Perceiv AI Leverages Big Data and Artificial Intelligence to Predict Disease Progression and Make…

Perceiv AI Leverages Big Data and Artificial Intelligence to Predict Disease Progression and Make Clinical Trials More Effective

Co-founders Christian Dansereau, PhD, and César Laurent, PhD, have put together a world-class team of experts in AI and deep learning in order to bring precision medicine to the masses and streamline clinical trials.

Challenge

Imagine you are packing your bag for a trip without knowing your destination. Do you bring a winter coat and a bathing suit, just to cover your bases? If you don’t know where you’re going, it’s impossible to know what to bring on your journey. You are packing blind.

We’ve grown to expect that we can know the forecast before we pack for a trip, yet on the long road towards developing effective new drugs and treatments, researchers and physicians aren’t as fortunate. Disease progression varies wildly from patient to patient, making it incredibly difficult to forecast which patients might benefit from a treatment or clinical trial in the future.

Researchers end up doing the equivalent of packing a little bit of everything — a one-size-fits-most approach to medicine that fails to address the heterogeneity of the patient population and disease progression. For example, a clinical trial for a new drug to prevent a second heart attack will go to enormous expense to recruit a large population size of people to test its efficacy. But only 10% of the recruited population will ultimately develop a second cardiovascular event. Not only were hundreds of millions of dollars spent on a trial that is irrelevant to 90% of its participants, but it also exposed patients who didn’t need the drug to unnecessary risks.

Or take a degenerative disease like Alzheimer’s. It’s incredibly hard for primary care physicians to predict which patients with mild cognitive symptoms are more likely to progress or remain stable. Given the uncertainty, doctors postpone their diagnosis and care until the condition clearly deteriorates to a point where diagnosis is obvious but unfortunately miss the optimal window for early treatment. Without an individualized forecast for a patient, there’s a lost opportunity for early intervention when patients are still functional.

When you calculate the amount of money these delayed diagnoses cost the healthcare system, the total is staggering. For Alzheimer’s disease alone, the lifetime cost of delayed diagnosis and interventions is estimated to be $7 trillion. And it cost over $5.7B to bring a new therapy to market, part of this cost comes from the high failure rate of clinical trials for Alzheimer’s disease. There’s simply too much variability in the disease progression to create a targeted intervention.

Origin Story

In 2017, Christian Dansereau was finishing his PhD in Computer Science from the University of Montreal. His research focused on how to apply machine learning and AI in the field of neuroscience.

“I was working with a pharmaceutical company assisting on late-phase clinical trials for Alzheimer’s disease and I saw their challenges in recruiting individuals whose progression was unknown. The trial had a huge risk of failure and ineffectiveness because of this uncertainty.”

What if, Dansereau wondered, you could better predict which patients would benefit from a particular intervention? What if you could take the wealth of patient data that can be gathered and leverage it to optimize diagnosis and prognosis? What if clinical trials didn’t need to be such a shot in the dark?

These were just the sort of research questions an academic like Dansereau could spend years turning over, but he wanted to do more than contemplate the problem. Over the course of his studies, he saw firsthand the stories and the lives behind the technology, the impact on patients and their families.

“I knew if you don’t take matters into your hands, it won’t happen. I needed to push to turn something from a research subject into a product that really impacted people’s health.”

In 2018, Dansereau launched Perceiv AI with his co-founder César Laurent, PhD. Laurent, a deep learning researcher from the prestigious Mila Institute, shared Dansereau’s passion for applying machine learning and AI to the healthcare field. Together they developed the initial software for the company, focusing on Alzheimer’s clinical trials as their first vertical, building off of Dansereau’s doctoral research.

A big turning point for the company came at the end of 2018 when they presented on the main stage at the Clinical Trials on Alzheimer’s Disease (CTAD) annual conference. After they walked off the stage, multiple pharmaceutical companies approached them to discuss the application of their AI for trial recruitment. These conversations lead to the company’s first contract in 2019.

Close to half a million in grant funding allowed Perceiv AI to build out its team in 2019 and 2020, bringing on Angela Tam, PhD, as Senior Scientist and Adrián Noriega de la Colina, MD, PhD, as the Clinical and Regulatory Lead. The team launched their first product offering, Foresight AD™, at the CTAD conference in 2021, a full-circle moment for Dansereau and Laurent. In three years, they took their idea from theory to a real-world tool.

