AI, Medical Research And Advancing Women’s Health

AI, Medical Research And Advancing Women’s Health

Prof. Dr. Anabel von Preußen–Ternès von Hattburg | VYBE Future | Pres. Club of Budapest DE | FoSB.

getty

When a friend called and shared that his wife was in the hospital, an image immediately sprang to mind: I remembered her at a gathering in our garden and her slender frame in a summer dress. By my estimate, she couldn’t have weighed more than 100 pounds. In contrast, her husband probably weighed closer to 250 pounds. What he said next stayed with me: “I found it odd. I get the same dose she does.” He paused. “That can’t be right, can it?”

His comment touched on a point that I’ve now devoted myself to researching: How well do we really understand the human body when, for so long, we have treated differences between individuals as mere confounding factors? After all, women are not simply smaller men.

A Lack Of Scientific Precision​

Hormones, metabolism, immune responses, body composition and the way the body processes medication can all differ. The U.S. Food & Drug Administration points out that biological sex can influence the safety and efficacy of drugs. But, for me, it’s not a question of gender. It is a question of scientific precision.

We often ask: Are women adequately represented in studies? The more important question is: Are the differences between women and men actually analyzed and do they subsequently alter diagnosis, treatment and prevention?​ Simply including a woman in a study is not enough. If her data is subsequently pooled with men, statistical “equal treatment” can render biological differences invisible.​

Between 2013 and 2023, a mere 8.8% of National Institutes of Health research funding in the United States went toward women’s health research. Meanwhile, women are living with conditions that have been under-researched for decades, particularly when it comes to women’s health: endometriosis, PCOS, menopause, autoimmune diseases, cardiovascular diseases and Alzheimer’s. Far too often, symptoms have been normalized rather than treated as subjects for research.​ This is the true “female health gap,” not merely a lack of research, but a lack of precise knowledge.

Enter artificial intelligence (AI). Many have pinned their hopes on AI to close precisely these gaps. It’s true this tech can analyze billions of data points, recognize patterns, help identify risks earlier and make medicine more personalized. But only if we feed it the right questions and data.

The Potential Role Of AI​

AI is not automatically objective.​ It learns from the past. And if the past involved poorer diagnosis for women, different assessments of their symptoms, less frequent data collection or research results that failed to differentiate by sex, then AI cannot simply make these biases vanish. Indeed, bias can widen the health gap even further. The crucial question regarding medical AI is therefore not just: How intelligent is our model? But rather: Whose bodies did we train it on? ​

That is why I would make five questions mandatory for every company, every health-tech startup and every research project:

1. Are women visible not only in the sample but also in the analysis?

2. Is data disaggregated by sex, age and relevant biological factors as a standard practice?

3. Is an AI model tested on diverse patient groups—or merely against an average value?

4. Does the AI ​​reveal its uncertainty, alternatives and limitations?

5. Where does the human fit into the decision-making process?

For me, the future of medicine is not AI versus the doctor. It is AI plus human judgment. The machine can recognize patterns. The doctor can understand the person behind them. The machine can calculate probabilities. The human can ask: What does this decision mean for the patient’s life?

In closing, women’s health should not be a niche specialty on the fringes of medicine. It should become a benchmark for the quality of modern medicine. After all, if our medicine fails to understand biological differences, it is not yet truly personalized. And if we feed old data into new AI systems without scrutiny, we haven’t built the medicine of tomorrow. We have simply automated the medicine of yesterday.​

Potential progress begins with a very simple sentence: “That can’t be right.” We need to finally start listening more closely, especially when that sentence comes from a female patient. The future of medicine will be determined by whether we finally ask the right questions in response.​


Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?


Read More

Zaļā Josta - Reklāma