Meet the Bread-Scanning AI System That’s Learning to Help Doctors Diagnose Cancer

Photo Courtesy: [Ken Ishii/Getty Images]

Have you ever gone to your local bakery or grocery store and splurged on bread and produce — then waited while the cashier entered all of the price codes for every item? If so, you know how time-consuming it can be to sort through different breads, fruits and veggies. In Japan, a new technology system called BakeryScan is making this process easier — and that technology may even end up saving lives in addition to the time spent waiting in line. 

Developed by computer systems engineer Hisashi Kambe and his company, Brain Co., BakeryScan uses a camera and artificial intelligence (AI) to assess different types of baked goods so cashiers don’t have to memorize hundreds of prices. Within seconds, the point-of-sale system can identify the type of bread that’s been placed onto the checkout station below the system’s camera. The cashier doesn’t need to enter or look up any codes and can focus on completing transactions much faster.

At most bakeries, placing barcodes on fresh goods is time-consuming. Until recently, cashiers had to learn to identify baked goods by sight and either memorize the price or a code associated with each type. But BakeryScan speeds up and optimizes the checkout process. The system can differentiate more than 100 products and track their prices. And, surprisingly, its uses don’t end there.

The AI in Brain Co.’s BakeryScan might also revolutionize healthcare. As it turns out, the system’s cameras can play a role in detecting cancer cells. Versions of BakeryScan technology are now in use in medical spaces in addition to bakery registers — and it’s possible that these scanners may be joining the lineup of healthcare tech that’s now utilizing AI.

BakeryScan Undergoes an Unexpected Medical Breakthrough

Photo Courtesy: [Shapecharge/iStock Images]

Historically, some of the best medical advancements were the result of pure luck — like Alexander Fleming creating penicillin after leaving mold unattended, for example. In the case of BakeryScan, though, it took a key observation to help researchers realize the system’s true potential.

By 2017, BakeryScan had become so popular that it was a mainstay in Japanese bakeries. The robotic assistant was featured in media everywhere, from news reports and social media posts to print ads and commercials. That year, a doctor who worked for the Louis Pasteur Center for Medical Research in Kyoto, Japan, saw a commercial for BakeryScan. Had that commercial not aired at that time, BakeryScan’s medical training may not have started. 

The physician reached out to Brain Co., explaining that bread’s molecular structure looks surprisingly similar to cancer cells, and asked if the technology could be used for medical purposes. The company hadn’t created tech for use in the medical field before that, but this idea had promise.

Out of this inquiry, Cyto-AiscAN was born. Cyto-AiscAN is the name of the adapted version of BakeryScan designed for medical use. Brain Co. was new to the medical field before Cyto-AiscAN was introduced, even though the company’s technology is typically focused on digital recognition systems in some way. But instead of using cameras to look at baked items for sale, this modified version of the AI can detect cancer cells.

BakeryScan Tech Becomes Cyto-AiscAN Innovation

Photo Courtesy: [Charday Penn/Getty Images]

What separates Cyto-AiscAN from BakeryScan is the size of the item that the system reads. Bread can fit in your hand; cancer cells are microscopic. Cyto-AiscAN doesn’t scan body parts under its camera, either. At least, not yet. Instead, doctors can place a very small sample of a tumor onto the type of slide that’s typically used for microscopes. And instead of distinguishing between a croissant and a doughnut, Cyto-AiscAN’s electronic readers can accurately tell the difference between healthy and cancerous cells.

Cyto-AiscAN might completely revolutionize healthcare as we know it. When diagnosing cancer, CT and MRI machines are often used for diagnostic imaging. Even mammograms require X-ray technology. What makes these methods of imaging potentially harmful is that they all involve radiology in some way.

Anything that’s radioactive can cause cancer or make it worse. This goes beyond diagnostic imaging and into cancer treatment as well. Radiation is one of the most widely used methods of treatment. While it may not be as damaging as chemotherapy, which uses a combination of medicines to treat patients, radiation side effects can still be painful and harmful to people living with cancer.

Cyto-AiscAN has the potential to improve cancer treatment in the future because it can eliminate the need to have someone with cancer undergo radiation scans as frequently. To confirm the system’s possible uses, researchers are currently experimenting with Cyto-AiscAN in two hospitals in Japan.

Cyto-AiscAN Is Only One Piece of the AI-Healthcare Puzzle

Photo Courtesy: [Bloomberg/Getty Images]

Thanks to Hisashi Kambe and other innovators, the future of healthcare is looking bright. While Cyto-AiscAN is still in a development stage — one that does look promising — AI is specifically in use for cancer treatment in ways other than diagnosis. For example, experts are currently experimenting with nanobots, which are microscopic robots about the width of a human hair, in the treatment of tumors. With some tumors, especially in the brain, doctors can’t inject the tumor or area around it with medicine because it can get caught in the bloodstream and make the patient sick. Using an air bubble, nanobots can reach areas that are harder to work on. There probably won’t be any “ultimate” treatment for all things cancer-related, but multiple people working on multiple ways to prevent or provide treatment for cancer is an ideal alternative.

BakeryScan, Cyto-AiscAN and nanobots are smaller pieces in a large puzzle. They don’t have every solution yet, but AI and technology may end up becoming a part of every facet of the healthcare experience. AI can help perform administrative duties, monitor a patient’s condition, use data to make informed medical decisions and more.

All of this progress is a great start, but much more work needs to be done. Cyto-AiscAN is only being tested in two hospitals. Finding facilities to test these new products on real people isn’t a simple task. A result of that is that it takes longer for new technology to become the norm. Yes, technology needs to be tested thoroughly before it’s adopted, but keeping an open mind could one day save lives.