The discovery, announced in late January by a team of researchers from Harvard’s Massachusetts General Hospital and the Massachusetts Institute of Technology, is part of a growing medical trend to use algorithms to predict everything from breast cancer and prostate cancer. the likelihood of tumor regrowth. Although research is increasing, scientists say more testing needs to be done before these products are fully released in clinical settings.
The tool is called Sybil, after the prophet in ancient Greek literature. It is a deep learning model, which means that computers analyze huge sets of data to identify and categorize patterns. Sybil was trained over six years of lung scans of patients in the United States and Taiwan, researchers said.
The results of the study showed that Sybil obtained scores scientifically considered “good” and “strong” for predicting lung cancer over six years. It was stronger with its one-year prediction rates, the study scientists noted.
Lung cancer is “the biggest killer of cancer because it’s relatively common and relatively difficult to treat,” said Florian Fintelmann, interventional radiologist at the Massachusetts General Cancer Center and co-author of the study. “If you catch lung cancer early, the long-term outcomes are significantly better.”
Cancer is the second leading cause of death worldwide, and as advances in artificial intelligence software and computing power have increased, it has become a ripe area for researchers to apply the technology in hopes. help doctors make diagnoses.
Researchers are using artificial intelligence to track the progression of prostate cancer, breast cancer, and tumor regrowth after undergoing treatment.
Much of the technology involves analyzing large amounts of medical scans, datasets or images and then feed them into complex artificial intelligence software. From there, computers are trained to pick up images of tumors or other abnormalities, which the researchers say can be more accurate and faster than the human eye.
In recent years, there has been an increase in new therapies to fight lung cancer, the study researchers said, but many patients still die from the disease due to obstacles.
Those who are old and poor may not benefit from screenings due to limited federal funding. Many patients diagnosed with lung cancer have never smoked or are former smokers who quit more than 15 years ago, MIT researchers have said, making them ineligible for screenings in the United States.
For those who can get screened, the most common way is through low-dose CT scans, called LDCTs. Researchers created Sybil to speed up the screening process, allowing software to analyze LDCT images without the help of radiologists to predict cancer risk up to six years in advance.
But building Sybil was a challenge, the study authors said. Peter Mikhael, a researcher and affiliate of MIT’s Jameel Clinic and its Computer Science and Artificial Intelligence Lab, described it as “trying to find a needle in the haystack.”
Most of them imaging data to form Sybil did not contain overt signs of cancer, as early-stage lung cancer is found in small parts of the lung and can be difficult to see with the naked eye spot. To ensure that the software could assess cancer risk, the study team “tagged hundreds of CT scans with visible cancerous tumors” and fed them into Sybil before releasing the software on the CT scans with limited signs of cancer, the researchers said.
The team used datasets from the National Lung Screening Trial, Massachusetts General Hospital and Chang Gung Memorial Hospital in Taiwan. According to the study, some of the data was heavily skewed by white people.
Medical experts warn that cancer-fighting software needs more study before it can be used in the clinic, according to government scientists and research studies.
Researchers from Harvard and the Netherlands said the skills needed to translate information generated by AI algorithms remain in the “nascent stage”. Moreover, the benefits that AI can bring to medicine are currently quite limited. Even with these detection tools, physicians still need to make diagnoses, design treatment plans, and manage overall care.
Other medical experts point out that more testing needs to be done to see how well the software works on a variety of patients, using different scanners and tools. More work also needs to be done to show that the software actually benefits people, either by helping them live longer, preventing cancer, or saving time and money. The operation of the algorithms should be transparent, not a “black box”, they said.
The MIT researchers said they would continue their work.
“An exciting next step in the research will be to prospectively test Sybil in people at risk for lung cancer who did not smoke or quit decades ago,” said Lecia Sequist, director of the Center for Innovation in Early Cancer Detection at Massachusetts General Hospital.