturing. I am surprised to see more progress for usage-based vs. performance-based models given the many challenges with FFS economics. It will be interesting to see how payers assess the "medical necessity" and therefore reimbursability of clinical AI as these models come to market. https://t.co/eLe18SmihV

418

9 Comments

Are you ushering in the robot apocalypse? 🤖 Alright, maybe it's not that dramatic. But many companies unknowingly cross ethical lines with AI, putting humanity at risk. That's why I wanted to share this free resource from IDEO. IDEO’s cards are a free resource to ensure your innovation never loses its human heart ❤️ Get ready to spark pivotal dialogues and align your team with the ethical side of tech. Grab your toolkit and lead with confidence #humancentereddesigned #humanintheloop #AI #AIethics #freeresource  

104

26 Comments

Data and AI in healthcare… if you can’t map it out on paper, you might not want to light it up with AI just yet… because technology that powerful can accelerate bad processes even more than good ones. So the game plans and the architecture matter. A little good old fashioned human “I” to guide the use of AI can make all the difference. 🌟 Exciting times in healthcare technology! Imprivata’s involvement in Chime’s recent Digital Health Most Wired roundtable highlighted the acceleration of data usage and the significant role of AI in healthcare governance. We’re leading the way, integrating patient data into EHRs and using technology to enhance care. Share our progress and how our innovations are setting industry benchmarks! Learn More: https://ow.ly/pRK230sBzZc 🚀 #HealthcareInnovation #Imprivata

23

💫 My new cover for Nature Portfolio Biomedical Engineering! A New Approach to Treating Ulcerative Colitis 💉 🧫 Ulcerative colitis, a chronic inflammatory disease, is typically treated with medications that suppress the immune system. Kin Man Au, Justin E Wilson, Jenny Ting and Andrew Wang, MD developed a targeted therapy: an injectable bioengineered scaffold placed under the skin. This scaffold creates a specific niche composed of colon-specific cells, support materials and immunomodulatory molecules, reduces intestinal inflammation and promotes colon healing. It also can prevent colitis-associated colorectal cancer and eliminate immune-triggered colitis. On the cover I depicted an injectable subcutaneous niche consisting of colon epithelial cells, decellularized colon extracellular matrix and immunosuppressive nanofibres. Check out the original paper here: https://lnkd.in/eCtRpaBq University of North Carolina at Chapel Hill, Harvard Medical School, UT Southwestern Medical Center, UNC Lineberger Comprehensive Cancer Center #biomedicine #cancer #immunotherapy #cell_niche #sciart #science #scientific_illustration #chemistry #medicine #ellamaru #sciencegirl #sci_artist #sci_design #cover #scientific_cover #medicalartist #phd #phdlife

