Quantitative elemental imaging with neutrons for breast cancer diagnosis: A GEANT4 study

Abstract

Neutron stimulated emission computed tomography (NSECT) has been proposed as an early cancer-detection technique with the capability of 3-D tomographic imaging for identification of malignant tumors. In previous work we have demonstrated the ability of the technique to differentiate between normal and malignant breast tumors based on the concentration of cancer-marking elements in the tissue. Here we present tomographic images from a breast phantom with benign and malignant tumors simulated in GEANT4. A simulated model of the NSECT system was developed in GEANT4, along with phantoms corresponding to the human breast with benign and malignant tumors. Each tumor within the breast was given a different concentration of cancer-marking trace elements based on values reported in literature. The phantom was scanned with a 5-MeV neutron beam over a 180-degree angle. Tomographic images were reconstructed for six elements of interest from 10 different spectral lines. The results showed excellent agreement between the location of the tumor and the concentration of trace element detected in gamma lines from bromine, cesium, sodium and zinc. These results demonstrate the ability of NSECT in quantitative elemental imaging for breast cancer detection.