“I worked closely with Naomi at Fern Health on several initiatives, and was continuously impressed by her work and leadership. Naomi's ability to articulate technical decisions through the lens of product benefits and tradeoffs made her a strong communicator across the team, and a pleasure to work with. She brings a deep knowledge of AI/ML to the table. Naomi also doesn't hesitate to ask the tough questions, and has a clear passion for the work she does. Any company would be lucky to have her on board!”
About
Technical leader with over a decade of success in spearheading product management and…
Experience
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Education
Publications
-
Regulation of innate and adaptive lymphocyte responses in metabolic health and disease.
Cell
-
Spatial and temporal mapping of human innate lymphoid cells.
Immunity
-
B2-adrenergic receptor-mediated negative regulation of group 2 innate lymphoid cell responses
Science
See publicationThe type 2 inflammatory response is induced by various environmental and infectious stimuli. Although recent studies identified group 2 innate lymphoid cells (ILC2s) as potent sources of type 2 cytokines, the molecular pathways controlling ILC2 responses are incompletely defined. Here we demonstrate that murine ILC2s express the b2-adrenergic receptor (b2AR) and colocalize with adrenergic neurons in the intestine. b2AR deficiency resulted in exaggerated ILC2 responses and type 2 inflammation in…
The type 2 inflammatory response is induced by various environmental and infectious stimuli. Although recent studies identified group 2 innate lymphoid cells (ILC2s) as potent sources of type 2 cytokines, the molecular pathways controlling ILC2 responses are incompletely defined. Here we demonstrate that murine ILC2s express the b2-adrenergic receptor (b2AR) and colocalize with adrenergic neurons in the intestine. b2AR deficiency resulted in exaggerated ILC2 responses and type 2 inflammation in intestinal and lung tissues. Conversely, b2AR agonist treatment was associated with impaired ILC2 responses and reduced inflammation in vivo. Mechanistically, we demonstrate that the b2AR pathway is a cell-intrinsic negative regulator of ILC2 responses through inhibition of cell proliferation and effector function. Collectively, these data provide the first evidence of a neuronalderived regulatory circuit that limits ILC2-dependent type 2 inflammation.
-
Arginase 1 is an innate lymphoid-cell-intrinsic metabolic checkpoint controlling type 2 inflammation.
Nature Immunology
Group 2 innate lymphoid cells (ILC2s) regulate tissue inflammation and repair after activation by cell-extrinsic factors such as host-derived cytokines. However, the cell-intrinsic metabolic pathways that control ILC2 function are undefined. Here we demonstrate that expression of the enzyme arginase-1 (Arg1) during acute or chronic lung inflammation is a conserved trait of mouse and human ILC2s. Deletion of mouse ILC-intrinsic Arg1 abrogated type 2 lung inflammation by restraining ILC2…
Group 2 innate lymphoid cells (ILC2s) regulate tissue inflammation and repair after activation by cell-extrinsic factors such as host-derived cytokines. However, the cell-intrinsic metabolic pathways that control ILC2 function are undefined. Here we demonstrate that expression of the enzyme arginase-1 (Arg1) during acute or chronic lung inflammation is a conserved trait of mouse and human ILC2s. Deletion of mouse ILC-intrinsic Arg1 abrogated type 2 lung inflammation by restraining ILC2 proliferation and dampening cytokine production. Mechanistically, inhibition of Arg1 enzymatic activity disrupted multiple components of ILC2 metabolic programming by altering arginine catabolism, impairing polyamine biosynthesis and reducing aerobic glycolysis. These data identify Arg1 as a key regulator of ILC2 bioenergetics that controls proliferative capacity and proinflammatory functions promoting type 2 inflammation.
Other authors -
-
Spatial Map of Human T Cell Compartmentalization and Maintenance over Decades of Life
Cell
Mechanisms for human memory T cell differentiation and maintenance have largely been inferred from studies of peripheral blood, though the majority of T cells are found in lymphoid and mucosal sites. We present here a multidimensional, quantitative analysis of human T cell compartmentalization and maintenance over six decades of life in blood, lymphoid, and mucosal tissues obtained from 56 individual organ donors. Our results reveal that the distribution and tissue residence of naive, central…
Mechanisms for human memory T cell differentiation and maintenance have largely been inferred from studies of peripheral blood, though the majority of T cells are found in lymphoid and mucosal sites. We present here a multidimensional, quantitative analysis of human T cell compartmentalization and maintenance over six decades of life in blood, lymphoid, and mucosal tissues obtained from 56 individual organ donors. Our results reveal that the distribution and tissue residence of naive, central, and effector memory, and terminal effector subsets is contingent on both their differentiation state and tissue localization. Moreover, T cell homeostasis driven by cytokine or TCR-mediated signals is different in CD4+ or CD8+ T cell lineages, varies with their differentiation stage and tissue localization, and cannot be inferred from blood. Our data provide an unprecedented spatial and temporal map of human T cell compartmentalization and maintenance, supporting distinct pathways for human T cell fate determination and homeostasis.
