Bayesian Health’s AI platform seeks to make the EHR more dynamic and predictive. READ MORE
Bayesian Health, a startup created by a machine learning researcher at Johns Hopkins, is building models for sepsis detection, patient deterioration and bedsores. The company recently emerged from stealth mode with backing from Andreessen Horowitz. READ MORE
Though predictive tools are widely used in the healthcare setting, there are no official rubrics or guidelines for evaluation or requirements for consistent regulatory oversight. But, what can often happen during deployment is something called “dataset shift.” Dataset shift occurs when a tool “underperforms because of a mismatch between the data set with which it was developed and the data on which it is deployed.” When dataset shift happens, the model can be considered “unsafe.”
As a result, it’s up to the predictive tool’s developers and implementers to monitor and evaluate the solution in order to ensure model safety, sensitivity and precision. Additionally, the tool’s users—clinicians—are a critical component of evaluation, as they are the individuals who can detect or see any dataset shift issues first.
Though evaluating for dataset shift can prove challenging, as it can happen for a variety of reasons (i.e. there can be population changes, physician behavior changes, etc.). It is essential to evaluate any data-driven predictive solution for dataset shift, in order to have safe and robust predictive tools at the bedside.
Bayesian Health’s CEO, along with other experts in the field, just published a letter in the New England Journal of Medicine (NEJM) outlining the causes of dataset shift and how to successfully recognize and mitigate dataset shift through both end-users and AI governance teams.
Bayesian Health’s predictive AI platform integrates the guidelines and rubric laid out in the NEJM, to monitor for dataset shift and ensure model safety. We believe this type of evaluation is an essential part of any AI tool, and this type of ongoing monitoring is how Bayesian’s platform is able to be deployed in a variety of populations/demographics, and adapt to differing or new physician and care team behaviors and practice patterns.
You can read the full NEJM article here, and you can learn more about the AI/machine learning strategies we use to overcome common hurdles faced by many in the field, here.
Bayesian Health, an AI and machine learning platform, launched Monday with an accompanying study revealing faster recognition and treatment times for patients. READ MORE
The formula for launching a machine learning company in health care looks something like this: Build a model, test it on historical patient data in a computer lab, and then start selling it to hospitals nationwide. READ MORE
Platform shows high sensitivity, precision and physician adoption; recent study demonstrates 1.85h earlier recognition and treatment of sepsis, driven by 89% physician adoption
NEW YORK, JULY 12, 2021—Bayesian Health, an AI-based clinical decision support platform enabling health systems to provide safer and higher quality care, today launched its solution to the commercial market. Serving health systems on leading electronic medical record (EMR) vendors, Bayesian’s platform makes the EMR proactive—dynamic and predictive—to catch life threatening disease complications early.
Bayesian’s AI platform sits within the EMR, analyzing patient data with industry-leading AI/machine learning models. The platform sends accurate and actionable clinical signals within existing workflows when a critical moment is detected, helping physicians and care team members accurately diagnose, intervene, and deliver timely care.
With a research-first foundation of over 21 patents and peer-reviewed research papers, Bayesian’s platform is based on technology licensed from the Johns Hopkins University. The platform is configured to scale within health systems, and targets high-priority areas with specific modules such as clinical deterioration, sepsis, pressure injury, and transitions of care.
“Creating an AI solution that can achieve high sensitivity and precision isn’t easy,” said Doug Given, MD, PhD, Managing Partner at Health2047 Capital Partners and Bayesian Health investor. “Bayesian’s platform was developed using the highest quality data, and tested by top physicians and clinicians. This, combined with the team’s technical expertise and research-first mentality, has allowed them to create an AI platform that can be applied to many safety and quality endpoints to transform care delivery in the hospital setting for the better.”
A recent large, five site study analyzing use and practice impact over two years for Bayesian’s sepsis module showed the platform drove 1.85 hour faster antibiotic treatment for sepsis where timely treatment directly impacts mortality rate. The platform also demonstrated high, sustained adoption by physicians and nurses (89% adoption), driven by the sensitivity and precision of the insights and user experience of the software.
Additionally, a single-site study showed a 14% reduction in ICU admissions, a 12% reduction in ICU length of stay, and a 10% reduction in need of supportive therapies. These improvements resulted in a $2.5M annualized benefit for the 250 bed study site hospital, from decreased ICU utilization, earlier accurate diagnoses, and fewer hospital acquired conditions. Further studies validating outcomes including measuring mortality and length of stay reduction will be completed later this year.
