CAD stems interobserver variability in lung lesions

May 1, 2006

Radiologists are often called to review scans of patients who have undergone therapeutic treatment for cancer. In such scenarios, accurate assessment of tumor volume is key to knowing whether the therapy is effective.

Radiologists are often called to review scans of patients who have undergone therapeutic treatment for cancer. In such scenarios, accurate assessment of tumor volume is key to knowing whether the therapy is effective.

Dr. Francesco Fraioli and colleagues from the University of Rome La Sapienza evaluated a CAD system's ability to assess tumor volume in nine patients undergoing chemotherapy for lung metastases. They applied RECIST criteria to consider the disease as stable, increased, or decreased and compared CAD findings to observers' measurements.

Researchers included 24 nodules in the study, ranging from 5 mm to 18 mm before therapy and 4 mm to 20 mm post-therapy. Observers agreed in the assessment of therapy response in 21 nodules: eight were considered increased in volume, and 13 were judged stable.

Observers and CAD measurements disagreed in three nodules: two considered as stable by radiologists and increased by CAD; one considered increased by radiologists and stable by CAD.

Fraioli, who presented the study at the 2006 European Congress of Radiology, concluded that CAD volumetric measurements allow an easy and objective evaluation, reducing interobserver variability.