Impact of tracer uptake rate on quantification accuracy of myocardial blood flow in PET: A simulation study DOI
Xiaotong Hong, Amirhossein Sanaat, Yazdan Salimi

и другие.

Medical Physics, Год журнала: 2025, Номер unknown

Опубликована: Май 8, 2025

Cardiac perfusion PET is commonly used to assess ischemia and cardiovascular risk, which enables quantitative measurements of myocardial blood flow (MBF) through kinetic modeling. However, the estimation parameters challenging due noisy nature short dynamic frames limited sample data points. This work aimed investigate errors in MBF a simulation study evaluate different parameter approaches, including deep learning (DL) method. Simulated studies were generated using digital phantoms based on cardiac segmentations from 55 clinical CT images. We employed irreversible 2-tissue compartmental model simulated 13N-ammonia scans under both rest stress conditions (220 cases each). The simulations covered K1 range 0.6 1.2 3.6 (unit: mL/min/g) myocardium. A transformer-based DL was trained dataset predict parametric images (PIMs) image validated 5-fold cross-validation. compared method with voxel-wise nonlinear least squares (NLS) fitting applied images, either Gaussian filter (GF) smoothing (GF-NLS) or nonlocal means (DNLM) algorithm for denoising (DNLM-NLS). Two patients coronary angiography (CTA) fractional reserve (FFR) enrolled test feasibility applying models data. showed clearer structures reduced noise traditional NLS-based methods. In terms mean absolute relative error (MARE), as values increased mL/min/g, overall bias myocardium estimates decreased approximately 58% 45% methods while reduction MARE 42% 18%. For data, 30% 70% GF-NLS contrast, DNLM-NLS (average: 42%) 20%) demonstrated significantly smaller changes varied. Regarding regional (±standard deviation), had 6.30% (±8.35%) K1, 1.10% (±8.21%) 6.28% (±14.05%) 10.72% (±9.34%) 1.69% (±8.82%) -10.55% (±9.81%) that an increase tracer uptake rate (K1) corresponded improved accuracy precision quantification, whereas lower resulted higher poorer estimates. Utilizing techniques approaches can mitigate noise-induced imaging.

Язык: Английский

Deep learning-based segmentation of ultra-low-dose CT images using an optimized nnU-Net model DOI Creative Commons
Yazdan Salimi, Zahra Mansouri, Chang Sun

и другие.

La radiologia medica, Год журнала: 2025, Номер unknown

Опубликована: Март 18, 2025

Abstract Purpose Low-dose CT protocols are widely used for emergency imaging, follow-ups, and attenuation correction in hybrid PET/CT SPECT/CT imaging. However, low-dose images often suffer from reduced quality depending on acquisition patient parameters. Deep learning (DL)-based organ segmentation models typically trained high-quality images, with limited dedicated noisy images. This study aimed to develop a DL pipeline ultra-low-dose Materials methods 274 raw datasets were reconstructed using Siemens ReconCT software ADMIRE iterative algorithm, generating full-dose (FD-CT) simulated (LD-CT) at 1%, 2%, 5%, 10% of the original tube current. Existing FD-nnU-Net segmented 22 organs FD-CT serving as reference masks training new LD-nnU-Net LD-CT Three bony tissue (6 organs), soft-tissue (15 body contour segmentation. The compared standard reference. External actual also compared. Results performance declined radiation dose, especially below (5 mAs). achieved average Dice scores 0.937 ± 0.049 (bony tissues), 0.905 0.117 (soft-tissues), 0.984 0.023 (body contour). LD outperformed FD external datasets. Conclusion Conventional performed poorly Dedicated demonstrated superior across cross-validation evaluations, enabling accurate available our GitHub page.

Язык: Английский

Процитировано

1
Potential of Radiomics, Dosiomics, and Dose Volume Histograms for Tumor Response Prediction in Hepatocellular Carcinoma following 90Y-SIRT DOI Creative Commons
Zahra Mansouri, Yazdan Salimi, Ghasem Hajianfar

и другие.

Molecular Imaging and Biology, Год журнала: 2025, Номер unknown

Опубликована: Март 10, 2025

Abstract Purpose We evaluate the role of radiomics, dosiomics, and dose-volume constraints (DVCs) in predicting response hepatocellular carcinoma to selective internal radiation therapy with 90 Y glass microspheres. Methods 99m Tc-macroagregated albumin ( Tc-MAA) SPECT/CT images 17 patients were included. Tumor responses at three months evaluated using modified evaluation criteria solid tumors categorized as responders or non-responders. Dosimetry was conducted local deposition method (Dose) biologically effective dosimetry. A total 264 DVCs, 321 radiomic features, dosiomic features extracted from tumor, normal perfused liver (NPL), whole (WNL). Five different feature selection methods combination eight machine learning algorithms employed. Model performance area under AUC, accuracy, sensitivity, specificity. Results No statistically significant differences observed between neither dose metrics nor radiomicas dosiomics non-responder groups. Y-dosiomics models any given set inputs outperformed other models. This also true for Y-radiomics SPECT SPECT-clinical achieving an specificity 1. Among MAA-dosiomic models, two showed AUC ≥ 0.91. While MAA-dose volume histogram (DVH)-based less promising, Y-DVH-based strong (AUC 0.91) when considered independently clinical features. Conclusion study demonstrated potential Tc-MAA SPECT-derived dosimetry establishing predictive tumor response.

