Saman Sotoudeh-Paima

Saman Sotoudeh-Paima

Auto-generated profile, based on publicly available information

Saman Sotoudeh-Paima

Center for Virtual Imaging Trials, Carl E. Ravin Advanced Imaging Laboratories, Department of Radiology, Duke University School of Medicine, 2424 Erwin Rd, Hock Plaza, Ste 302, Durham, NC 27705

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 March 2022 in Computers in Biology and Medicine
Verified
Journal Article
Closed access
SSSaman Sotoudeh-Paima
Saman Sotoudeh-Paima
AJAta Jodeiri
Ata Jodeiri
FHFedra Hajizadeh
Fedra Hajizadeh
HSHamid Soltanian-Zadeh
Hamid Soltanian-Zadeh

16 July 2024 in IEEE Transactions on Biomedical Engineering
Verified
Journal Article
Closed access
MZMojtaba Zarei
Mojtaba Zarei
SSSaman Sotoudeh-Paima
Saman Sotoudeh-Paima
CMCindy McCabe
Cindy McCabe
EAEhsan Abadi
Ehsan Abadi
ESEhsan Samei
Ehsan Samei
Computed Tomography (CT) quantification is affected by the variability in image acquisition and rendition. This paper aimed to reduce this variability by harmonizing the images utilizing physics-based deep neural networks (DNNs). Methods: An adversarial generative network was trained on virtual CT images acquired under various imaging conditions using a virtual imaging platform with 40 computational patient models. These models featured anthropomorphic lungs with different levels of pulmonary diseases, including nodules and emphysema. Imaging was conducted using a validated CT simulator at two dose levels and varying reconstruction kernels. The trained model was tested on an independent virtual test dataset and two clinical datasets. Results: On the virtual test set, the harmonizer improved the structural similarity index from 79.3 $\pm$ 16.4% to 95.8 $\pm$ 1.7%, normalized mean squared error from 16.7 $\pm$ 9.7% to 9.2 $\pm$ 1.7%, and peak signal-to-noise ratio from 27.7 $\pm$ 3.7 dB to 32.2 $\pm$ 1.6 dB. Moreover, the harmonized images yielded more precise quantification of emphysema-based imaging biomarkers for lung attenuation, LAA -950 from 5.6 $\pm$ 8.7% to 0.23 $\pm$ 0.16%, Perc 15 from 43.4 $\pm$ 45.4 HU to 20.0 $\pm$ 7.5 HU, and Lung Mass from 0.3 $\pm$ 0.3 g to 0.1 $\pm$ 0.2 g. In clinical data, the harmonizer reduced biomarker variability by an average of 70%. For lung nodules, harmonized images improved the detectability index by 6.5-fold and DNN-based precision by 6%. Conclusion: The proposed harmonizer significantly enhances image quality and quantification accuracy in CT imaging. Significance: The study demonstrated the potential utility of image harmonization for consistent CT image quality and reliable quantification, which is crucial for clinical applications and patient management.

1 April 2026 in Radiology: Cardiothoracic Imaging
Verified
Journal Article
Closed access
SSSaman Sotoudeh-Paima
Saman Sotoudeh-Paima
DLDavid Lynch
David Lynch
SHStephen Humphries
Stephen Humphries
ESEhsan Samei
Ehsan Samei
EAEhsan Abadi
Ehsan Abadi
A physics-based image harmonization method was developed and validated; it improved CT-based lung density reproducibility by aligning images to a reference quality index of noise, resolution, and lung volume.

10 May 2025 in British Journal of Radiology
Verified
Journal Article
Closed access
SSSaman Sotoudeh-Paima
Saman Sotoudeh-Paima
ESEhsan Samei
Ehsan Samei
EAEhsan Abadi
Ehsan Abadi
Objectives: To evaluate how photon-counting CT (PCCT) reproducibly quantifies lung density compared to conventional CT (energy-integrating CT or EICT) and assess the relevance of image quality. Methods: In this retrospective analysis of a prospective study, subjects underwent acquisitions using EICT and PCCT systems (August-December 2021) on the same day. PCCT projections were reconstructed with 4 kernels and 2 voxel sizes using 2024 reconstruction software (ReconCT, Siemens). PCCT and EICT data were compared using lung density measurements: percentage of low-attenuation areas <−950 Hounsfield unit (HU) (LAA-950), 15th percentile of lung density histogram (Perc15), and mean lung density (MLD). Measurement reproducibility and its relation to image quality parameters (modulation transfer function and noise) were analyzed. Results: Fifty-four subjects (mean age, 67.4 years ± 9.6, 30 female) were studied. PCCT images had the closest agreement to EICT under identical voxel size and similar kernels, yielding lower LAA-950 (−0.9% ± 2.7, P < 0.05), higher Perc15 (1.4 HU ± 7.9, P > 0.05), and higher MLD (2.4 HU ± 10.8, P > 0.05). At higher PCCT resolution (1024 × 1024 matrix, 0.2 mm slice thickness), the smoother quantitative kernel (Qr36) improved agreement, with differences of −0.5% ± 3.0, −0.1 HU ± 9.8, and 2.8 HU ± 11.1, for LAA-950, Perc15, and MLD (P > 0.05). Image quality parameters strongly correlated with density measurements (RPerc152=85.3%). Conclusions: PCCT yields reproducible lung density measurements comparable to EICT, with optimal reproducibility dependent on optimized reconstruction parameters and explainable by image quality parameters. Advances in knowledge: Lung density reproducibility is critical for integrating PCCT into EICT-based workflows. In our study, PCCT aligned more closely with EICT using similar kernels at identical voxel sizes, or smoother kernels at smaller voxel sizes.

1 May 2023
Verified
Conference Paper
Closed access
AKA. Kavuri
A. Kavuri
MNM.G. Nejad
M.G. Nejad
SSS. Sotoudeh-Paima
S. Sotoudeh-Paima
HMH.P. McAdams
H.P. McAdams
DLDavid Lynch
David Lynch
EAE. Abadi
E. Abadi