Screws tighten on lung cancer with cross-field AI analytics
Deep neural networks are capable of tying oncological findings from genetic testing with those from medical imaging and biopsy analysis to not only validate previously discovered connections among and between the three fields but also uncover new ones.
So found researchers at UCLA whose clinical focus for the challenge was non-small cell lung cancer (NSCLC).
The team’s work was published online May 8 in the Journal of Medical Imaging.
Radiological scientist William Hsu, PhD, and colleagues trained their neural networks on 262 public datasets, later testing it on an additional set of 89.
Using an AI-trialing method called gene masking, the team drew out associations between gene subsets and one or another finding on CT scans and molecular histology slides.
They found the neural networks outperformed competing cancer classifiers at identifying and distinguishing between different cancer types.
Additionally, the system reproduced known associations across the three fields while illuminating heretofore unreported connections.
“This work demonstrates neural networks’ ability to map gene expressions to radiomic features and histology types in NSCLC and to interpret the models to identify predictive genes associated with each feature or type,” the authors write.
Commenting to the International Society for Optics and Photonics, Hsu suggests the direction clinical applications might take going forward.
“While radiogenomic associations have previously been shown to accurately risk stratify patients, we are excited by the prospect that our model can better identify and understand the significance of these associations,” Hsu says. “We hope this approach increases the radiologist’s confidence in assessing the type of lung cancer seen on a CT scan. This information would be highly beneficial in informing individualized treatment planning.”
The study is available in full for free.
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.