Radiologists reading only axial CT images could be missing more than 12% of traumatic spinal fractures. They could improve their sensitivity by incorporating other views as well, according to researchers at Harvard Medical School.
Radiologists reading only axial CT images could be missing more than 12% of traumatic spinal fractures. They could improve their sensitivity by incorporating other views as well, according to researchers at Harvard Medical School.
Spinal fractures are seen in 4% to 6% of blunt trauma patients, and they are being increasingly diagnosed with CT imaging, said Dr. Ajay Singh, an associate professor of medicine at Harvard, at the American Roentgen Ray Society meeting.
Singh and colleagues performed multislice CT scans on 63 patients whose spinal fractures had been diagnosed on chest and abdomen CT. Ninety-six of the fractures were in the lumbar spine, and 59 in the thoracic.
Two radiologists blindly interpreted the images, which researchers randomly combined with 30 normal chest and abdomen CT scans. The radiologists first made a diagnosis using axial images alone. They then interpreted the images using axial images along with sagittal and coronal reconstructions.
For the thoracic spine, the first reader's sensitivity improved from 71% using axial images only to 91% using the combined views. This reader's sensitivity in the lumbar spine improved from 93% to 99%.
The second reader had similar results, boosting sensitivity from 77% to 94% in the thoracic spine and from 87% to 95% in the lumbar spine with the additional views.
Overall, information provided by sagittal and coronal reconstructions led to a 12.1% to 12.7% improvement in fracture diagnosis for both readers, Singh said.
Compared with traditional x-rays, MSCT images are less likely to be degraded by technical factors, and motion artifacts are minimized, offering more opportunities for fracture diagnosis, Singh said.
For more information from the online Diagnostic Imaging archives:
Digital imaging makes inroads in orthopedics
Speedy CT makes the most of the 'golden hour' in trauma care
CT slices yield to 'computerized volumetric imaging'
Axial slices fail to catch pathologies seen on sagittal or coronal views
AI Mammography Platform Shows Promising Results for Detecting Subclinical Breast Cancer
October 3rd 2024Mean artificial intelligence (AI) scoring for breasts developing cancer was double that of contralateral breasts at initial biennial screening and was 16 times higher at the third biennial screening, according to a study involving over 116,000 women with no prior history of breast cancer.
FDA Clears Software for Enhancing CCTA Assessment of Atherosclerosis
October 1st 2024Through analysis of coronary computed tomography angiography (CCTA) images, the PlaqueIQ software provides quantification and classification of atherosclerosis, a common cause of myocardial infarction (MI) and ischemic stroke.