While a growing number of women are empowered by learning about their health using wearables, the way Artificial Intelligence (AI) and femtech interact needs attention.
AI algorithms can process vast amounts of medical information and identify patterns that may not be immediately apparent to human healthcare providers.
Femtech uses AI in many ways, from fertility tracking to menstrual cycle management, pregnancy monitoring, and even menopause support.
AI in femtech helps women
Artificial intelligence is increasingly being incorporated into femtech and healthcare products.From helping doctors to diagnose to assisting people to track and understand their bodies, AI does much of the groundwork in processing data and contributing to decision-making.
Here are four examples of areas where AI is being used in women’s health:
- Breast cancer
Screening for breast cancer requires analysis of ultrasound imaging. This analysis can be done by an AI, which has been taught using millions of reference images of what a normal mammogram looks like versus a mammogram with cancer.
This way, the AI can reference this ‘learning’ when reviewing a mammogram and detect abnormalities quickly and precisely.
A radiologist and an oncologist review these findings , thus helping them do their jobs. In a recent Swedish study, the the screenings which received help from the AI detected 20% more cases of breast cancer than radiologists alone.
AI is also used for other types of cancer screenings, such as cervical cancer, using a similar principle of analysis and detection of abnormalities. An example of this is the The EVAPro digital colposcope.
Increasingly, AI is being used to track and diagnose menopause symptoms, such as the app Midday, which we have written an article about.
Midday uses a sensor for hot flash detection built into your smartwatch and AI to help predict when hot flashes might happen.
Thermaband is another product which uses AI to track hot flashes, heart rate, and temperature thus helping women to understand their bodies, symptoms, and to cool them down before a hot flash strikes.
Ava is a bracelet which is worn while sleeping and which uses machine learning algorithms to track five physiological signals: pulse rate, breathing rate, sleep, heart rate variability, and temperature. It uses these to predict a woman’s next fertile window and ovulation.
Ava has been shown to detect an average of 5.3 fertile days per cycle at 89 percent accuracy and is the first FDA-approved fertility tracking wearable.
Bloomlife is a patch worn by pregnant people which measures fetal movement, fetal heart rate, contractions, and maternal health in a non-invasive way.
The device sends data to an app on the wearer’s smartphone and helps to predict and manage pregnancy complications. Health care providers can remotely review this data to ensure that mom and baby are doing well.
Natural Cycles is an app that works in combination with a built-in thermometer in a smartwatch or the Oura ring to track your body temperature and predict when you’re fertile. It is a non-hormonal, zero-side effect, FDA-cleared way of tracking your cycle and is 93% effective with typical use and 98% effective with perfect use.
You can read our review of it on TechTruster here. Body temperatures change over the course of the menstrual cycle, with a low but significant rise around ovulation day. Natural Cycles uses AI (a smart algorithm, as they call it) to predict one’s fertile window.
How we can benefit
Even though women make up about half of the world’s population, there is significantly less research on women’s health than men’s.
That’s why it’s important to learn about how AI can help women’s health by doing more research about topics such as pregnancy complications, endometriosis, maternal mortality, breast and cervical cancers, and other women’s health concerns.
AI-powered devices and applications are being developed to address specific topics such as fertility tracking, menstrual cycle management (including non-hormonal treatment), identifying cancer in early stages, pregnancy monitoring, and menopause support.
AI algorithms can analyze medical imaging data (such as mammograms and pap-smears) with high precision, leading to earlier and more reliable diagnoses. These technologies offer convenience and empowerment by putting more control over their health in the hands of women.
Considerations about data, privacy, and ethics
While the benefits of using AI and femtech to improve women’s health are numerous, there are potential downsides, too, and they need to be addressed.
Using AI in women’s health brings up important ethical considerations and we have to ask questions such as: what is happening to this massive amount of data being collected?
Collecting and analyzing sensitive personal data raises concerns about privacy and security. Women must be assured that their personal information will be protected when using AI-powered healthcare solutions.
Furthermore, there is a need for transparency in how AI algorithms are developed and validated. Bias within these algorithms could lead to inaccurate diagnoses or unequal access to healthcare resources for certain groups of women.
We must ensure that AI-driven health tech solutions are accessible to all women, regardless of their socioeconomic status, education level, or technological proficiency, and ensure that the benefits of AI are not limited to certain demographics.
Algorithm vigilance against bias
While AI has the potential to impact women’s health positively, it is crucial to carefully consider these ethical issues when using AI to collect and analyze health data.
Striking a balance between innovation, efficiency, and accessibility on the one hand and safeguarding patient rights on the other should be at the top of the checklist in this field (read more about this controversy here).
The World Health Organization recognizes the importance of algorithm vigilance and has presented six core principles for AI in healthcare in its “Ethics and governance of artificial intelligence for health: WHO guidance”.
These principles also highlight the importance of data protection and informed consent, transparency, accountability, inclusiveness, and equity.
From underrepresentation to equity
We must also keep in mind that women have historically been excluded from clinical trials. Therefore, if AI seeks out information from past medical records, this will not be as helpful as similar information for a male body.
This underrepresentation means that much of the medical knowledge and data collected may not adequately reflect women’s unique health needs and responses. This, in turn, can lead to gender-specific health issues being overlooked or less understood.
Addressing underrepresentation in healthcare and medical research is crucial for developing AI-driven health tech that truly meets the needs of women.
This includes promoting greater inclusivity in clinical trials, advocating for more research on women’s health issues, and actively feeding the AI diverse and representative data to improve the quality and accuracy of the AI material and analysis.
By addressing these challenges, the field of AI in women’s health can strive for equity and improved healthcare outcomes for all.
Where to learn more
Menopause.tech – A site in progress, created by the author of this article.