How AI is helping to make breast cancer history – TechCrunch

Each October for the past four decades, Breast Cancer Awareness Month has helped raise the profile of the most common cancer on Earth, a cancer that claims nearly three-quarters of a million lives each year.

Despite recorded cases dating back to ancient Egypt, breast cancer has been considered an “unspeakable” disease for millennia. Women were expected to suffer in silence and in “dignity”.

This stigma fueled academic ignorance, with breast cancer languishing as a relatively little studied disease until just a few decades ago. For most of the last century, a woman suffering from breast cancer would have been offered radiation therapy and / or surgery – often radical surgery, leaving her disfigured for little benefit – while the treatment of other cancers were progressing.

Breast cancer mortality barely changed from the 1930s to the 1970s, until a concerted effort by feminist and women’s liberation groups elevated the study and treatment of breast cancer to its place. returns to heavily male-dominated hospitals and research institutes. Treatment turned into a generation.

In the 1970s, a woman diagnosed with breast cancer had about a 40% chance of surviving for the next 10 years. Today, that likelihood has nearly doubled, thanks to new drugs, advanced screening methods, and more subtle and effective surgery.

The emphasis on early diagnosis has been essential to this transformation. The earlier breast cancer is detected, the easier it is to treat. Artificial intelligence is playing an increasingly critical role in identifying breast cancer. This year, the UK’s National Health Service (NHS) announced a study how AI could screen for breast cancer. While intended to augment, not replace, human physicians, it would help alleviate the shortage of x-ray machines – 2,000 more are needed to clear the NHS backlog in analyzes caused by the pandemic.

Startups are also using AI to deal with this shortage. Brittany Kheiron Medical Technologies plans to use AI to screen half a million women for breast cancer. spain the blue box develops a device capable of detecting breast cancer from urine samples. india Niramaï is working on a low-cost tool that could help screen large numbers of women in rural and semi-urban areas.

But identifying patients at high risk for relapse is just as crucial to improving outcomes. About one in ten breast cancer patients will relapse after their initial treatment, reducing their chances of survival.

Identifying them early has always been difficult, but my team, in collaboration with Gustave Roussy, a French cancer hospital, has developed an AI tool capable of identifying 8 out of 10 patients at high risk of relapse. AI helps get patients the treatment they need sooner while sparing low-risk patients frequent and unsettling exams. Meanwhile, pharmaceutical companies are speeding up breast cancer drug trials by recruiting high-risk patients faster.

The confidentiality of patient data can be an understandable barrier to a speedy search. Hospitals are cautious about sending data offsite, and no pharmaceutical company wants to share valuable data with its competition. But AI is helping solve these problems, enabling the development of new treatments faster, safer, and cheaper.

Federated learning, a new form of AI that trains on data from multiple institutions without the data leaving hospitals, is being used across Europe to give researchers access to critical, but previously inaccessible, data.

We will also use AI to deepen our understanding of why the more aggressive forms of breast cancer are resistant to certain drugs, helping us to develop new, bespoke drugs that better distinguish healthy cells from tumor cells than cancer. chemotherapy.

As the influence of AI grows, it is equally important to improve outcomes to recognize that healthcare is a fundamentally human endeavor. No algorithm will ever be able to comfort a patient in their darkest moments, and no machine will ever be able to instill and inspire the resilience that every patient needs to overcome their disease.

I and all the other doctors know that treating illness is as much about understanding the patient as it is about understanding their affliction. Clinician empathy is related to greater patient satisfaction and less distress, motivating a patient to continue with difficult treatment. Fortunately, the AI ​​technology that is increasingly helping breast cancer treatment is designed to empower and empower physicians.

Breast cancer is no longer “unspeakable” for the millions of people who are diagnosed with it each year. The sea of ​​pink ribbons that heralds the start of October signals how far we have come in our battle against one of our oldest enemies, the one we are defeating. We may never be able to completely eradicate breast cancer. But with AI helping diagnose patients earlier and allowing for rapid development of treatments, it’s possible that in a few decades we won’t need a month of breast cancer awareness.