A convolutional neural network trained and internally validated on more than 15,000 lung CT scans has correctly reclassified indeterminate pulmonary nodules (IPNs) into low-risk or high-risk categories.
The AI system proved its mettle in external validation testing. Here it reclassified more than a third of nearly 600 real-world nodules, malignant and benign, with significantly better accuracy than conventional risk models with which researchers and developers compared it.
The study was conducted at Vanderbilt University and is published in full for free in the American Journal of Respiratory and Critical Care Medicine.
In their discussion section, lead author Pierre Massion, MD, and colleagues comment that their neural network’s demonstration of high performance may mean the technology has potential for changing patient management. If so, it would accomplish this mainly by obviating unnecessary invasive procedures and reducing delays in diagnosis.
Vanderbilt’s internal coverage of the development notes that adherence to current guidelines offered by the American College of Radiology and the American College of Chest Physicians can be variable, leading to subjective nodule classifications.
The present study, the school states, “is the first to validate a risk-stratification tool on multiple independent cohorts and to show reclassification performance that is significantly superior to existing risk models.”
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.