Our healthcare system is a world of contradictions. It performs miracles on a daily basis, like healing cancer patients, transplanting organs, and enabling trauma victims to recover against all odds. At the same time, it fails us continuously, in the form of preventable deaths, missed diagnoses, and inequitable care. It also increasingly hampers clinicians with the administrative burden of EHRs, which has led to crisis-level burnout rates that were only exacerbated by the pandemic. If we continue to operate as we have, our country faces a massive health quality problem and capacity shortage challenge that could cripple our ability to deliver cost-effective care to all who need it.
One of the many ways by which innovators have responded to this dire need for improvements in efficiency and yield is through the creation of new infrastructure platforms that industrialize things like healthcare operations (e.g., on-demand home-based lab draws), financial services (e.g., tools for digital payments that don’t rely on snail mail or point-of-service card swipes), and data interoperability (e.g., patient record retrieval engines).
However, we have yet to see the advent of robust infrastructure for industrializing clinical intelligence, despite it being perhaps the most crucial component of the healthcare value chain — largely because it’s been hard to reliably scale the clinical judgement of doctors. Cracking this nut requires invention of techniques that work specifically with complex health data, and presentation of actionable insights in a manner that clinical users can trust, since the cost of getting it wrong is so high. Many attempts to transfer AI developed for other use cases, like advertising or manufacturing, into care delivery have failed both technically and commercially for these reasons.
Fortunately, we’re now at a tipping point where software and artificial intelligence can realistically supercharge clinician decision making. We’ve long awaited clinical AI solutions that can be integrated into live point-of-care workflows, demonstrate clear clinical utility, exhibit transparent precision, and be readily adopted by physicians. While there have been numerous academic demonstrations of the potential of AI in clinical decision making, the recent proliferation of digitized medical data, APIs for integration into EHR workflows, and value-based incentives to invest in preventative care have created the necessary conditions for viable clinical AI applications in the real world.
Bayesian Health has brought this new capability to the front lines of healthcare, in the form of a clinical AI platform that delivers real-time patient risk stratification and triage to front-line clinicians, in-workflow. Clinicians use the Bayesian suite of workflow products to rapidly make data-informed decisions about which patients would benefit from proactive intervention and what the best course of action might be, while tapping into a trove of intelligence about how other patients with similar characteristics and clinical status have progressed on that basis. Bayesian’s published outcomes study shows how the platform’s clinical signals can help physicians diagnose and treat sepsis 1.85 hours faster, with an impressively high 89% clinician adoption rate.
Founder and CEO Dr. Suchi Saria, a renowned health AI expert and director of the Machine Learning and Healthcare Lab at Johns Hopkins University, is setting a new standard for the level of clinical rigor and peer-reviewed validation that a digital health product exhibits to pass muster in high-stakes, patient-facing clinical settings. Health systems are increasingly seeking to partner with experts like her to implement clinical AI strategies that deliver utility and ROI, responsibly. Bayesian also represents a new “clinical intelligence infrastructure” that can insert informed AI into unbundled care models to increase leverage across our currently constrained physician supply.
We couldn’t be more excited to be leading an investment into Bayesian Health and for Julie to join its board, as Suchi and her team fuse together clinically validated technology with deep knowledge about complex care delivery environments to help our healthcare system heal itself.
* * *