Like many diseases, cervical cancer is treatable if caught early, yet millions of women miss out on routine Pap smears for the disease each year, killing a disproportionate number of black women.
Cervical cancer screening rates have gotten worse during the pandemic, especially for low-income women and women from minority racial and ethnic groups.
A group of recent graduates of the Johns Hopkins Carey Business School thinks that artificial intelligence can solve this problem. In a class they took last fall, they developed the concept of a “smart tampon,” an at-home cervical test that they hope will make disease screening more accessible and ultimately reduce discrepancies.
They’re not the only ones who have high hopes that artificial intelligence and machine learning technologies will play a role in the future of healthcare.
A growing number of researchers in Maryland and around the country believe the technology will change the way patients are treated, allowing them to diagnose them earlier and more accurately and better spot signs that they may be at risk for chronic disease. disease or condition.
Just in the past few months, Johns Hopkins University and the University of Maryland have launched centers to further integrate artificial intelligence into medicine.
Beyond healthcare, the market for AI technologies is booming and extending into settings as diverse as courtrooms and classrooms. According to market researchers, the market is expected to be worth more than $1 trillion by 2030.
But as the technology has grown in prominence, there have also been concerns.
Just like people, AI algorithms — and the large data sets they rely on — can be biased. If used irresponsibly, the technology could reinforce the ways in which the system already discriminates against marginalized groups and could worsen them.
But in medicine, some researchers believe that if the technology advances thoughtfully — and physicians are educated about its limitations — it could make health care more affordable and accessible while reducing disparities.
“AI has the potential to be a transformative technology in the practice and delivery of medicine,” said Ritu Agarwal, co-director of the Carey School’s Center for Digital Health and Artificial Intelligence. “But there is much more to do before we can realize its full potential. .”
Like many college projects, the smart tampon idea started with a group of students sitting around a friend’s living room.
In a class they took with business school professor Tinglong Tai, the students were tasked with selecting a healthcare challenge and figuring out how to solve it using artificial intelligence techniques.
During group brainstorming, someone asked, “Is there anything that sucks in healthcare? Is everyone afraid?”
For two of the women on the panel, Hayley Hoaglund and Madeline Howard, the answer came quickly: a Pap smear, which is done by The often uncomfortable procedure of scraping cells from the cervix to diagnose cervical cancer.
Over the next few weeks, the group — which also included Charlie Acosta and Tomas Delia — developed a concept for a device that would act like a tampon inserts and scans a person’s cervix the same way an iPhone scans a user’s face.
scan will be Upload to an app that will use artificial intelligence software to compare it to other cervical scans in its database. If the software detects abnormal cells in a patient photo, it notifies the provider or the patient to seek additional testing.
The group believes the tests could be done every three years, perhaps during routine physicals or even at home. Its ease of use could help bridge the gap in disadvantaged communities with high cervical cancer mortality rates. A recent study found that black women were 41 percent more likely than white women to develop cervical cancer and 75 percent more likely to die from it.
Cervical cancer kills approximately 4,000 people each year.
Ultimately, Hoaglund said, the goal is to encourage more people who have had cervical surgery to get cervical cancer screening — especially those who would otherwise not be able to get to a doctor’s office for a Pap smear, and those who have been sexually assaulted or have violated the law. , making the current program a potential triggering program.
“Smart tampons” are far from hitting drugstore shelves. The students did not have a prototype, and although Hoaglund said they were interested in finding investors, the project has been on hold since they graduated.
Elsewhere in Baltimore, researchers have invested heavily in launching a test that uses artificial intelligence to screen patients’ blood for signs of tumors.
Delfi Diagnostics, a biotech company founded in 2018 by Dr. Victor Velculescu, a professor at the Johns Hopkins School of Medicine, raised $225 million in its latest funding round.
The company’s scientists are developing technology that uses a type of artificial intelligence called machine learning to analyze DNA fragments to sense the presence of cancer and determine where tumors are located in patients.
Delfi is currently in a large clinical trial across the country looking for a technology to detect lung cancer, Velculescu said. Next, the company’s scientists hope to put a similar technology that can detect liver cancer to use with a similar test.
Like cervical cancer, lung and liver cancers are easier to treat if they are detected early. But unlike the more invasive procedures now required to detect these types of cancer, Delfi’s test will be able to be performed on a blood sample drawn at a doctor’s office, Velculescu said.
“Our goal is to develop tests that are very cheap and easy to use so that everyone — regardless of socioeconomic level and so on — can be screened,” Velculescu said.
On a North Bethesda rooftop in early November, officials from a coalition of Maryland educational and medical institutions announced the launch of a center that will study the use of artificial intelligence in medicine and support advances in the field.
University of Maryland 3 – The Health Computing Institute will use patient data from the University of Maryland Health System that has been de-identified in a series of projects aimed at improving care across the state, said Dr. Mark Gladwin, MD Dean of the University of Maryland Medical System at the University of Maryland School of Medicine in Baltimore.
In addition to UMHS, partners include the University of Maryland at Baltimore and the University of Maryland at College Park.
Laboratory and office space for the center is under construction in North Bethesda and is expected to open in 2028, the University of Maryland, Baltimore said in a news release. But Gladwin said he expects scientists to be hard at work in rented office space six months from now.
Officials hope the institute can use artificial intelligence to track how newly-approved drugs perform in different populations and spot patterns in electronic medical records that would help doctors spot diseases early.
The researchers also hope to use the technology developed at the center to train surgeons and medical students and eventually conduct clinical trials.
Gladwin said he is well aware of how AI algorithms can leave members of marginalized communities behind or at a disadvantage. But, he said, UM Health has one of the most diverse patient populations in the world.
“We hope that by having a more inclusive, diverse patient population, we’ll help ensure that advances in artificial intelligence reach all populations,” he said. “That’s really important.”
The Johns Hopkins Center for Digital Health and Artificial Intelligence also wants to develop AI technologies that work for everyone.
It’s part of the center’s broader mission, Agarwal said: to find new ways to use artificial intelligence to improve health care outcomes, whether in terms of quality, patient safety, equity, access or cost.
Another Hopkins professor, Goudong Gao, co-directs the center with Agarwal. He described a number of ongoing projects, including one that will use a type of artificial intelligence to study doctors’ clinical records to determine whether race and socioeconomic factors influence the breast cancer treatments they recommend.
The center also hopes to study a way to remove stigmatizing language from clinical records, removing phrases more commonly used to describe members of marginalized communities — such as calling patients “difficult.”
Agarwal said it will take a lot of work—likely several years—to fully understand how to create value from AI technologies in healthcare.
“Taking these technologies and disseminating and implementing them in a system that has been fairly rigid for decades?” she said. “It’s going to be challenging.”