Regional dopamine transporter (DaT) uptake in sub-cortical regions, as derived from DaT SPECT images, may help provide markers to measure severity of Parkinson disease. However, to achieve this goal, there is a need for methods to reliably reconstruct, segment, and then quantify these uptakes from DaT-SPECT images. To address this need, we have developed new methods for SPECT reconstruction that can provide improved quantification. We have also developed novel deep-learning-based techniques to segment regions such as globus pallidus, which are visually impossible to demarcate in SPECT images.


Z. Liu, H. Moon, R. Laforest, J. Perlmutter, S. A. Norris, A. K. Jha, “Fully automated 3D segmentation of dopamine transporter SPECT images using an estimation-based approach”, (arxiv)

H.S. Moon, Z. Liu, M. Ponisio, R. Laforest, and A. K. Jha, A physics-guided and learning-based estimation method for segmenting 3D DaT-Scan SPECT images. Journal of Nuclear Medicine61(supplement 1), pp.10-10, 2020 (selected for Young Investigator Symposium, Highest scored abstract in the Data Sciences category) (link)

M. A. Rahman, R. Laforest, A. K. Jha, “A list-mode OSEM-based attenuation and scatter compensation method for SPECT”, IEEE Symposium on Biomedical Imaging 2020 (link)

Z. Yu, M. A. Rahman, A. K. Jha, “A Transmission-less Attenuation Compensation Method for Brain SPECT Imaging”, IEEE International Symposium on Biomedical Imaging 2020