AI tool analyzes tumor images, determines lung cancer types
An NYU School of Medicine-led research team has developed an artificial intelligence tool that analyzes tissue and can tell apart two lung cancer types.
Researchers also contend that the tool can identify genetic changes in patients’ tumors.
The team trained a deep convolutional neural network using whole-slide images obtained from The Cancer Genome Atlas to accurately and automatically classify them into adenocarcinoma and squamous cell carcinoma—the most prevalent subtypes of lung cancer—or normal lung tissue.
Results of their study, published last week in the journal Nature Medicine, found that their AI tool could distinguish with 97 percent accuracy between adenocarcinoma and squamous cell carcinoma and that its performance was comparable to that of pathologists.
In addition, researchers trained the convolutional neural network to predict the 10 most commonly mutated genes in adenocarcinoma and found that six of them can be predicted from pathology images, with an accuracy that ranged from 73 to 86 percent, depending on the gene.
“Visual inspection of histopathology slides is one of the main methods used by pathologists to assess the stage, type and subtype of lung tumors,” state the authors. “These findings suggest that deep-learning models can assist pathologists in the detection of cancer subtype or gene mutations. Our approach can be applied to any cancer type.”
The problem with the current genetic tests used to confirm the presence of mutations is that they can take two weeks to get the results, according to the authors.
“Delaying the start of cancer treatment is never good,” says Aristotelis Tsirigos, the study’s senior author and associate professor in the Department of Pathology at NYU Langone’s Perlmutter Cancer Center. “Our study provides strong evidence that an AI approach will be able to instantly determine cancer subtype and mutational profile to get patients started on targeted therapies sooner.”
“In our study, we were excited to improve on pathologist-level accuracies, and to show that AI can discover previously unknown patterns in the visible features of cancer cells and the tissues around them,” adds Narges Razavian, co-corresponding author and assistant professor in the departments of Radiology and Population Health. “The synergy between data and computational power is creating unprecedented opportunities to improve both the practice and the science of medicine.”
Going forward, Tsirigos says that if the accuracy of the AI tool can be improved to more than 90 percent in determining which genes are mutated in a given cancer, then it will become usable in clinical practice.