Using AI to Accelerate the Discovery of Insights and Evidence from Clinical Text

Share article:

Why did you decide to start a healthcare AI company?

Pentavere was born as a result of a tragic medical error that occurs when healthcare providers do not have the lifesaving information they need when it is needed most.

The mother of Pentavere’s co-founder and Chief Technology Officer, Steven Aviv, fractured her ankle and went to the hospital for what should have been a routine surgical procedure. The fact that she required a certain type of medication was missed by her medical team because it was buried in a clinical note. Due to this tragic oversight, she never left the hospital.

Steven is a true technologist who spent over 20 years designing and building systems for the world’s largest financial service companies so that they could make huge amounts of money in real time. After the tragic experience with Steven’s mother, Pentavere was founded in response to a simple question:

“Why is it that financial service or social media companies are using AI to make billions of dollars, but in healthcare, we still cannot use AI technology to efficiently and cost effectively leverage clinical data and more fully access, discover, and analyze clinical data in order to save lives?”

What has Pentavere’s journey looked like since founding the company?

We have stayed true to our mission, working hard every day solving the problem that we were founded to solve. We embedded ourselves within one of Canada’s leading hospitals, at our own expense, in order to fully understand how healthcare systems work. Through that experience we learned how to develop and deploy AI technology in a way that overcomes siloed and fragmented healthcare systems to unlock lifesaving information buried in multiple complex sources of clinical data. We forged relationships with world-renowned researchers, key opinion leaders, and global pharmaceutical companies interested in the problem we were solving. Through collaborations with these partners, we received over $7 million dollars of funding support to develop and validate our breakthrough and proprietary DARWEN™ AI system.

Operationalizing, deploying, and scaling AI is a universal challenge in healthcare. How does Pentavere approach this?

We’ve approached operationalizing AI by confronting the challenging realities of healthcare head on: healthcare data is messy, it is fragmented, it is siloed, data quality is inconsistent, and there is a lack of integration between systems. We’ve successfully tackled all of these challenges to operationalize, deploy and scale our AI in real-world settings. We’ve done that while always respecting patient privacy, because privacy is a non-negotiable foundation for us at Pentavere. We intentionally designed Pentavere’s architecture and deployment approach to incorporate all privacy considerations, and to pragmatically deal with messy systems and the breadth and depth of complex clinical documentation. As a result, large well respected healthcare providers and pharmaceutical companies have shared data with us so that we can discover insights that change the way patients are cared for and treated. Our results have been published in high-impact, prestigious, peer-reviewed medical journals validating DARWEN™ and DARWEN™’s ability to deliver best-in-class results in real-world healthcare settings.

Pentavere’s focus is on applying AI to accelerate discovery within clinical and medical information, and there are a number of applications. Where are you seeing the biggest impact?

What we are seeing is that there is tremendous need and demand for technology that can accelerate discovery of insights from clinical data sources.

The clinical text that DARWEN™ AI can rapidly and inexpensively discover, catalog, organize and analyze includes electronic health records, physician notes, pathology reports, physician transcriptions and many other sources. Through this, DARWEN™ AI empowers researchers, clinicians, healthcare providers, and pharmaceutical companies to apply data to develop breakthrough health solutions, including the ability to learn how existing treatments and medications can be better utilized by current patient populations; apply existing treatments and medications to new patient populations; uncover unmet medical needs; and accelerate new drug discovery and reduce drug costs in clinical trials by accelerating and automating data management and medical review.

We constantly hear from the many healthcare leaders we are privileged to work with that the impact of discoveries from clinical text can be even greater than the discovery of a new molecule. One of our customers said it best:

“The ability to use the whole of the patient experience and clinical observations to inform diagnosis and best management is a game changer that could prove as impactful as decoding the human genome.”

Everyone is talking about personalized healthcare. How do you see Pentavere’s AI accelerating this?

Healthcare is complex because every patient is unique and every patient is important. Whether these are patients with a rare disease, patients with heart failure, or patients who are a target for a particular oncology immunotherapy – every individual has different histories, socio economics, genetics, comorbidities, and lifestyle decisions. All these factors impact how treatment decisions are made, and patient outcomes as a consequence of those decisions.

This lifesaving information is captured electronically in clinical text, but the vast majority is not analyzed.  To deliver personalized care, we need to be able to identify, analyze, understand and treat the complexity of individual patients. In order to do this, we need to understand the totality of the patient journey and how different treatments and combinations of treatments work in different patients. DARWEN™ AI’s ability to analyze and understand all the unique combinations of diagnoses, patient comorbidities, past treatments, and outcomes is the missing piece to a more cost effective and personalized healthcare system.

What role do you see AI playing when it comes to promoting health equity?

The majority of health data that is used to treat patients, and which forms the bulk of literature and is considered the gold standard for clinical practice, is data that is collected in clinical trials. However, there is a growing body of literature that describes the lack of diversity of patients in clinical trials. Outside of clinical trials, it is common practice to process, curate, and analyze real-world patient data through manual abstraction, which is extremely slow, time-consuming, error-prone, expensive, and biased. Consequently, healthcare decisions are made on limited data representing extremely small portions of the population, with even more severe gaps in representation for ethnic, diverse, and marginalized populations.

With DARWEN™, we unlock otherwise inaccessible value and insights from millions of clinical data points buried in huge amounts of clinical text that reflect the experiences of large, diverse, and real-world patient populations. We identify different representative population cohorts, analyze diversity, equity, and inclusion of clinical trial cohorts, assess inequity in treatment and access to care, and understand social determinants of health, all of which are ways Pentavere is working towards the goal of promoting health equity.

What do you see as the greatest opportunities for AI in healthcare?

There is growing awareness in the field of medicine that the use of AI has potential to significantly transform the healthcare industry and accelerate the development of lifesaving health solutions. Just like the printing press transformed how society exchanged information, AI can transform the practice of medicine.

We are in the early days of AI and even earlier days of AI in healthcare. What we have seen is that 80% of the information that we produced in healthcare is not being optimally used to improve health outcomes. Unlocking the incredible value, insights and evidence that is contained in this clinical information and using AI to help us understand and improve health delivery and medicine will be one of the most significant and revolutionary changes of our generation.  At Pentavere, we are excited to be leading and participating in this revolutionary journey with other incredible partners and organizations.

Aaron Leibtag is the CEO & Co-founder of Pentavere, a digital health company that has developed a breakthrough proprietary AI engine, DARWEN™, that dramatically accelerates the discovery of insights buried in clinical text data. Aaron has fostered successful partnerships with, and secured funding from, many of the largest global pharmaceutical companies and leading hospitals to develop DARWEN™ AI, which has achieved industry leading accuracy published in prestigious journals and presented at influential global conferences. 

 Aaron has a proven leadership track record leading organizational transformation resulting in liquidity transactions for multiple private equity backed consumer enterprises. Aaron is the former vice chair of the Sinai Health System Volunteer Advisory Committee, and board member at the Museum of Modern Contemporary Art.