CANARY in the reading room: Software characterizes lung nodule risk

Newly developed software offers a method to assess high-resolution chest CT scans and noninvasively characterize pulmonary nodules, according to a pilot study published in the April issue of the Journal of Thoracic Oncology.

The tool, computer-aided nodule assessment and risk yield, or CANARY, was shown to accurately identify pulmonary adenocarcinoma, according to Tobias Peikert, MD, of the Mayo Clinic in Rochester, Minn., and colleagues.

“On the basis of the CANARY analysis and the consensus histopathology of the lesion, we successfully designed and optimized a decision algorithm for the noninvasive risk stratification into aggressive and indolent lesions,” wrote the authors. “Through automated volumetric quantitation of the lesions, CANARY provides the opportunity for the noninvasive preoperative characterization and risk stratification of pulmonary nodules of the lung adenocarcinoma spectrum.”

Pixel-by-pixel, CANARY matches CT imaging data of a lung nodule example cases, establishing signatures without clinical input and based solely on radiologic characteristics, according to the authors.

In the pilot trial, measurements of histopathologic tissue were developed in a training set of 54 pulmonary nodules of the adenocarcinoma spectrum and validated in 86 consecutively resected nodules.

Peikert and colleagues reported the sensitivity, specificity, positive predictive value and negative predictive value for detecting aggressive lesions were 95.4, 96.8, 95.4 and 96.8 percent, respectively, in the training set, and 98.7, 63.6, 94.9 and 87.5 percent, respectively, in the validation set.

To hear more about CANARY from Peikert, the study's senior author, watch the video below:

brightcove.createExperiences();