Other authors -
-
Human memory T cells: generation, compartmentalization and homeostasis.
Nature Reviews Immunology
Memory T cells constitute the most abundant lymphocyte population in the body for the majority of a person's lifetime; however, our understanding of memory T cell generation, function and maintenance mainly derives from mouse studies, which cannot recapitulate the exposure to multiple pathogens that occurs over many decades in humans. In this Review, we discuss studies focused on human memory T cells that reveal key properties of these cells, including subset heterogeneity and diverse tissue…
Memory T cells constitute the most abundant lymphocyte population in the body for the majority of a person's lifetime; however, our understanding of memory T cell generation, function and maintenance mainly derives from mouse studies, which cannot recapitulate the exposure to multiple pathogens that occurs over many decades in humans. In this Review, we discuss studies focused on human memory T cells that reveal key properties of these cells, including subset heterogeneity and diverse tissue residence in multiple mucosal and lymphoid tissue sites. We also review how the function and the adaptability of human memory T cells depend on spatial and temporal compartmentalization.
Other authors -
-
Distribution and compartmentalization of human circulating and tissue-resident memory T cell subsets.
Immunity
Knowledge of human T cells derives chiefly from studies of peripheral blood, whereas their distribution and function in tissues remains largely unknown. Here, we present a unique analysis of human T cells in lymphoid and mucosal tissues obtained from individual organ donors, revealing tissue-intrinsic compartmentalization of naive, effector, and memory subsets conserved between diverse individuals. Effector memory CD4(+) T cells producing IL-2 predominated in mucosal tissues and accumulated as…
Knowledge of human T cells derives chiefly from studies of peripheral blood, whereas their distribution and function in tissues remains largely unknown. Here, we present a unique analysis of human T cells in lymphoid and mucosal tissues obtained from individual organ donors, revealing tissue-intrinsic compartmentalization of naive, effector, and memory subsets conserved between diverse individuals. Effector memory CD4(+) T cells producing IL-2 predominated in mucosal tissues and accumulated as central memory subsets in lymphoid tissue, whereas CD8(+) T cells were maintained as naive subsets in lymphoid tissues and IFN-γ-producing effector memory CD8(+) T cells in mucosal sites. The T cell activation marker CD69 was constitutively expressed by memory T cells in all tissues, distinguishing them from circulating subsets, with mucosal memory T cells exhibiting additional distinct phenotypic and functional properties. Our results provide an assessment of human T cell compartmentalization as a new baseline for understanding human adaptive immunity.
Other authors -
-
Expansion of HIV-specific T follicular helper cells in chronic HIV infection.
J Clin. Invest.
HIV targets CD4 T cells, which are required for the induction of high-affinity antibody responses and the formation of long-lived B cell memory. The depletion of antigen-specific CD4 T cells during HIV infection is therefore believed to impede the development of protective B cell immunity. Although several different HIV-related B cell dysfunctions have been described, the role of CD4 T follicular helper (TFH) cells in HIV infection remains unknown. Here, we assessed HIV-specific TFH responses…
HIV targets CD4 T cells, which are required for the induction of high-affinity antibody responses and the formation of long-lived B cell memory. The depletion of antigen-specific CD4 T cells during HIV infection is therefore believed to impede the development of protective B cell immunity. Although several different HIV-related B cell dysfunctions have been described, the role of CD4 T follicular helper (TFH) cells in HIV infection remains unknown. Here, we assessed HIV-specific TFH responses in the lymph nodes of treatment-naive and antiretroviral-treated HIV-infected individuals. Strikingly, both the bulk TFH and HIV-specific TFH cell populations were significantly expanded in chronic HIV infection and were highly associated with viremia. In particular, GAG-specific TFH cells were detected at significantly higher levels in the lymph nodes compared with those of GP120-specific TFH cells and showed preferential secretion of the helper cytokine IL-21. In addition, TFH cell expansion was associated with an increase of germinal center B cells and plasma cells as well as IgG1 hypersecretion. Thus, our study suggests that high levels of HIV viremia drive the expansion of TFH cells, which in turn leads to perturbations of B cell differentiation, resulting in dysregulated antibody production.
Other authors -
-
Application of the systematic "DAmP" approach to create a partially defective C. albicans mutant.
Fungal Genet. Biol.
