Computer-aided detection may lend an extra punch to the diagnosis of pulmonary embolism with multislice and dual-energy CT scanners. Studies by French and German researchers have shown it helps detect small, hard-to-spot clots lodged in the lungs’ periphery.
Computer-aided detection may lend an extra punch to the diagnosis of pulmonary embolism with multislice and dual-energy CT scanners. Studies by French and German researchers have shown it helps detect small, hard-to-spot clots lodged in the lungs' periphery.
Multislice CT angiography has become the diagnostic standard in pulmonary embolism, with a well-established accuracy in the detection of proximal lesions. But CTA detection of peripheral PE remains challenging and time-consuming, and it may lead to false positives. Can CAD pick those small clots missed by the radiologist's naked eye in the first read? Yes, it can, said principal investigator of a study from France Dr. Marion Dewailly from the University of Lille.
Dewailly and colleagues enrolled 74 patients who underwent diagnosis of peripheral acute PE with dual-source CTA and a CAD software (PECAD 7, Siemens). Two radiologists interpreted high- and low-quality angiograms done under varied conditions, including ECG-gated and nongated scanning.
The investigators found neither the scanning conditions nor the overall image quality influenced CAD results, which proved sensitive for detection of peripheral PE. Dewailly released her group's results Saturday in Vienna.
CAD detected 80 of 93 peripheral clots found in 21 patients. CAD's sensitivity for peripheral, segmental, and subsegmental clots was 86%, 78.3%, and 93.3%, respectively. Researchers found a mean number of 6.4 false positives per patient, mainly linked to hilar lymph nodes and perivascular airspace consolidation. The sensitivity for CAD was 84% for ECG-gated scans, 82% for dual-energy scans, and 76% for nongated scans. The difference was not statistically significant.
"CAD can complement the visual assessment of pulmonary clots by the radiologist. And dual energy can help with automatic detection of peripheral clots and identification of perfusion defects," she said.
In a study from Germany, researchers at sites in Nordhausen, Neustadt, and Jena compared CAD's performance against that of radiologists. They performed 160 CT examinations on patients suspected with PE and evaluated whether highlighted lesions were true or false positives.
The researchers found that CAD allows a better analysis of peripheral PE and that most false-positive markers highlight perivascular soft tissue. They also found an association between the total number of CAD markers per case and D-dimer results, which bore diagnostic and prognostic connotations.
CAD proves to be a useful tool to differentiate peri- and intravascular tissue and to diagnose mainly peripherally located emboli, according to principal investigator and presenter Dr. Ansgar Malich.
"We use that tool as a second reader," Malich said.
Could a Deep Learning Model for Mammography Improve Prediction of DCIS and Invasive Breast Cancer?
April 15th 2024Artificial intelligence (AI) assessment of mammography images may significantly enhance the prediction of invasive breast cancer and ductal carcinoma in situ (DCIS) in women with breast cancer, according to new research presented at the Society for Breast Imaging (SBI) conference.
Mammography-Based AI Abnormality Scoring May Improve Prediction of Invasive Upgrade of DCIS
April 9th 2024Emerging research suggests that an artificial intelligence (AI) score of 75 or greater for mammography abnormalities more than doubles the likelihood of invasive upgrade of ductal carcinoma in situ (DCIS) diagnosed with percutaneous biopsy.
Mammography Study: AI Improves Breast Cancer Detection and Reduces Reading Time with DBT
April 3rd 2024An emerging artificial intelligence (AI) model demonstrated more than 12 percent higher specificity and reduced image reading time by nearly six seconds in comparison to unassisted radiologist interpretation of digital breast tomosynthesis (DBT) images.