Mapping CT scan locations on computational humans may improve patient dose monitoring

A new algorithm can automatically map CT scan locations of patients on computational human phantoms may trump manual mapping techniques for patient dose monitoring, clinical trials and epidemiologic studies, detailed in a study published online Sept. 5 in the Journal of Digital Imaging.  

Researchers led by Choonsik Lee, PhD, dosimetry unit head and a senior investigator at the National Cancer Institute, developed the algorithm to compare a two-dimensional skeletal mask generated from CT scans with that of an entire body computational human phantom.  

First, the algorithm was used to automatically map scan locations of the CT images on a computation human phantom, according to the researchers. Radiologists then manually mapped the same CT images to compare and organ doses for both automated and manual mapping locations were calculated and compared.  

Results included the following:  

• Visual comparison showed excellent agreement between manual and automatic mapping locations for neck, chest, and abdomen-pelvis CT scans.  
• The difference in mapping locations averaged over the start and end in the five patients was less than 1 centimeter for all neck, chest, and abdomen-pelvis scans: 0.9, 0.7, and 0.9 cm for neck, chest, and abdomen-pelvis scans.  
• Five cases out of ten in the neck scans show zero difference between the average manual and automatic mappings.  
• Average of absolute dose differences between manual and automatic mappings was 2.3, 2.7, and 4.0 percent for neck, chest, and abdomen-pelvis scans.  

“The automatic mapping algorithm provided accurate scan locations and organ doses compared to manual mapping,” Lee et al. wrote.  

The research was funded by the intramural program of the National Institutes of Health, National Cancer Institute, Division of Cancer Epidemiology and Genetics.