An understanding of gene function often relies upon creating multiple kinds of alleles. Functional analysis in Candida albicans, a major fungal pathogen, has generally included characterization of mutant strains with insertion or deletion alleles and over-expression alleles. Here we use in C. albicans another type of allele that has been employed effectively in the model yeast Saccharomyces cerevisiae, a "Decreased Abundance by mRNA Perturbation" (DAmP) allele (Yan et al., 2008). DAmP alleles…
An understanding of gene function often relies upon creating multiple kinds of alleles. Functional analysis in Candida albicans, a major fungal pathogen, has generally included characterization of mutant strains with insertion or deletion alleles and over-expression alleles. Here we use in C. albicans another type of allele that has been employed effectively in the model yeast Saccharomyces cerevisiae, a "Decreased Abundance by mRNA Perturbation" (DAmP) allele (Yan et al., 2008). DAmP alleles are created systematically through replacement of 30 noncoding regions with nonfunctional heterologous sequences, and thus are broadly applicable. We used a DAmP allele to probe the function of Sun41, a surface protein with roles in cell wall integrity, cell-cell adherence, hyphal formation, and biofilm formation that has been suggested as a possible therapeutic target (Firon et al., 2007; Hiller et al., 2007; Norice et al., 2007). A SUN41-DAmP allele results in approximately 10-fold reduced levels of SUN41 RNA, and yields intermediate phenotypes in most assays. We report that a sun41Δ/Δ mutant is defective in biofilm formation in vivo, and that the SUN41-DAmP allele complements that defect. This finding argues that Sun41 may not be an ideal therapeutic target for biofilm inhibition, since a 90% decrease in activity has little effect on biofilm formation in vivo. We anticipate that DAmP alleles of C. albicans genes will be informative for analysis of other prospective drug targets, including essential genes.
Other authors -
Patents
-
CANCER-SPECIFIC MOLECULES AND METHODS OF USE THEREOF
Filed US WO2020210537
See patentDisclosed herein are systems and methods for identifying candidate targets that may be used for therapeutic developments. The systems and methods may comprise receiving and analyzing biologically relevant data to identify information or events such as splicing events that may be statistically significant or specific to certain diseases, illnesses or conditions. Also provided are systems and methods for modulating the statistically significant events using compositions or molecules including…
Disclosed herein are systems and methods for identifying candidate targets that may be used for therapeutic developments. The systems and methods may comprise receiving and analyzing biologically relevant data to identify information or events such as splicing events that may be statistically significant or specific to certain diseases, illnesses or conditions. Also provided are systems and methods for modulating the statistically significant events using compositions or molecules including small molecule compounds, oligonucleotides, proteins or cells.
-
Neoantigens, Methods of Detection and Use Thereof
Filed US 62/896230
Honors & Awards
-
Ruth L. Kirschstein National Service Award
NIH
-
NIH DP5 Early Independence Award Nominee
Weill Cornell Medicine
Selected as the institution-wide NIH DP5 Nominee by Weill Cornell
-
Richard C. Parker Memorial Award
Columbia University
The Richard C. Parker Memorial Award recognizes a student who has accomplished outstanding and innovative work during their graduate training.
-
Research Honors
Carnegie Mellon University
Undergraduate honors thesis
-
University Honors
Carnegie Mellon University
Summa cum laude
Languages
-
Russian
Native or bilingual proficiency
-
English
Native or bilingual proficiency
Organizations
-
AdvaMed
Executive Leadership Group Member
- Present -
NIH SBIR/STTR
Chartered Review Group Member - Applied AI/ML
- Present -
American Association of Immunologists
-
- Present
Recommendations received
3 people have recommended Naomi
Join now to viewOther similar profiles
Explore more posts
-
Morgan Cheatham
One key reason why healthcare technologies like AI, telemedicine, and software haven't structurally lowered costs or broadened access for the entire system after decades of experimentation (excluding pure financial engineering of payer contracts) is the limited reimagining of care delivery itself, with asynchronous telemedicine as a notable exception. In the same way that digitizing physical paper failed to exploit the capabilities of a new computing medium for word processing, hosting physician-patient conversations over video does not transform the actual delivery of care. In telemedicine, many of the inputs (e.g., clinician time) and therefore cost structures of in-person care persist, which is why we often see fee-for-service telemedicine companies struggle to achieve economies of scale (i.e., growing revenues non-linearly while growing costs linearly). Until we deconstruct the components of a medical appointment – from intake and triage, to chart review, history gathering, physical examination, diagnosis, treatment, and follow-up – we will continue to be underwhelmed by the disconnect between the transformative potential of these technologies and real-world patient/system impact. I’m optimistic that the emergence of reliable and performant medical AI agents will urge us to perform this deconstruction and redesign the fundamental building blocks of a “medical appointment.”
