7 min read

Dionysus Health

Dionysus Health
From the Washington Post article on Dionysus

This week my company, Dionysus, is in the spotlight with a story in the Washington Post about its transformational power to help millions of families who suffer from postpartum depression.

You want to support my work but don't need a keynote from a mad scientist? Please become a paid subscriber to this newsletter and recommend to friends!

Dionysus, a History

In the US last year, more than 500,000 new mothers suffered from postpartum depression (PPD), based on estimates from the CDC. Worse, most cases are never diagnosed, and many families lack access to the appropriate care they need, leading to medical complications, job loss, and even suicide. “A lot of women suffer from postpartum depression and we never talk about it,” CVS Health CEO Karen Lynch told Fortune, envisioning a health care system in which mothers are informed about the risk of postpartum depression.

Now my new startup, Dionysus Health, has announced the first ever biological test for postpartum depression. Combining epigenetics and artificial intelligence, our test predicts postpartum depression risk before delivery, rather than waiting for symptoms to surface. With $7 million in grants from the NIH-RADx program and the Department of Defense, my company is bringing our test to maternal healthcare clinics this year, starting with a very prestigious partner…that I’ll have to wait to name for now.

The roots of this breakthrough stretch back to a neurotech conference at MIT in 2019. Inspired by a presentation on the genetics of gender in mice, I joined a larger group of scientists to explore why the differences between men and women—genetics, hormones, behavior, epigenetics, and more—lead to different rates of Alzhiemer’s, bipolar, and depression.

Inspired by the early findings, Dr. Andrea Cubitt and I, founded Dionysus Health to turn our research into practical tools to prevent suicide. We built a platform that combined diverse biological and behavioral data with machine learning to identify personalized causal factors in chronic stress and suicidality. Exploring the nuanced role of hormones, genetics, and epigenetics in health led to an astonishing body of research by Dr. Zachary Kaminski and Dr. Jennifer Payne. They had identified a unique epigenetic profile in mothers diagnosed with PPD, the most common complication of pregnancy. While genes are the basic building blocks of biology, epigenetics is the process by which our environment and even our own behavior turns individual genes on and off within different parts of our bodies. You may have a genetic risk for depression, but only if it is epigenetically activated (e.g. via a traumatic childhood) might you actually experience depression. In the team’s research, moms diagnosed with PPD showed a specific epigenetic pattern.

Incorporating this breakthrough epigenetic research, we developed a test that can discover the subtle difference that predicts postpartum depression even before the mothers have given birth. Using AI to analyze the epigenetic profile of different mothers, Dionysus’s new blood test allows doctors to identify the most at-risk mothers and begin treatment before delivery rather than waiting for symptoms to harm the family.

Over the last year, we have been awarded a $6 million grant from the Department of Defense, the largest employer of women in the US that fully covers their healthcare. The grant will support research to further confirm that our test works equally well across all US mothers, regardless of race or socioeconomic status. We’ve also received an additional $1 million in grants from RADx, an NIH initiative developed out of the need for rapid diagnostic testing in the wake of COVID. This piece of the funding is supporting our first trial in an underserved maternal healthcare desert.

Based on our growing body of research, I believe our technology can be expanded to predict perimenopausal depression—a causal factor in the higher rates of Alzheimers in women—and to predict which medications are likely to be effective. This would eliminate wasted time searching through various SSRI and other medications.

Perhaps the most revolutionary aspect of Dionysus goes beyond AI or epigenetics. I have founded an independent, non-profit data trust, The Human Trust that will work to end maternal healthcare deserts in the United States. More importantly, The Trust will also hold all of the private data generated by Dionysus, never to be sold or used for any purpose other than the health of those moms. Women shouldn’t have to choose between a life-saving test or the privacy of their body.

Stage & Screen

If your company, university, or conference just happen to be in one of the above locations and want the "best keynote I've ever heard" (shockingly spoken by multiple audiences last year)?

<<Please support my work: book me for a keynote or briefing!>>

SciFri: The Ferryman and Children of Time

Perhaps inevitably, the ending of The Ferryman let me down. I still recommend it as a fine read. I’d give you some comparison books but that would give away the mystery since the book chooses to break no new ground. I can live with retreading the same water, but one challenge I’ve had since becoming a neuroscientist (25 years ago) is that brain-related scifi often runs hard against my unsuspended disbelief. Hollywood, give me a call—lazily bad science is unacceptable.

So, along with my weak tepid recommendation of The Ferryman, I’ll plug Adrian Tchaikovsky’s Children of Time and its sequels. It is a wonderful exploration of different forms of intelligence embedded in a couple of excellent stories. (Another from decades ago is The Golden Age.) I’m looking forward to his next book, Alien Clay.

It's Not Just in Your Head

In 2017, postpartum depression eroded $14 billion from the US economy (adjusted for inflation, closer to $18 billion today). If we break it down case by case, the estimated societal cost is $32,000 per affected pregnancy. To combat this problem, I co-founded a biotech start-up, Dionysus, featured in the Washington Post this week, that combines machine learning with epigenetics to predictively test for postpartum depression. Our research shows that when PPD is diagnosed early and women have access to treatment, the potential for societal cost savings is massive.

To demonstrate the potential savings, I built a financial model to estimate the cost savings when women are diagnosed early and receive appropriate treatment in comparison with the current costs of untreated PPD. The core financial assumptions of the model come from a report on the current societal costs of untreated PPD. It breaks down the costs into different categories: direct treatment ($2.8 billion), preterm complications ($3.3 billion), productivity loss ($4.7 billion) and child impact costs ($3.4 billion). My model uses empirically derived estimates of false positives, false negatives and successful detection taken from our published research.

The graph below represents an idealized world in which we can test all expectant mothers in the US. We ran three different models: NLP (natural language processing), Epi (epigenetics), Casc (AI + epigenetic). In each case, we estimated the total societal savings (blue bars) as well as the immediate savings on direct treatment costs (orange bars).

When the epigenetics and AI are combined the system saves nearly $4 billion in direct treatment costs, dramatically changing the economics of PPD. And because our test is predictive, run early in the 3rd trimester, this change strongly incentivizes insurers to treat those at higher risk before they experience symptoms. With predictive diagnosis and access to care, Dionysus is on the verge of transforming the lives of millions of families affected by PPD and dramatically reducing the costs this burden has on society.

Vivienne L'Ecuyer Ming

Follow more of my work at
Socos Labs The Human Trust
Dionysus Health Optoceutics
RFK Human Rights GenderCool
Crisis Venture Studios Inclusion Impact Index
Neurotech Collider Hub at UC Berkeley