The recently inaugurated American College of Radiology Imaging Network randomized trial comparing virtual colonoscopy with its traditional counterpart has cooled the debate about which technique is better. Participants at the annual American Roentgen Ray Society meeting in May seemed to reserve judgment pending results from the National CT Colonography Trial.
The recently inaugurated American College of Radiology Imaging Network randomized trial comparing virtual colonoscopy with its traditional counterpart has cooled the debate about which technique is better. Participants at the annual American Roentgen Ray Society meeting in May seemed to reserve judgment pending results from the National CT Colonography Trial.
The research continues, however, into making virtual colonoscopy quicker, easier, and more accurate. Two studies presented at the meeting reported the effects of a prepless protocol. The researchers found that eliminating the irksome preparation might improve patient compliance, but it also interferes with radiologists' accuracy and proved troublesome for computer-aided detection.
Dr. Abraham Dachman, a professor of clinical radiology at the University of Chicago, and his colleagues in the U.S. and Belgium compared 12 patients who had a prepless protocol with 14 who underwent a mild prep. Stool tagging was used for both groups. Studies from the prepless group were more difficult to interpret, requiring 2D reads in both lung and bone windows. Residual stool often made the 3D reads impossible.
The prepless group didn't score much better on comfort: They gave the experience a rating of 1.2, compared with 1.8 for the mild prep group, on a 10-point comfort scale in which 1 was best and 10 was worst.
Prepless patients also had three times as many segments with stool than the other group as well as a larger percentage of mucosal surface covered by stool. This statistic is important, Dachman said, because it may indicate the number of polyps hidden in the stool.
"No one has done this type of data analysis before," he said.
In a second study from the same institution, 24 of the original 26 patients were again separated into a prepless and a mild prep group. Researchers found that CAD systems that had been trained on cleansed colons faltered when patients had not undergone cathartic cleansing.
Dr. Hiroyuki Yoshida, an associate professor of radiology at the University of Chicago, and colleagues trained an in-house CAD system using 121 independently collected CT colonography cases in which all colons had been rigorously cleansed but no tagging was used.
During the training, the CAD system posted 90% sensitivity with a 1.2 false-positive rate per data set. When investigators used the CAD system for the same patient cohort as the Dachman study, the results differed considerably.
While CAD detected all polyps in both groups, it reported an average of 5.8 false positives for the prepless group. The rate dropped to an average of 2.8 false positives for the prepped cohort.
In the prepless patients, 75% of the false positives were caused by untagged stool and 12% by partially tagged stool. In the prepped group, untagged stool accounted for 21% of the false positives and folds in the colon for 36% of false positives.
The researchers concluded that some cleansing is still required to enhance interpretation and polyp detection.
What is the Best Use of AI in CT Lung Cancer Screening?
April 18th 2025In comparison to radiologist assessment, the use of AI to pre-screen patients with low-dose CT lung cancer screening provided a 12 percent reduction in mean interpretation time with a slight increase in specificity and a slight decrease in the recall rate, according to new research.
The Reading Room: Racial and Ethnic Minorities, Cancer Screenings, and COVID-19
November 3rd 2020In this podcast episode, Dr. Shalom Kalnicki, from Montefiore and Albert Einstein College of Medicine, discusses the disparities minority patients face with cancer screenings and what can be done to increase access during the pandemic.
Can CT-Based AI Radiomics Enhance Prediction of Recurrence-Free Survival for Non-Metastatic ccRCC?
April 14th 2025In comparison to a model based on clinicopathological risk factors, a CT radiomics-based machine learning model offered greater than a 10 percent higher AUC for predicting five-year recurrence-free survival in patients with non-metastatic clear cell renal cell carcinoma (ccRCC).
Could Lymph Node Distribution Patterns on CT Improve Staging for Colon Cancer?
April 11th 2025For patients with microsatellite instability-high colon cancer, distribution-based clinical lymph node staging (dCN) with computed tomography (CT) offered nearly double the accuracy rate of clinical lymph node staging in a recent study.