Japanese team uses AI to predict cancer risk based on fatty liver images, A group of Japanese researchers announced that it has developed a deep learning model to predict the risk of developing cancer based on fatty liver images.
The team, led by the University of Tokyo, created the model using single digital images of fatty liver disease collected from 46 people who developed liver cancer within seven years of the biopsy and 639 others who did not.
For the artificial intelligence application project, the researchers collected such images from a total of 2,432 people who had undergone the live tissue examination at nine medical institutions in the country.
The model has been proven to predict the risk of developing cancer with an accuracy of 82.3%, compared to 78.2% for biopsy-based manual analyses, the researchers said.
They also saw that AI assessed cell dysplasia and decreasing fat deposition, despite the progression of fatty liver disease, as risk factors for cancer and predicted a high probability that mild fibrosis would develop into cancer.
“It is also possible to create deep-learning models to predict cancer development from images collected through gastric and intestinal mucosal biopsies,” said group member Ryosuke Tateishi, associate professor at the university’s Graduate School of Medicine.
The team’s research paper has been published in the online edition of Journal of Hepatology, a specialty American medical journal.