38 Comments -
Rishad Usmani, MD
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
8 Comments -
Joshua Liu
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
21 Comments -
Joshua Liu
A health system asked if I could share what SeamlessMD’s secret sauce was for engaging patients… so I’ve decided to open source my response here. They wanted to know what principles and success factors have enabled our customer partners to achieve industry-leading results like: ⬇️ Cost of care by $1,000+ per patient ⬆️ Increased net profit of $1,000+ per patient ⬇️ 1 to 2 days length of stay ⬇️ 89% readmissions reduction by up to 89% ⬇️ 68% ED visit reduction by up to 68% ⬇️ 89% mortality ⬇️ 65% phone calls to providers Here are the main principles and success factors I shared: 1️⃣ It starts with a high touch partnership to ensure success. The People + Process we assign to a health system partnership matter even more than the Technology 2️⃣ Clinical expertise applied to the product matters more than Tech features. Our product’s secret sauce is our world class, pre-built care plan content. 3️⃣ Key elements of our care plan content that drive high patient engagement include: - Brand personalization - Condition-specific patient experiences - Personalization to a patient’s demographics & comorbidities - Condition-specific health status & symptom tracking - Automated self-care guidance for 80%+ of patient issues 4️⃣ Obsession with patient accessibility further optimizes patient engagement. - Gr. 6 reading level - Colour-blind tested - Screenreader friendly - WCAG 2.0 Level AA requirements met and more 5️⃣ Digital care plan implementation is Clinical Change Management. A successful implementation requires a proven playbook and hands-on expertise for: - Achieving care pathway standardization and alignment - while being efficient with clinician time commitment - Embedding digital care journeys into the clinical workflow: What workflows make the most sense for receiving alerts and monitoring dashboards given our limited staff resources? - Having the clinical acumen to going deep into the nitty gritty clinical workflows. E.g. for a pregnancy care journey - how will our solution know if there is an infant loss? How should the care plan change in that case? 6️⃣ Clinicians must be engaged to have a sense of ownership so they promote it to patients We’ve figured ou the optimal time commitment that ensures all clinicians are engaged during implementation (hint: physicians have very limited time!) 7️⃣ Let busy clinicians act as the Editors of Clinical Content. By having a clinical Patient Education team act as proactive Ghost Writers. - The key principles are to (1) earn clinician trust and (2) reduce friction to clinician engagement. 8️⃣ The ideal approach to patient enrollment varies from health system to health system. Sometimes a hybrid approach might even be needed 9️⃣ You must believe that the job’s not done until you achieve business outcomes. We obsess over the data and push our partners to measure ROI. More details in the slides below 👇 *** I’m curious: Which of these resonated most?
7 Comments -
Bobby Guelich
Here’s your recap of last week’s health IT news 🗞️ 👇 🤖 AI Clinician Assistant • Suki is partnering with Zoom to generate clinical notes as well as other integrated AI features. • Net Health announced it would integrate AI ambient scribe Tali AI. • Aidoc is partnering with NVIDIA to create a guideline to accelerate AI adoption, the Blueprint for Resilient Integration and Deployment of Guided Excellence (BRIDGE). 📞 AI Contact Center • Authenticx released a new AI assistant that helps users uncover business insights from their data. • Infinitus Systems, Inc. closed a $51.5M Series C with participation from Memorial Hermann Health System and others. 🏥 AI Facility Management • Cognosos, Inc. acquired Cox Prosight. 🌡️ AI Remote Patient Monitoring and Triage • Biofourmis has merged with CopilotIQ. 📈 Data & Analytics • Avandra Imaging launched out of stealth; it plans to sell to biopharma, medical research, and AI companies. • Gradient Health acquired DataAppraisal. • HealthEx launched and announced $14M in Seed and Series A funding. 📋 Patient Administration • Notable released Flow Builder, which allows users to develop their own AI agents for operational workflows within EHRs and other systems. • MIla Health Inc released its flagship AI solution of the same name, and announced implementations at provider organizations including Aayu Clinic, MIMIT Health, and Best Practices Inpatient Care. 🔁 RCM • Oracle’s patient accounting solution has been integrated with FinThrive, which enables improved functionality, performance, and a unified login experience. • Blue Shield of California and Salesforce are partnering on a solution that promises to give patients prior auth responses within the course of their appointment. ⚙️ Clinical Operations • Dyania Health closed a $10M Series A, including participation from Cleveland Clinic Ventures. 📱 Patient Support & Communications • Psych Hub released a new solution designed to support patients seeking mental health support, directing them to both non-clinical and clinical resources, thereby assisting behavioral health providers to reach new patients. • Mayo Clinic has partnered with Lin Health to support patients after treatment for fibromyalgia. 🍎 VBC • WellSky acquired Bonafide Health, allowing it to provide a more integrated platform for providers to run their businesses. 🧠 Food For Thought • 𝗦𝘁𝗿𝘂𝗴𝗴𝗹𝗶𝗻𝗴 𝘁𝗼 𝗸𝗲𝗲𝗽 𝘂𝗽 𝘄𝗶𝘁𝗵 𝗮𝗹𝗹 𝘁𝗵𝗲 𝗔𝗜 𝗴𝗼𝘃𝗲𝗿𝗻𝗮𝗻𝗰𝗲 𝗳𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸𝘀? Andrew Hines, CTO at Canvas Medical, shared a breakdown (plus some hot takes) on all of the many entities currently vying for market share in the AI governance space https://lnkd.in/eMNpXw4t --- Want the latest healthcare tech news delivered to your inbox each week? Sign up here 📬 👉 https://lnkd.in/e2UJ3ckP
4 Comments -
Samantha C. Register, MBA
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.