Язык: Английский

Процитировано

0
Deep Learning-Based CT-Less Cardiac Segmentation of PET Images: A Robust Methodology for Multi-Tracer Nuclear Cardiovascular Imaging DOI Creative Commons
Yazdan Salimi, Zahra Mansouri, René Nkoulou

и другие.

Deleted Journal, Год журнала: 2025, Номер unknown

Опубликована: Май 6, 2025

Abstract Quantitative cardiovascular PET/CT imaging is useful in the diagnosis of multiple cardiac perfusion and motion pathologies. The common approach for segmentation consists using co-registered CT images, exploiting publicly available deep learning (DL)-based models. However, mismatch between structural images PET uptake limits usefulness these approaches. Besides, performance DL models not consistent over low-dose or ultra-low-dose commonly used clinical imaging. In this work, we developed a DL-based methodology to tackle issue by segmenting directly images. This study included 406 from 146 patients (43 18 F-FDG, 329 13 N-NH 3 , 37 82 Rb images). Using previously trained nnU-Net our group, segmented whole heart three main components, namely left myocardium (LM), ventricle cavity (LV), right (RV) on was resampled resolution edited through combination automated image processing manual correction. corrected masks SUV were fed V2 pipeline be fivefold data split strategy defining two tasks: task #1 #2 components. Fifteen as external validation set. delineated compared with standard reference Dice coefficient, Jaccard distance, mean surface segment volume relative error (%). Task average coefficient internal 0.932 ± 0.033. 15 cases comparable reaching an 0.941 0.018. 0.88 0.063, 0.828 0.091, 0.876 0.062 LM, LV, RV, respectively. There no statistically significant difference among coefficients, neither acquired radiotracers nor different folds ( P -values > 0.05). overall prediction components less than 2%. We acceptable accuracy robust test set nuclear proposed can overcome unreliable segmentations performed

Язык: Английский

Процитировано

0
Impact of tracer uptake rate on quantification accuracy of myocardial blood flow in PET: A simulation study DOI
Xiaotong Hong, Amirhossein Sanaat, Yazdan Salimi

и другие.

Medical Physics, Год журнала: 2025, Номер unknown

Опубликована: Май 8, 2025

Cardiac perfusion PET is commonly used to assess ischemia and cardiovascular risk, which enables quantitative measurements of myocardial blood flow (MBF) through kinetic modeling. However, the estimation parameters challenging due noisy nature short dynamic frames limited sample data points. This work aimed investigate errors in MBF a simulation study evaluate different parameter approaches, including deep learning (DL) method. Simulated studies were generated using digital phantoms based on cardiac segmentations from 55 clinical CT images. We employed irreversible 2-tissue compartmental model simulated 13N-ammonia scans under both rest stress conditions (220 cases each). The simulations covered K1 range 0.6 1.2 3.6 (unit: mL/min/g) myocardium. A transformer-based DL was trained dataset predict parametric images (PIMs) image validated 5-fold cross-validation. compared method with voxel-wise nonlinear least squares (NLS) fitting applied images, either Gaussian filter (GF) smoothing (GF-NLS) or nonlocal means (DNLM) algorithm for denoising (DNLM-NLS). Two patients coronary angiography (CTA) fractional reserve (FFR) enrolled test feasibility applying models data. showed clearer structures reduced noise traditional NLS-based methods. In terms mean absolute relative error (MARE), as values increased mL/min/g, overall bias myocardium estimates decreased approximately 58% 45% methods while reduction MARE 42% 18%. For data, 30% 70% GF-NLS contrast, DNLM-NLS (average: 42%) 20%) demonstrated significantly smaller changes varied. Regarding regional (±standard deviation), had 6.30% (±8.35%) K1, 1.10% (±8.21%) 6.28% (±14.05%) 10.72% (±9.34%) 1.69% (±8.82%) -10.55% (±9.81%) that an increase tracer uptake rate (K1) corresponded improved accuracy precision quantification, whereas lower resulted higher poorer estimates. Utilizing techniques approaches can mitigate noise-induced imaging.

Язык: Английский

Процитировано

0