349

10 Comments

I’ve spoken with 50+ CMIOs / CIOs over the past year on how they’ve piloted new Health Tech for clinicians (e.g. AI scribes, generative AI drafts for patient portal messages, etc.). There’s no one right answer, but there are 5 main approaches I’ve heard for choosing clinicians to pilot with: 𝟭. 𝗠𝗼𝘀𝘁 𝘁𝗲𝗰𝗵 𝘀𝗮𝘃𝘃𝘆 𝗰𝗹𝗶𝗻𝗶𝗰𝗶𝗮𝗻𝘀 You want to move fast so you look for clinicians who raise their hand and are eager to try the solution. The risk? Tech savvy clinicians may be biased in favour of new tech. So they may not be highly representative of the population. 𝟮. 𝗟𝗲𝗮𝘀𝘁 𝘁𝗲𝗰𝗵 𝘀𝗮𝘃𝘃𝘆 𝗰𝗹𝗶𝗻𝗶𝗰𝗶𝗮𝗻𝘀 Or you do the opposite. If the tech is so intuitive, so easy to use that even the least tech savvy folks can and will use it… then you can be confident it can scale enterprise-wide. The risk? Maybe you don’t actually care if the least tech savvy folks can use it, because even 50% adoption is good enough to have a meaningful ROI. E.g. data I’ve seen shows clinician adoption of AI scribes can range 40% to 80% at health systems. But if you piloted with the least likely users (who don’t end up using it)… you would call the pilot a failure and miss out on the up to 80% who would’ve used it. 𝟯. 𝗖𝗹𝗶𝗻𝗶𝗰𝗶𝗮𝗻𝘀 𝘄𝗶𝘁𝗵 𝘁𝗵𝗲 𝗯𝗶𝗴𝗴𝗲𝘀𝘁 𝗽𝗮𝗶𝗻 𝗽𝗼𝗶𝗻𝘁𝘀 If you’re looking to demonstrate the biggest ROI, you could pilot with the clinicians who would benefit the most. For Generative AI drafts for patient portal messages, this is often primary care physicians as they spend more time in their inbox than other specialties. For AI Scribes you could check EHR data to see who is spending the most time on clinical documentation, who is doing the most pajama time in the EHR, etc. The risk? You might end up investing in a solution that works for one clinical group but not for others. 𝟰. 𝗟𝗮𝗿𝗴𝗲𝘀𝘁 𝗴𝗿𝗼𝘂𝗽 𝗼𝗳 𝗰𝗹𝗶𝗻𝗶𝗰𝗶𝗮𝗻𝘀 More data = shorter time to determining value. So it would make sense to pilot with largest specialty with the most clinicians. The risk? Again you would be assuming a solution or approach that works in one specialty is scalable to all… it may not be. 𝟱. 𝗦𝗼𝗺𝗲 𝘂𝗻𝗶𝗾𝘂𝗲 𝗶𝗻𝘀𝗶𝗴𝗵𝘁 𝗮𝗯𝗼𝘂𝘁 𝗽𝗼𝘁𝗲𝗻𝘁𝗶𝗮𝗹 𝘂𝘀𝗲𝗿𝘀 For example some health systems have chosen AI Scribe pilot groups by asking medical office assistants to identify physicians who care the least about how a note is written and thus are more willing to delegate to an AI scribe. The risk? It’s super custom to the situation and it may be hard to replicate this approach with various new tech. 𝙒𝙝𝙖𝙩 𝙖𝙥𝙥𝙧𝙤𝙖𝙘𝙝 𝙬𝙤𝙪𝙡𝙙 𝙮𝙤𝙪𝙧 𝙝𝙚𝙖𝙡𝙩𝙝 𝙨𝙮𝙨𝙩𝙚𝙢 𝙩𝙖𝙠𝙚? #digitalhealth #healthcareinnovation

53

20 Comments

Have you heard of Digital Therapeutics, or #DTx? 🎮 This emerging field is using technology like virtual reality to transform healthcare. As #DTx expert Jillian Ahrens says, "The true power of VR is that it can change somebody's reality instantly." By pulling the patient into an interactive virtual world, you can direct the brain's attention elsewhere & immediately decrease the perceived pain and anxiety. Now imagine the possibilities: relieving pain without medication, accelerating physical therapy, & reducing anxiety during medical procedures. The potential impact is enormous. Gain more insights at: https://lnkd.in/g2KpWT9d #DTx #digitaltherapeutics #VR #healthcareinnovation

14

3 Comments

A regulation is nothing more than entranchment of error into a system. CON laws? Administrative regulatory bodies? Value systems decided in closed door sessions? Healthcare has so much entrenched error it has limited innovation. Is this by design? Wanna see the $2 imaging exam? The $.35 cancer drug…remove entrenched error.

2

The healthcare system is broken, or at least hugely unoptimized, from any reasonable perspective (flatlined outcomes, falling patient/provider satisfaction, rising costs, etc.) The market for new health technology is framed around this flawed system. Almost all health tech innovation funding, private and increasingly public, is steered towards near-term product-market fit. The overall health ecosystem is incentivized to perpetuation decline and dilute transformative innovation. Perhaps it is time for a new approach at a systems level? AI MINDSystems Foundation #health #tech #science #research #outcomes #ecosystem

19

6 Comments