-
Morgan Cheatham
As the discourse shifts from models to compound AI systems / agents, we need better AI benchmarks to evaluate multi-modal and multi-step task performance, especially in healthcare and life sciences. When we wrote the first paper demonstrating ChatGPT's performance on the USMLE, we chose the US Medical Licensing Exam as a benchmark for accessibility, speed, and ease. This benchmark was never intended to represent AI model performance on real-world clinical tasks. Today, I still see so many research teams and startups using benchmarks (like the USMLE) that are ill-fitted for assessing the true clinical or scientific performance and utility of the models they are developing for real-world contexts. Benchmark development may be seen as a "less sexy" area of research, but it is of paramount importance. Years after the rise of the transformer, we still lack adequate benchmarks for so many single-step tasks in biomedicine. With compound AI systems (i.e., architectures that integrate multiple AI models to perform complex tasks) emerging, we need new benchmarks for agentic behaviors. I'd argue that developing an agent with novel capabilities without at least proposing a companion benchmark (if an industry standard does not yet exist) may hinder the adoption of said agent, especially for high-stakes workflows. Designing more benchmarks that capture/simulate real-world clinical and scientific workflows will help us mitigate the major discrepancies observed between in silico and in vivo performance and better support safe + effective deployment of AI in biomedicine. There are already brilliant people focused here, and we need more. DMs are open if you're researching or working in this area of multi-step/multi-modal benchmarking in healthcare and life sciences! #healthcare #ai #artificialintelligence #generativeai
18 Comments -
Bobby Guelich
Here’s your recap of last week’s health IT news 🗞️ 👇 🤖 AI Clinician Assistant • Avaamo launched an AI ambient scribe solution with support for over 50 specialties with pre-trained models. • Wound AI, an AI clinical decision support solution built by VANTIQ, Telemedicine Solutions, LLC, and NTT DATA, launched and is now available 📞 AI Contact Center • Premier Inc. awarded a group purchasing agreement to TeleVox Healthcare. 🏥 AI Facility Management • Medely has integrated with Oracle, allowing users to leverage Oracle’s Cloud HCM capabilities. 🌡️ AI Remote Patient Monitoring and Triage • The Alliance for Smart Healthcare Excellence—established by care.ai, now a part of Stryker, and developed with assistance from CHIME—was established to promote the Smart Hospital Maturity Model (SHMM). 📈 Data & Analytics • Epic expanded its set of APIs available to developers to support USCDI v3. • Health Catalyst will acquire Intraprise Health, a cybersecurity provider; it was also selected for analytics and interoperability by CyncHealth, Iowa’s health information exchange (HIE). • Health Commons Project acquired OneHealthPort • Socially Determined released a new application that helps organizations access data for minors. 🧾 Orders & Fulfillment • Surescripts is expanding its Medication History for Populations product to health plans, enabling care management teams to identify gaps in care and improve outcomes. 🧑💻 Virtual & At-Home Care • Mass General Brigham launched a partnership with the New England Center and Home for Veterans to provide hospital at home services for veterans experiencing homelessness. • MedStar Health expanded its partnership with DispatchHealth to support care transitions. 🍎 VBC • i2i Population Health is partnering with CureMD to surface actionable data to providers at the point of care. 🧠 Food For Thought • Brendan Keeler shared an interesting post differentiating AI solutions across augmentative vs. autonomous and then further breaking down each. One take we found interesting was, “Active interfaces are fun, but ultimately have some of the same problems as Siri - they require explicit engagement and cognitive overhead to use. This means their adoption and impact often fall short of expectations - users must actively remember to use them and overcome the friction of engagement.” https://lnkd.in/eR39qFpK --- Want the latest healthcare tech news delivered to your inbox each week? Sign up here 📬 👉 https://lnkd.in/e2UJ3ckP
-
Bobby Guelich
Here’s your recap of last week’s health IT news 🗞️ 👇 🥼AI Clinician Assistant • Suki : The AI scribe announced an SDK and APIs enabling developers to integrate ambient documentation into their applications. • Apricot: The gen AI pre-seed startup focused on reducing documentation overhead for home health nurses launched. • Tampa General Hospital + Palantir Technologies: The health system and the software platform are partnering to develop an AI care coordination and decision support platform. ☎️ AI Contact Center • Mediktor + Sensely: The AI symptom triage solution acquired the conversational member engagement platform. 📋 Patient Administration • Oracle (Cerner) + Loyal: The EHR now offers integration with the patient engagement solution through its Healthcare Marketplace. • HubSpot: The CRM and marketing platform announced it will now offer HIPAA support and sensitive data tools. 🏥 Provider and Practice Administration • LeanTaaS: The AI-enabled facility management solution released updates including: mobile-first live view of OR schedule, predictive staffing automation, and a gen AI solution allowing decision-makers to glean staffing and scheduling insights. ⚕️ Clinical Operations • Keragon: The HIPAA-compliant automation platform launched out of stealth with a $3M funding round. 