With an estimated 400,000 preventable deaths a year, costing over $17B, patients, physicians, and health systems are all suffering as a result of reactive care. Patients are dying from preventable health events. Physicians and care team members are at risk of burnout with data overload and increased caseloads. Health systems are dealing with tighter staffing ratios, reduced operating margins, and increased value-based contracts.
“The number of data points being generated every single day for any given patient in a hospital is enormous, and continues to rise. And yet most physicians have never interacted with any AI technology that actually helps them analyze these data, or gives them any clinical insights,” said Dr. Vineeta Agarwala MD PhD, an internal medicine physician and General Partner at Andreessen Horowitz. “Early results from the real-world use of Bayesian Health’s platform is showing us that it can fit into our workflow, and augment how clinicians triage and diagnose patients.”
Historically, clinical decision support tools have been associated with high alert fatigue and have failed to build trust with physicians. Using cutting edge AI/ML strategies such as a wait and watch strategy and real-time feedback loops to increase precision, and strategies to make the models stronger, Bayesian’s technology accuracy is 10x higher than other solutions in the marketplace. Further, context and transparency with each clinical signal builds confidence among experts in the user experience.
“Having spent over two decades in AI and machine learning research, I know there’s immense potential to create AI tools to drive better care outcomes,” said Suchi Saria, PhD, founder and CEO of Bayesian Health. “Health data are messy, and it requires deep AI expertise to deliver strategies that can successfully analyze this data. But what’s even harder is what happens after the model is created; even with great models, you still need the solution to be adopted and trusted to realize better outcomes. We’re doing it differently, being one of the first solutions to deliver accurate and actionable clinical signals that physicians and nurses are actually acting upon.”
Bayesian Health and Dr. Saria have been highlighted by press, such as Bloomberg News and PBS, for the development of the platform, and has won several awards for excellence in care delivery, including The Armstrong Award for Excellence in Quality and Safety at Johns Hopkins Medicine; Node.Health’s Best in Class Digital Health Intervention Award; and Society of Critical Care Medicine’s Annual Scientific Award.
Advisory and investor team members to the company include Julie Yoo, General Partner at Andreessen Horowitz and Bayesian Health Board Member; Tasso Argyros, Founder & CEO, ActionIQ and Bayesian Health Board Member; Vijay Pande, PhD, General Partner, Andreessen Horowitz; Doug Given, MD, PhD, Managing Partner at Health2047 Capital Partners; and R. Jacob Vogelstein, PhD, Co-Founder & Managing Partner of Catalio Capital Management, LP.
Bayesian Health has raised $15 million in venture funding led by Andreessen Horowitz. Health 2047 Capital Partners, Lifeforce Capital, and Catalio Capital Management, LP also participated in the round of funding.
Bayesian Health is on a mission to make healthcare proactive by empowering physicians with real-time data to save lives. Just like the best physicians continually incorporate new data to refine their prognostication of what’s going on with a patient, Bayesian Health’s research-backed AI platform integrates every piece of available data to equip physicians with accurate and actionable clinical signals that empower them to accurately diagnose, intervene, and deliver proactive, higher quality care.
Why I’m bringing AI to the bedside
Several years ago, I lost my nephew to sepsis — he was only 26 years old when this happened. It was devastating for my family. Sepsis is a life-threatening syndrome that is preventable only when it’s recognized early and treated in a timely way. Unfortunately, in my nephew’s case, it was recognized once his system was already in septic shock, a state where it’s much harder to rescue the patient.
I think about my nephew a lot. Because I know that everyone has a story like mine. With over 200,000 preventable deaths a year in the US, it’s incredibly likely that you, your neighbor or friend has lost someone dear to them at the hands of a preventable condition.
When my nephew was diagnosed with sepsis, my family called on me as I was the “sepsis expert.” Having spent over 5 years researching and creating AI / machine learning strategies to find early warning signs of life threatening conditions, focusing on sepsis, it made sense that I should have the answer. But I realized that my research was just that — it was research. It wasn’t impacting real world outcomes yet, but I knew it could transform the way care is delivered in the in-patient setting and save thousands of lives.
Right now, care is reactive, leading to conditions like sepsis being diagnosed too late. But it’s not the fault of our clinicians. Our clinicians are working their hardest but our system is not gearing them up for success. It’s not surprising that conditions like sepsis are missed. It’s hard to be by the patient’s bedside 24/7. Early signs can be subtle and providers rarely have time to click through dozens of screens culling data from this encounter and past encounters to determine patient specific baselines and hunt for subtle deviations that may be anomalous.
But, now is also a time when we can be building our systems for success. Interoperability rules are making it possible to stitch together a comprehensive longitudinal patient view. High quality artificial intelligence and machine learning driven systems can scan, analyze, and help focus attention to not miss critical events. The EMR provides a digital layer that spans nearly every care delivery setting and provides the infrastructure layer within which we can make care real-time, dynamic, patient and provider-context aware.
