About 12% of women in the US will develop invasive breast cancer over the course of their lifetime, and there are expected to be close to 250,000 new cases of breast cancer in 2016 alone.
With a breast cancer diagnosis comes fear, anxiety, stress, added expenses and–sometimes–a missed diagnosis. Approximately 5% to 17% of breast cancers are missed by a radiologist. Mammograms can also result in a false positive or an overdiagnosis, which leads to additional and unnecessary cancer treatment for the patient. According to a study by Health Affairs, between 22% and 31% of all diagnosed breast cancers are overdiagnosed, resulting in an additional $4 billion in health-care spending annually.
To help take human error out of diagnosing breast cancer, a Chicago high school student wants to use artificial intelligence software to help doctors make more informed treatment decisions and give patients a more accurate diagnosis.
Lane Tech rising senior Abu Qader and his co-founder Vedad Mešanovi? have launched GliaLab, an artificial intelligence MedTech startup that works with existing medical imaging devices to give smarter breast cancer diagnoses. The startup uses machine learning and big data to create a faster and more accurate ways of diagnosing cancerous tumors, says Qader, with the goal of saving lives and preventing unnecessary spending on treatment.
“Thousands of women are misdiagnosed with breast cancer every year, and it usually leads to death or enormous treatment costs,” Qader said. “Where we come in is right when the radiologist is making (the diagnosis) … We have a lot faster, a lot more accurate, and a lot cheaper method of telling you what type of cancer tumor you have.”
Qader says GliaLab’s software is between 93% and 99% accurate, and gives results in real-time. And Qader added that while some companies are building entirely new machines to provide this same type of artificial intelligence, GliaLab works with the devices doctors already use, which helps keep the price down for patients.
Qader first conceived the idea for GliaLab during his sophomore year at Lane Tech when he built an early version of the AI software for a class project. Qader, who calls himself a self-taught computer scientist, continued working on the project after the class finished, brought on a co-founder to handle the business side of things, and launched GliaLab last December.
The startup has partnerships with hospitals that are willing to beta test its program, Qader said, and the company is talking with angel investors in hopes of raising around $400,000. In the future, the startup plans to partner with companies that make imaging machines so GliaLab can embed its software in their devices, and it plans to expand its AI beyond just breast cancer diagnoses.
“We really think that with the technology we have and what we’re creating, we can actually affect lives,” he said.
Images via GliaLab