Developed an AI cash register that instantly identifies bread. High expectations for application to cancer researchBRAIN Co.,Ltd.
<Series 1 (Total 2 series)>
The reality of a declining birthrate and aging population along with a shrinking workforce casts a shadow over Japan's future. The key phrase in this context is "improved productivity." If you can't increase the workforce, devise a way to increase the amount of work one person can do. Brain has developed the product "BakeryScan" that achieves this improved productivity in bakeries. How did the company arrive at this product and technology? Here is our two-part interview with President Hisashi Kanbe.
AI automatically identifies bread on trays
Some popular bakery shops even have a line, mainly due to cumbersome register transactions. Bakery cashiers need to correctly identify the types of bread, enter each one into the register, check out the customer, and even pack the bread.
Brain has developed "BakeryScan", a system in which AI instantly identifies the types of bread on the tray and displays the total amount. The checkout is done at a self-service cash register, so the cashier can basically focus on packaging and just check that the AI didn't make any identification mistakes.
"It takes one month for fast learners to learn about 100 types of bread, and three months for slow learners. Eliminating the need for this labor was a big deal. It takes less than a day to learn to use it. In addition, we have added an AI function that learns on its own to improve accuracy."
With "BakeryScan," you can simplify register transactions such as bread identification, calculating the total, and checkout.
Praised by stores and customers, and even Google's AI staff
As soon as it was released in March 2013, there was a great response, and it was featured in various media. And in 2017, a Google AI employee tweeted about the BakeryScan, saying that Japan has an amazing AI cash register. In addition to the publicity from receiving the Good design Award of Small and Medium Enterprises 2015, as well as the Excellence Award at the 6th Monodzukuri Nippon Grand Awards, stores and customers that actually use the product have increasingly voiced their satisfaction. The number of installed units has been steadily increasing, with up to 600 as of August 2019.
"In addition, we have received inquiries from all quarters to apply this technology. Commercialization is already underway, including image identification for more than 200 types of amulets at shrines and temples, as well as technical support for research and studies classifying and analyzing the data of about 15 million images in the field of industrial technology. We've also been selling a device for several years that identifies a scanned drug from out of 10,000 drug types."
Applying the technology to cancer cell identification. Attracting attention from the medical community
Furthermore, the company is now focusing on developing technology to detect cancer cells. The Louis Pasteur Center for Medical Research (Kyoto Prefecture) learned about the product on TV and contacted Brain in 2017.
"Even though many of the technologies currently used for cancer cell detection also use AI, it's deep learning AI. However, deep learning requires a massive amount of data, has a misdiagnosis rate of about 10%, and sometimes the basis for the result is unknown. On the other hand, our product's image identification technology does not require much data in the first place, and identifies individual differences, so at least the basis of the diagnosis is clearly shown. And with its ability to learn, it's getting more and more accurate."
The company presented at a conference in 2018 for its debut in medical circles, and an invitation to an international conference soon arrived. The key "cancer identification accuracy" is almost 100% for some cancers, and now the company will try identifying other types of cancer. A bakery cash register system developed by a Japanese company leads to minimizing the threat of cancer, a long-time enemy of mankind... This could be in store for the near future.
However, the journey so far has not been smooth. In the next part we will delve into the details of how the company ended up creating new value in the form of BakeryScan.
Series "You can only win real freedom by taking risks"
Series 1 Developed an AI cash register that instantly identifies bread. High expectations for application to cancer research
Series 2 You can only win real freedom by taking risks
Brain Co., Ltd.
President Hisashi Kanbe
Brain was founded as a PC shop in 1982 in Nishiwaki City, Hyogo Prefecture, the site of a thriving textile industry. Later, while developing phrase conversion software and NHK's character information display system, as well as developing technology for image identification for fibers, computer system R&D became a pillar of its business. "BakeryScan" was developed in 2013, and various solutions driven by this technology have been rolled out.
Coverage date November 15, 2019
Solves the Problems of “High Barrier,” “Complexity of Implementation/Operation,” and “Security Concerns” in the Implementation of IT in SMEs
Committed to reducing environmental impact using bioplastics that have new possibilities
Mitsubishi Chemical Corporation
New material manufactured using paper powder with a unique kneading technique. Usage of plastics, and greenhouse gas emissions are both reduced
Eco Research Institute Ltd.