Under the Hood

Perceiv AI predicts disease progression using a platform powered by machine learning and AI. It’s a forecasting tool that allows for precision medicine: treatments and interventions that take into account individual clinical risk factors, cognition, genetic biomarkers, environment, imaging, and other information across different modalities to more accurately address a condition. Instead of a one-size-fits-most approach to healthcare, Perceiv AI’s platform empowers targeted research and care to find out what specific strategies work best based on the many factors in play.

Their first product, Foresight AD™, specifically forecasts the progression of cognitive decline in Alzheimer’s disease. They spent several years building a comprehensive database of patient information using imaging, genetic, phenotypic, molecular, and clinical variables to train their algorithms to make its predictions. It took deep expertise to integrate multiple data sets across multiple modalities together, but the proof has certainly been in the pudding. While other strategies can reduce the sample size of a trial by 15%, Foresight AD™ proved able to reduce a sample size by 51%. An optimally enrolled clinical trial significantly reduces costs while increasing the power of a study to prove efficacy of a treatment.

Because of their success with Foresight, Perceiv AI is also looking into applications in the clinic to assist physicians in a timelier diagnosis so that interventions start sooner; this is critical with the new disease modifying drugs that will become available in a year or so.

Perceiv AI plans to expand from clinical trials to the exam rooms of primary care physicians as a clinical decision support tool. They have already applied for breakthrough designation from the FDA to start their path towards bringing that product to market. Their vision is to integrate directly with the physician’s interface and EMR platform, so that as the physician gathers information from a patient through different modalities, the Perceiv AI algorithm can predict likely progression and recommend followup tests and treatments.

In the long run they are also looking into expanding to other neurological disease areas, like MS and Parkinson’s, where they can leverage the same learnings and wealth of information they have aggregated for Alzheimer’s.

“There is a lot of space in the market,” explains Dansereau. “It’s very nascent. People are just starting to understand the use of AI in these studies and in the clinic.”

“We are in dire need of precision medicine and a more thoughtful approach to treatment. You see it happening in the oncology space currently and it’s about time to bring that thinking and success into other areas as well. We can reduce cost and show more value for patients.”

Why We’re Proud to Invest

StartUp Health is backing Perceiv AI because of the way they combine visionary thinking and processes with the academic know-how. Their team brings together highly skilled people from the worlds of neuroscience and AI to offer a product that increases the efficacy of trials and treatments for patients. With the prognostic tools they produce, researchers and physicians no longer have to “pack blind.”

Thanks to their connections at various academic institutions, their advisory team includes such luminaries as Yoshua Bengio, PhD, known for his pioneering work in deep learning, Serge Gauthier, MD, FRCPC, the Director of the Alzheimer’s Disease Research Unit at McGill University, and Betsabeh Madani Hermann, MEng, MBA, an advocate for translational research and thought-tech innovations, among other respected innovators and thought leaders. These advisors represent the wealth of insight and knowledge backing Perceiv AI’s work and keep the company striving for novel applications and new verticals to tackle.

Finally, we’re proud to back Perceiv AI because of the way they ensure that their insights are actionable. It would be easy to get caught in the ether of the vast amounts of data and information they process and its implications. But Perceiv AI has a clear North Star: the patients. People drive their processes. Their digital biomarkers are highly specific and thoroughly validated so they are actually useful and usable in clinical settings.

Join us in welcoming Perceiv AI to the StartUp Health family!

Connect with the Perceiv AI team via email.

Passionate about ending degenerative brain conditions? If you’re an entrepreneur or investor, contact us to learn how you can join our Alzheimer’s Moonshot.

Investors: Contact us to learn how you can invest in Health Transformers.

Digital health entrepreneur? Don’t make the journey alone. Learn more about the StartUp Health Community and how StartUp Health invests.

Sign up for StartUp Health Insider™ to get funding insights, news, and special updates delivered to your inbox.

Follow us on social media for daily updates on Health Transformers: Twitter, LinkedIn, Facebook, and Instagram.

Perceiv AI Leverages Big Data and Artificial Intelligence to Predict Disease Progression and Make… was originally published in StartUp Health on Medium, where people are continuing the conversation by highlighting and responding to this story.

biomarkers
healthcare
digital health
artificial intelligence
ai
big data
machine learning
software
medicine
pharmaceutical
application
therapy
fda
research
clinical trials