Done right, care can be better, more proactive, and safer while reducing provider burnout. This is exactly what Bayesian Health does. Just like the best providers continually integrate new data to refine their prognostication of what’s going on with a patient, Bayesian’s technology integrates every piece of available data to equip doctors and nurses with accurate, timely and trusted clinical signals that enable them to deliver the right care, at the right time, to save lives.
Bringing a Research and Science-Led Approach
Health AI is hard — the data are messy and challenging and requires approaches that draw signals that cut through the noise. Gaining provider trust and adoption is key to improving outcomes but they’ve been burnt on low quality solutions before. And how best to deliver these new kinds of solutions so it’s effective, safe, reliable, trustworthy requires new ideas, systematic measurement, and studies demonstrating use and efficacy. Bayesian’s platform is based on nearly two dozen studies (and counting) discussing novel algorithmic advances for improving precision, measuring and tackling bias, increasing quality of evaluations, and reporting. Further, we’ve run studies involving 4,000+ providers to date studying adoption and outcomes. This is the ethos with which we’re going to grow Bayesian — innovation, transparency, scientific mindset, and results.
I’ve spent my life’s work doing research in Artificial Intelligence, Machine Learning, and most recently with a focus in Human-Machine teaming. AI/ML are going through a renaissance — self-driving cars, preventing financial fraud, speeding up search, and enabling smart automation to improve our daily lives everywhere. In healthcare, we will see the same benefits.
Realizing improved patient outcomes isn’t easy. But I’m so optimistic for what’s possible when things are done right. Over the last five years, I’ve received hand written letters from patients and their families. These letters have fueled my desire to keep pushing, harder, to bring Bayesian’s technology to the bedside, helping patients just like my nephew.
As health systems look to adopt new technology platforms that engage frontline caregivers to improve patient outcomes, understanding how a platform interacts with users and sustains engagement is critical to understanding if the technology will be successful. Even the very best solutions won’t have any meaningful impact unless they are adopted by primary users, and used frequently and consistently.
Initiating this type of behavior change isn’t an easy task as health systems, clinicians, and frontline staff are overwhelmed and overburdened; systems are working with tighter staffing ratios and reduced margins, and many solutions provide vast amounts of data in multiple systems without the tools to assist frontline staff with analyzing and prioritizing alerts and relevant information.
Bayesian Health builds its products around several principles of behavior change, targeting three of the hardest stages—the preparation, action, and maintenance stages—to encourage continued use and maximize impact.
Read our latest whitepaper on how Bayesian secures adoption and long-term engagement.
Many hospitalized COVID patients experience acute respiratory failure (ARF), but it has been difficult to predict which patients will need advanced respiratory support (e.g., mechanical ventilation) and which will not. Many predictive models have been reported for COVID patients, but most predict only the most severe outcomes (ICU admission, death, etc.) and only make a single prediction at the time a patient is admitted to the hospital. In an article published this week in Critical Care Explorations, we describe ARC (Anticipating Respiratory failure in COVID), a model that continuously monitors patients throughout a hospital encounter and identifies patients at risk of experiencing ARF. We collaborated with researchers at the University of Washington to validate the model on data collected from eight hospitals across two geographically distinct regions. At 75% specificity, the model achieved 77% sensitivity and predicted ARF at a median time of 32 hours prior to onset.
In addition to ARC, Bayesian has also trained models that predict ARF in non-COVID patients. Early identification of these patients would facilitate proactive care, leading to better patient outcomes and lower costs for health systems. Even prior to COVID, respiratory failure accounted for ~30% of unplanned transfers to intensive care units.
Read the full paper here.
Predicting events using clinical data is challenging because individual data points are often missing and/or noisy. For a given patient, input data are often collected at different times and with varying frequency. For any given measurement, the frequency can also vary between patients, often based on a clinician’s suspicion of potential complications. Learning strategies that do not account for this variation will be much less accurate. This study published in IEEE Transactions on Pattern Analysis and Machine Intelligence describes two solutions to this issue. The first is applying smart interpolation techniques that account for missing data based on patient context. The second is introducing a “Wait and Watch” policy that trades off delays in making predictions against the cost of false alerts. Compared to the standard approaches (e.g., logistic regression), these techniques achieve 3-4X higher sensitivity at comparable positive predictive value.
Bayesian Health applies these and many other leading-edge learning techniques to optimize performance and clinical actionability in our predictive models.
Scalable Joint Models for Reliable Uncertainty-Aware Event Prediction
Read the full paper here.