🧑💻 Virtual and At-Home Care • AvaSure + CLEW: The virtual care platform is partnering with the AI-powered remote patient monitoring and analytics company to improve patient outcomes. 🍎 VBC • Evolent + Machinify, Inc.: The VBC org acquired AI Authorization assets from the AI operating system platform. • ProgenyHealth, LLC + Wildflower Health: The maternity and NICU case management solution is partnering with the women’s health VBC enabler to “help health plans fill gaps, collapse silos and integrate into existing clinical and operational workflows.” • Health Catalyst + Carevīve: The population health data and analytics company has completed acquisition of the cancer care management solution. 🧠 Food for Thought: • This article provided an interesting look at Nebraska Medicine’s efforts to design a next-generation hospital room, including the four technologies they’ve selected to anchor their work. https://lnkd.in/em-u-a5v • McKinsey released a survey of 200 global health system executives on their investment priorities. Among the notable findings, 88% listed AI as an area of high potential impact, yet only 40% have implemented any solution thus far and over 20% don’t have plans to invest in the next two years. https://lnkd.in/eK3CeqQ2
4 Comments -
Nick Tarazona, MD
👉🏼 ChatGPT's Accuracy in Triage of Nephrology Patient Communications 🤓 Justin H Pham 👇🏻 https://lnkd.in/eYkqgU4h 🔍 Focus on data insights: - 📈 In the first trial, ChatGPT correctly triaged 93% of messages, demonstrating strong reliability. - 🔍 The AI system maintained consistency across trials, with a similar accuracy rate in both assessments. - ⚖️ Internal agreement between ChatGPT's responses was high, indicating robust performance and minimal variability in results. 💡 Main outcomes and implications: - 🤖 High accuracy in triaging suggests that AI could be effectively integrated into outpatient settings to streamline patient communication. - ⏱️ Potential reduction in response times for patient messages, leading to improved patient care and operational efficiency. - 🏥 Encourages further exploration of AI applications in healthcare for enhanced decision-making and resource management. 📚 Field significance: - 🔬 Highlights the role of artificial intelligence in transforming patient care pathways within nephrology and beyond. - 🌐 Sets a precedent for future studies assessing AI capabilities in various medical specialties. - 📊 Supports the development of AI-driven tools aimed at optimizing triage processes and improving clinical outcomes. 🗄️: [#AI #Triage #Nephrology #PatientCare #HealthcareInnovation #ChatGPT #DataInsights #Efficiency]
-
Joshua Liu
I’ve talked to many CMIOs and CIOs the past year. Wanna know the Number 1 mistake they’ve made when choosing a Digital Health solution? Evaluating based primarily on the Tech features (oh wow shiny features!), and not doing real due diligence on the Clinical Content & Workflows that come with the actual product. Here are the types of frustrations they tell me: → “The vendor said they had 100 out-of-the-box care plans… but the content is incredibly bare bones. My clinicians end up creating them from scratch.” → “Our clinicians are demanding a point solution now (e.g. for Ortho) because the clinical content in this supposed enterprise platform aren’t good enough.” → “We need content for this new use case, and the vendor has no interest in growing their content library.” → “The vendor says they’re willing to develop the content, but it’s going to take a whole year. That’s a LONG time.” The reality is Digital Health Tech is only as good as the Clinical Content and Workflows you put in them. The parallel reality is clinicians do not have hours and hours of time to create content, workflows and algorithms from scratch. They are happy to edit, but they have zero time to build from scratch. But what CMIOs and CIOs cannot do is simply trust the quality of clinical content a Digital Health vendor claims to have. Because unless you’ve practiced in that specialty, you don’t truly know if the content is good enough. So what do you do as a CMIO and CIO? You get your clinicians to weigh-on on the product you’re considering. Bring your clinicians to a product demo, and let them due diligence the quality of content, workflows, etc. for their own specialty. E.g. let the Orthopedic surgeon leaders diligence the quality of the Orthopedic content. They can tell you pretty quickly if the content is great or poor. If it meets their needs or not. They’ll ask very clinical questions, and tell right away if the vendor has actual clinical subject matter expertise or if they were talking out of their butt the whole time. Save yourself, your clinicians and your health system a ton of headache. P.S. If you’ve never heard of SeamlessMD, we provide digital care journeys for health systems to guide patients across pre/post surgery, oncology, women’s health, chronic care and more. We ACTUALLY have a library of customizable, evidence-based care plans that cover 150+ procedures and conditions, backed by 40+ studies and evaluations showing improved cost, length of stay, readmissions, ER visits, mortality and more. So if you’re a health system leader who is struggling with Digital Health vendors who don’t actually provide the Clinical Content and Workflows to make the Tech work… give me a shout.
