6 days ago
AI Can Revolutionize Women's Healthcare – Here's How
People who look closely at clinical data and medical research are recognizing that in the past, women have been left out.
One aspect of this is a narrow view of what women's health encompasses – and a historic lack of paying attention to the unique ways that women's bodies differ from men's.
A New York Academy of Sciences article expresses some of these disparities – that women were not generally included in medical research until the mid-1990s, that a woman's diagnosis tends to happen later than a man's in many cases, and that some conditions more common to women take a long time to be diagnosed, or receive less research.
Brooke Grindlinger writes about a panel at SXSW in 2024 that talked about how much of this clinical research on women has been relegated to specific and narrow applications to reproductive health.
'There's so much more to women's health than that,' said Christina Jenkins, MD, a panelist and a general partner at Convergent Ventures. 'We consider 'women's health' as a specific practice… focused on things that are unique to women, which are those reproductive organs and [associated conditions]
Of course, AI has also brought better diagnosis and new approaches to mammography: Maureen Salamon writes at Harvard Health Publishing about applications to breast cancer, while acknowledging that this is far from the only aspect of women's health that's being explored with the new technologies.
There's also a focus on the vast trove of data that come from wearables, and thoughts on how they can contribute.
'In coming years, A.I. may be able to produce personalized breast cancer risk assessment scores, which offer a more detailed picture of an individual woman's risk for the disease,' Salamon writes. 'Currently, women's breast cancer risks are calculated using questionnaires that ask about factors such age, ethnicity, family history of breast cancer, breast biopsy history, breast density, when they first had a menstrual period, and — for those who have children — how old they were when they first gave birth. All of these issues can influence breast cancer risk.'
In a recent TED talk, Lily Janjigian went over some of the ways that we are addressing women's health right here at MIT.
We've heard a lot about this in recent conferences and events where people are excited about clinical improvements with AI. But Janjigian's story is unique – as she points out in her TED talk, she was on a sports team at MIT, and ended up developing stress fractures. Then she looked around and asked why these rates of injury were so much higher among female athletes.
(getty image: a female athlete)
The result was a focus on whether women's health is getting the billing that it deserves.
'We don't know a lot about women's health,' she said, referencing a statistic from McKinsey, that less than one percent of medical research looks into women's health, beyond work on female cancers. 'This ends up (leading to) really unfair outcomes for women.'
Then she found MIT's Female Medicine through Machine Learning office (check it out here) where the focus is on exploring women's health with the new tech, and, as Janjigian pointed out, looking at patterns in large data sets for things like endometriosis.
She showed how scientists can group three elements of data: genetic data, biological data, and symptoms – and bring those together for diagnosis and patient care.
'AI can finally let us questions about women's about women's health that we haven't been able to answer in the past. So why are we not going and deploying it everywhere we can?'
The answer, Janjigian suggested, is that in some ways, AI is not a perfect solution.
It can extend bias, she noted, with the wrong approach.
'AI systems are a reflection of what and who we choose to value, so let's make sure that we're all part of that from the start,' she said.
More on AI's Power
Good research calls for a deliberative approach, but there's another reason I think that AI will be helpful here
It has to do with the attention mechanism, and how traditional research has worked. Stephen Wolfram, for one, is fond of talking about how AI's attention differs from that of humans, and what that means for our use of AI tools.
In light of that, the above experts talk about the disparities – how human research has focused on men's issues and men's health.
Well, when you're using AI data, you're working on the data sets that come in. So there's an opportunity to reduce the bias, and bring a broader lens to healthcare in general, and to women's health in particular.
In other words, the same capabilities that let AI do great work in radiology diagnosis could help to focus in on how certain conditions affect women, with really great surveys of female clinical trials and resulting solutions.
This is something a lot of people are excited about, and we should keep watching as we continue to integrate these solutions into our lives.
