Neural Notes: Ovum AI wants to close the gender data gap in healthcare

Welcome back to Neural Notes: a weekly column where I look at some of the most interesting AI news of the week. In this edition: Ovum AI is an upcoming women’s healthtech app that aims to democratise healthcare for women and address the lack of diverse data sets in the space globally.

Melbourne healthtech startup Ovum AI is gearing up for its pilot phase later this month. Founded by Dr Ariella Heffernan-Marks, Ovum aims to go beyond standard health-tracking apps. Pioneering Australia’s first-ever longitudinal AI women’s data set, it aims to offer a holistic, personal health assistant designed with and for women.

Ovum’s holistic approach to health

More than 3.3 million Australian women currently use apps to manage aspects of their reproductive health and fertility, but Ovum’s approach will go beyond these functionalities.

While it’s yet to officially launch, it has been in the works since 2022, with Dr Ariella Heffernan-Marks at the helm.

Ovum has already attracted investment from Wollemi Capital and venture studio Nakatomi. It is also planning a crowdfunding round this year.

Ovum integrates comprehensive health data into a centralised AI platform, including the incorporation of data from the likes of wearables. And its planning for the long term with longitudinal data, which captures the evolution of variables over extended periods.

This is invaluable in healthcare, where it enables researchers and practitioners to track changes, assess the effectiveness of interventions and understand the progression of health conditions over many years.

For an application like Ovum AI, employing longitudinal data means the platform can systematically monitor and analyse changes in a woman’s health metrics throughout her lifetime.

“We know that AI produced 85% of erroneous outcomes for women due to gender bias and it’s a massive issue, which is what I am trying to solve,” Dr Heffernan-Marks said on a call with SmartCompany.

According to Dr Heffernan-Marks, the scenario Ovum finds itself in is somewhat reminiscent of the chicken and the egg. How can you create an AI for women when AI currently has a substantial gender bias?

“In our MVP we’ve trained in a combination with an award-winning venture studio, AI engineers, UX, professional marketing strategist and with my medical expertise. We also have only accessed specific resources that are evidence-based and representative of women,” Dr Heffernan-Marks said.

“Then through our pilot, and through training and off real women, we hope to create our own machine learning and large language model that will be completely representative of women. Our own data set.”

Dr Heffernan-Marks emphasises that Ovum does and will continue to have privacy at its core. Personal information of users won’t be shared, and they won’t even be required to add their full birth date during the sign-up stage,

This ongoing collection and analysis will facilitate the creation of highly personalised healthcare plans, adapt to the unique and changing health needs of women at different life stages, and ultimately, provide insights that could lead to groundbreaking advancements in women’s healthcare and management.

“One in two women navigate a chronic health issue in Australia and by leveraging the power of AI, our bespoke personal health assistant works to understand and empower women with resources and confidence to manage their health over their lifetime,” Dr Heffernan-Marks said.

Addressing systemic health disparities

Ovum’s launch comes at a crucial time. Global health data has long been under-representative of women, leading to significant disparities in health outcomes. This is especially pronounced in conditions like endometriosis, which can take between 7 and 12 years to diagnose.

Dr Heffernan-Marks founded Ovum to combat these challenges head-on, driven by firsthand experiences of discrimination in healthcare based on gender, age, sexuality, disability, migration status, and income.

“The healthcare system has systemically underfunded and underrepresented women’s health, and with women being under or misdiagnosed, my vision is to create an accessible resource that is designed with women, for women,” Dr Heffernan-Marks said.

With the femtech sector projected to be worth US$108 billion by 2032, Ovum is set on not just participating in the market but leading it with a focus on global and diverse data sets.

The startup is particularly keen on including a wider spectrum of women in its data collection. This includes trans and non-binary individuals, to better understand and address their unique health needs.

“That’s why the pilot is so important because we need to be trialing it before going onto the market with women from diverse backgrounds to understand what are the issues, is it producing biases and how can we minimise that before going on to the market,” Dr Heffernan-Marks said.

The commitment to creating a diverse global data set is critical, as current AI models often perpetuate existing biases. Ovum aims to correct these disparities by ensuring its AI learns from a broad and inclusive range of health experiences.

Ovum’s plan for an equitable access model

Recognising the financial barriers many face in accessing quality healthcare, Ovum is exploring innovative funding models to make the app free for all users.

The goal is to rely on industry funding to cover costs, allowing free user access to its features. However, Dr Heffernan-Marks said the company has been floating the idea of payment models.

“If we’re at the stage of needing to go to market and need some form of revenue for product development — we have spoken to women and found that they are willing to pay around $10 a month,” Dr Heffernan-Marks said.

Another option is a pay-it-forward model, where those who can afford it may opt to pay more, subsidising access for those who cannot. According to Dr Heffernan-Marks, this suggestion received highly positive feedback from the women they spoke to.

At the present time it’s unclear exactly when Ovum will launch, but there are hopes for it to be this year. As Dr Heffernan-Marks notes, the pilot will be crucial in determining the next steps.

Every piece of feedback and data collected will be crucial in shaping a health assistant that truly understands and meets the needs of women– especially when the current healthcare system can be so quick to dismiss them.