27 Comments -
Avinash Jayaswal, MD, MS
This truly highlights the tension between innovation and market dominance in health tech. Particle Health's challenge against Epic could set a significant precedent for data interoperability and patient rights. As regulatory scrutiny on monopolistic practices intensifies, it will be interesting to see how this unfolds.
-
Nick Tarazona, MD
👉🏼 ChatGPT Responses to Frequently Asked Questions on Ménière's Disease: A Comparison to Clinical Practice Guideline Answers 🤓 Rebecca A Ho 👇🏻 https://lnkd.in/e7Nvpwzw 🔍 Focus on data insights: - ChatGPT-4.0 had significantly better resource reporting compared to ChatGPT-3.5. - Both models showed room for improvement in comprehensiveness, readability, and accuracy. 💡 Main outcomes and implications: - ChatGPT-4.0 demonstrated higher quality resources in responses. - Readability scores for both models fell below the recommended level. - Improvement needed in providing more comprehensive and less misleading information to patients. 📚 Field significance: - Evaluating AI models like ChatGPT against clinical guidelines can help identify areas for enhancement in patient education. - Enhancing AI models' readability and accuracy is crucial for effective patient communication in healthcare settings. 🗄️: [#Ménière'sDisease #AIModelEvaluation #PatientEducation #HealthcareCommunication]
-
Joe Warbington 📊☁️
My colleague Rafa Arranz ❄️ shows how to talk to your clinical data safely, securely, and without hallucinations using Snowflake Cortex Analyst. I'll always have a spot in my heart for data visualizations, but directly asking for insights from vast data is where the world is today thanks to ChatGPT. #Snowflake #AIDataCloud #snowflakeCortex #healthcare #healthdata #healthtech
2 Comments -
Nick Tarazona, MD
👉🏼 ChatGPT compared to national guidelines for management of ovarian cancer: Did ChatGPT get it right? - A Memorial Sloan Kettering Cancer Center Team Ovary study 🤓 Lindsey Finch 👇🏻 https://lnkd.in/esWx_ZsJ 🔍 Focus on data insights: - 📊 64% of unprompted GPT responses were accurate and complete, indicating strong performance in comparison to NCCN guidelines. - 🔍 GPT-4 outperformed NCCN in areas such as risk factors, surgical management, and surveillance, showcasing its potential utility in these domains. - ⚠️ Inaccuracies were noted in medical management questions, highlighting the need for caution in relying solely on AI for clinical decision-making. 💡 Main outcomes and implications: - 🏥 The study suggests that while GPT-4 can provide valuable insights, it should not replace expert clinical judgment, especially in complex cases. - 📈 The findings indicate a growing role for AI in oncology, but emphasize the importance of supervised use to mitigate risks associated with inaccuracies. - 🔄 Continuous evaluation and improvement of AI models are necessary to enhance their reliability in clinical settings. 📚 Field significance: - 🌐 This research contributes to the ongoing discourse on integrating AI into healthcare, particularly in oncology, where timely and accurate information is critical. - 🧠 It underscores the potential of AI tools to assist healthcare professionals, while also calling for rigorous validation against established guidelines. - 🔗 The study paves the way for future investigations into the effectiveness of AI in various medical specialties, promoting a balanced approach to technology adoption. 🗄️: [#ovariancancer #AIinHealthcare #ChatGPT #NCCNGuidelines #oncology #medicalmanagement #surgicalmanagement #riskfactors #healthcareinnovation]
-
Mario Amaro, MD
We have yet to see any AI-native hardware in healthcare for the direct care market and this IMO is the biggest opportunity right now. 1. helps patients with care management + provides access to health information anywhere anytime 2. adds agentic care models on-device for clinical use (doctors) and personal use (patients) 3. eliminates expensive admin/ops barriers that have been sidelining new entrants from moving into independent practice AI-native hardware (e.g. AI-powered wearables) will create an entirely new category of care delivery in 2025. #HealthCareOnLinkedIn #PrivatePractice #SeniorCare Cline cc Nikhil Basu Trivedi
2 Comments -
Andy Mychkovsky
