HomeTechnologyArtificial intelligenceMIT researchers develop an AI model that can detect future lung cancer...

MIT researchers develop an AI model that can detect future lung cancer risk | MIT News

The name Sybil has its origins in the oracles of ancient Greece, also known as sibyls: female figures relied upon to impart divine knowledge of the unseen and the almighty past, present, and future. Now the name has been dug up from ancient times and donated to an artificial intelligence tool for lung cancer risk assessment being developed by researchers at MIT’s Abdul Latif Jameel Clinic for Machine Learning in Health, Mass General Cancer Center (MGCC) and Chang Gung Memorial Hospital ( CMH).

Lung cancer is the number 1 deadliest cancer in the world, resulting in 1.7 million deaths worldwide in 2020killing more people than the next three deadliest cancers combined.

“It is the biggest cancer killer because it is relatively common and relatively difficult to treat, especially if it has reached an advanced stage,” says Florian Fintelmann, MGCC thoracic interventional radiologist and co-author of the new work. “In this case, it’s important to know that if you catch lung cancer early, the long-term outcome is significantly better. Your five-year survival rate is closer to 70 percent, while if you catch it when it’s advanced, the five-year survival rate is just under 10 percent.”

Although there has been a wave of new therapies to fight lung cancer in recent years, the majority of lung cancer patients still succumb to the disease. Low-dose computed tomography (LDCT) scans of the lungs are currently the most common way patients are screened for lung cancer, hoping to find it in its earliest stages, when it can still be surgically removed. Taking the screening a step further, Sybil analyzes the LDCT image data without the help of a radiologist to predict a patient’s risk of developing a future lung cancer within six years.

In their new article published in the Journal of Clinical Oncology, Jameel Clinic, MGCC and CGMH investigators showed that over six years Sybil achieved C-Indices of 0.75, 0.81 and 0.80 from several sets of lung LDCT scans from the National Lung Cancer Screening Trial (NLST), Mass General Hospital (MGH) and CGMH respectively – models with a C-index score greater than 0.7 are considered good and greater than 0.8 are considered strong. The ROC-AUCs for one-year forecast with Sybil scored even higher, ranging from 0.86 to 0.94, with 1.00 being the highest possible score.

Must Read
New machine learning tool discovers mutational signature that links bladder cancer to tobacco smoking

Despite its success, the 3D nature of lung CT scans made it challenging to build Sybil. Co-first author Peter Mikhael, an MIT doctoral student in electrical engineering and computer science associated with Jameel Clinic and the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), compared the process to “trying to find a needle in a haystack.” .” The imaging data used to train Sybil showed largely no signs of cancer, because early-stage lung cancer occupies small areas of the lung — just a fraction of the hundreds of thousands of pixels that make up each CT scan. Denser areas of lung tissue are known as lung nodules, and while they have the potential to cause cancer, most are not and can be the result of healed infections or airborne irritants.

To ensure Sybil could accurately assess cancer risk, Fintelmann and his team labeled hundreds of CT scans with visible cancerous tumors that would be used to train Sybil before testing the model on CT scans with no detectable signs of cancer.

MIT electrical engineering and computer science PhD student Jeremy Wohlwend, co-author of the paper and Jameel Clinic and CSAIL affiliate, was surprised by how high Sybil scored despite not having any visible cancer. “We found that while we [as humans] couldn’t quite see where the cancer was, the model might still have some predictive power about which lung would eventually develop cancer,” he recalls. “Know [Sybil] was able to highlight which side was the most likely side was really interesting to us.

Must Read
Biomarker tests offer 'real potential' to treat lung cancer

Co-author Lecia V. Sequist, a medical oncologist, lung cancer expert and director of the Center for Innovation in Early Cancer Detection at MGH, says the results the team achieved with Sybil are important “because lung cancer screening is not being used to its full potential in the US or worldwide, and Sybil may be able to help us bridge this gap.

Lung cancer screening programs are underdeveloped in regions of the United States most affected by lung cancer due to a variety of factors. These range from stigma against smokers to political and policy landscape factors such as Medicaid expansion, which varies from state to state.

In addition, many patients diagnosed with lung cancer today have never smoked or are former smokers who quit more than 15 years ago — traits that make both groups ineligible for CT screening for lung cancer in the United States.

“Our exercise data consisted only of smokers, as this was a necessary criterion for participation in the NLST,” says Mikhael. “In Taiwan, they screen nonsmokers, so our validation data is expected to include people who did not smoke, and it was exciting to see Sybil generalize well to that population.”

“An exciting next step in the research is testing Sybil prospectively in people at risk for lung cancer who either did not smoke or who quit decades ago,” says Sequist. “I treat such patients every day in my lung cancer clinic and it is understandable that they find it difficult to accept that they would not have been candidates for screening. Perhaps that will change in the future.”

Must Read
UK Accepts Marketing Authorization Application for Sugemalimab in Metastatic NSCLC

There is a growing population of lung cancer patients who are classified as non-smokers. Female nonsmokers are more likely to develop lung cancer than nonsmokers. worldwide, more than 50 percent of women diagnosed with lung cancer are nonsmokers, compared with 15 to 20 percent of men.

MIT professor Regina Barzilay, a paper co-author and head of the Jameel Clinic AI faculty, who is also a member of the Koch Institute for Integrative Cancer Research, credits MIT and MGH’s joint efforts on Sybil to Sylvia, the sister of a close friend of Barzilay and one of Sequist’s patients. “Sylvia was young, healthy and athletic – she never smoked,” recalled Barzilay. “When she started coughing, neither her doctors nor her family initially suspected that the cause might be lung cancer. When Sylvia was finally diagnosed and met with Dr. Sequist, the disease had progressed too far to reverse its course. As we mourned after Sylvia’s death, we couldn’t stop thinking how many other patients have similar trajectories.”

This work was supported by the Bridge Project, a collaboration between the Koch Institute at MIT and the Dana-Farber/Harvard Cancer Center; the MIT Jameel Clinic; Quanta computers; Stand up against cancer; the MGH Center for Innovation in Early Cancer Detection; the Bralower and Landry families; Upstage lung cancer; and the Eric and Wendy Schmidt Center at MIT and Harvard’s Broad Institute. The Linkou CGMH Cancer Center under the Chang Gung Medical Foundation provided data collection assistance and R. Yang, J. Song and their team (Quanta Computer Inc.) provided technical and computer support for analyzing the CGMH dataset. The authors thank the National Cancer Institute for access to NCI’s data collected through the National Lung Screening Trial, as well as the patients who participated in the study.



Please enter your comment!
Please enter your name here

Most Popular

Recent Comments