Digital health, we need to talk about your ROI claims & reducing overall healthcare costs. Companies often target a specific clinical subset, but then make claims that $x of spending towards the digital health solution yields $x+y savings in overall healthcare spend. When you peel back the onion, a lot of the overall healthcare "savings" is generated by a supposed reduction in emergency department or hospital admissions. Below is a visual breakdown of 2018 Nationwide Emergency Department (ED) Sample (NEDS) that shows the most common reasons for ED visits. For an employer-sponsored population (non-Medicare, non-Medicaid), the reasons for going to the ED are vast and acute. Retrospectively, you can always pick two periods in time where spending was higher at start and lower at the end (i.e., producing savings). But that really doesn't prove it's your instrument or company that generated the savings. Rather, many patients are poly-chronic and/or subject the randomness of disease/injury. Multiple companies treating the same employee may even use the same savings estimates... More sophisticated buyers don't believe you can reliably produce the claims some of your sales team's are making. It's not your fault the savings are difficult to produce, humans just are complicated creatures. I get your sales and marketing teams need to beat up each other, but let's be realistic. Now I do think site of care savings for chronic treatment of disease is the most reliable way to produce savings, but it’s only applicable to those receiving expensive hospital-based care. ---- credit: AHRQ Healthcare Cost and Utilization Project
22 Comments -
Nick Tarazona, MD
👉🏼 Navigating Nephrology's Decline Through a GPT-4 Analysis of Internal Medicine Specialties in the United States: Qualitative Study 🤓 Jing Miao 👇🏻 https://lnkd.in/egMP7CfV 🔍 Focus on data insights: - 📉 Declining interest in nephrology shows an 11% drop in candidates for the 2024 fellowship match. - 🏥 Only 66% of nephrology positions were filled, highlighting a significant gap compared to other specialties. - ⚖️ Evaluation criteria such as work-life balance and career demand scored lower for nephrology, contributing to its declining appeal. 💡 Main outcomes and implications: - 📊 Nephrology ranks low among internal medicine specialties, only above sleep medicine, suggesting a reevaluation of its status in the field. - 🔍 Key areas of concern include intellectual challenge, procedural involvement, and financial compensation, which may deter prospective candidates. - ⚠️ The trend emphasizes a potential physician shortage, necessitating strategies to enhance nephrology's attractiveness to medical residents. 📚 Field significance: - 🔄 The findings raise crucial alarms about the future of nephrology as a specialty, particularly in light of an overall shortage of physicians. - 🧠 It suggests a need for targeted interventions to improve perceptions and realities of nephrology practice. - ❓ Understanding factors that influence specialty choice can guide educational and policy reforms to address this decline. 🗄️: [#nephrology #internalmedicine #fellowshipmatch #physicianshortage #careerchoice #GPT4 #dataanalysis #medicaleducation]
-
Jasmeet Anand
Interesting piece by #JAMA on the FDA's perspective on #AI in healthcare and biomedicine. It highlights how AI presents challenges but also great opportunities to impact health outcomes. The opportunities are immense as it relates to improving clinical research, drug discovery, drug development, clinical trial support, post-marketing surveillance, etc. What is essential however is the need to have optimized tools, regulation and established frameworks to ensure that AI models are safe and effective across diverse settings. This is an exciting time for ongoing AI innovation!
-
Jan Beger
This paper discusses the need for regulatory and reimbursement frameworks that support the flexible interactions and groupings of digital health technologies (DHTs) to enhance their application in clinical settings. 1️⃣ DHTs are increasingly being grouped into packages for specific clinical tasks, creating new paradigms in medical treatment. 2️⃣ Current frameworks are insufficient for the complex interactions within DHT suites, necessitating updated regulations. 3️⃣ DHT suites refer to a collection of interconnected digital health devices and applications that are grouped together to perform specific clinical tasks in defined settings. These suites leverage the combined functionalities of individual devices to deliver comprehensive healthcare solutions. Examples of DHTs include wearables, mobile health devices, telehealth platforms, and health information technology systems. 4️⃣ The pandemic accelerated the adoption and temporary regulatory changes for DHTs, highlighting the need for permanent frameworks. 5️⃣ There are significant variations in how different countries perceive, evaluate, and reimburse DHTs. 6️⃣ Ensuring interoperability and addressing cybersecurity concerns are major challenges for the adoption of DHT suites. 7️⃣ Assessing the cost-effectiveness and gathering evidence for DHT suites introduces new complexities, especially regarding initial investments. 8️⃣ Future Directions: The article calls for new regulatory and HTA frameworks to support the innovative use of DHT suites in healthcare. ✍🏻 Rebecca Mathias, Peter McCulloch, Anastasia Chalkidou, Stephen Gilbert. Digital health technologies need regulation and reimbursement that enable flexible interactions and groupings. npj Digital Medicine. June 2024. DOI: 10.1038/s41746-024-01147-z
14 Comments
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More