Search Results (14)

Background: The generation of synthetic CT images from MRI scans represents a crucial step toward enabling MRI-only clinical workflows and supporting multi-modal integration in medical imaging, particularly in radiotherapy planning. Despite significant advancements in deep learning models, many current methods still struggle to reconstruct high-density structures, especially bone, and exhibit limited accuracy in density values. This shortcoming is largely attributed to the passage of excessive or noisy features through skip connections in the traditional U-Net architecture, which degrade the quality of information transmitted to the decoder, negatively impacting the clarity of anatomical boundaries and the pixel-wise accuracy of the resulting synthetic image. Methods: In this work, we propose an enhanced 3D U-Net architecture in which the Convolutional Block Attention Module (CBAM) is systematically integrated within each skip connection. The CBAM sequentially applies channel and spatial attention to adaptively reweight encoder feature maps before fusion with the decoder, thereby emphasizing anatomically relevant structures while suppressing irrelevant feature propagation. The model was trained and evaluated on the SynthRAD2023 (Task 1—Brain) MRI–CT dataset. To rigorously assess the contribution of the attention mechanism, a dedicated ablation study was conducted comparing three variants: 3D U-Net with Squeeze-and-Excitation (SE), Coordinate Attention (CA), and the proposed CBAM module. Performance was evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Normalized Cross-Correlation (NCC). Results: The ablation study demonstrated that the CBAM-enhanced model consistently outperformed both SE- and CA-based variants across all quantitative metrics. Specifically, the proposed method achieved an MAE of $38.2\pm 5.4$ HU and an RMSE of $51.0\pm 12.0$ HU, representing the lowest reconstruction errors among the evaluated models. In addition, it obtained a PSNR of $29.45\pm 2.10$ dB, SSIM of $0.940\pm 0.031$, and NCC of $0.967\pm 0.015$, indicating superior structural preservation and strong voxel-wise correspondence between synthesized and reference CT volumes. These results confirm that the sequential integration of channel and spatial attention provides a statistically and practically meaningful improvement for high-fidelity MRI-to-CT synthesis. Conclusions: Generating high-resolution brain CT images from brain MRI scans using a 3D U-Net network enhanced with a CBAM module can contribute to supporting the clinical workflow by providing additional diagnostic data without the need for extra radiological examinations, thereby enhancing diagnostic efficiency and reducing radiation exposure. This technique helps reduce patient exposure to radiation and improves accessibility in resource-limited settings. Furthermore, this method is valuable for retrospective studies, surgical planning, and image-guided therapy, where complete multi-modal data may not always be available. Full article

Objective: Axillary lymph node (ALN) status in breast cancer is pivotal for guiding treatment and determining prognosis. The study aimed to explore the feasibility and efficacy of a radiomics model using voxel-wise dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) time-intensity-curve (TIC) profile maps to predict ALN metastasis in breast cancer. Methods: A total of 615 breast cancer patients who underwent preoperative DCE-MRI from October 2018 to February 2024 were retrospectively enrolled and randomly allocated into training (n = 430) and testing (n = 185) sets (7:3 ratio). Based on wash-in rate, wash-out enhancement, and wash-out stability, each voxel within manually segmented 3D lesions that were categorized into 1 of 19 TIC subtypes from the DCE-MRI images. Three feature sets were derived: composition ratio (type-19), radiomics features of TIC subtypes (type-19-radiomics), and radiomics features of third-phase DCE-MRI (phase-3-radiomics). Student’s t-test and the least absolute shrinkage and selection operator (LASSO) was used to select features. Four models (type-19, type-19-radiomics, type-19-combined, and phase-3-radiomics) were constructed by a support vector machine (SVM) to predict ALN status. Model performance was assessed using sensitivity, specificity, accuracy, F1 score, and area under the curve (AUC). Results: The type-19-combined model significantly outperformed the phase-3-radiomics model (AUC = 0.779 vs. 0.698, p < 0.001; 0.674 vs. 0.559) and the type-19 model (AUC = 0.779 vs. 0.541, p < 0.001; 0.674 vs. 0.435, p < 0.001) in cross-validation and independent testing sets. The type-19-radiomics showed significantly better performance than the phase-3-radiomics model (AUC = 0.764 vs. 0.698, p = 0.002; 0.657 vs. 0.559, p = 0.037) and type-19 model (AUC = 0. 764 vs. 0.541, p < 0.001; 0.657 vs. 0.435, p < 0.001) in cross-validation and independent testing sets. Among four models, the type-19-combined model achieved the highest AUC (0.779, 0.674) in cross-validation and testing sets. Conclusions: Radiomics analysis of voxel-wise DCE-MRI TIC profile maps, simultaneously quantifying temporal and spatial hemodynamic heterogeneity, provides an effective, noninvasive method for predicting ALN metastasis in breast cancer. Full article

Background: Accurate and rapid intraoperative tumor margin assessment remains a major challenge in surgical oncology. Current gold-standard methods, such as frozen section histology, are time-consuming, operator-dependent, and prone to misclassification, which limits their clinical utility. Objective: To develop and evaluate a novel hyperspectral imaging (HSI) workflow that integrates deep learning with three-dimensional (3D) tumor modeling for real-time, label-free tumor margin delineation in head and neck squamous cell carcinoma (HNSCC). Methods: Freshly resected HNSCC samples were snap-frozen and imaged ex vivo from multiple perspectives using a standardized HSI protocol, resulting in a 3D model derived from HSI. Each sample was serially sectioned, stained, and annotated by pathologists to create high-resolution 3D histological reconstructions. The volumetric histological models were co-registered with the HSI data (n = 712 Datacubes), enabling voxel-wise projection of tumor segmentation maps from the HSI-derived 3D model onto the corresponding histological ground truth. Three deep learning models were trained and validated on these datasets to differentiate tumor from non-tumor regions with high spatial precision. Results: This work demonstrates strong potential for the proposed HSI system, with an overall classification accuracy of 0.98 and a tumor sensitivity of 0.93, underscoring the system’s ability to reliably detect tumor regions and showing high concordance with histopathological findings. Conclusion: The integration of HSI with deep learning and 3D tumor modeling offers a promising approach for precise, real-time intraoperative tumor margin assessment in HNSCC. This novel workflow has the potential to improve surgical precision and patient outcomes by providing rapid, label-free tissue differentiation. Full article

Hybrid positron emission tomography/magnetic resonance imaging (PET/MR) opens new possibilities in multimodal multiparametric (m2p) image analyses. But even the simultaneous acquisition of positron emission tomography (PET) and magnetic resonance imaging (MRI) does not guarantee perfect voxel-by-voxel co-registration due to organs and distortions, especially in diffusion-weighted imaging (DWI), which would be, however, crucial to derive biologically meaningful information. Thus, our aim was to optimize fusion and voxel-wise analyses of DWI and standardized uptake values (SUVs) using a novel software for m2p analyses. Using research software, we evaluated the precision of image co-registration and voxel-wise analyses including the rigid and elastic 3D registration of DWI and [18F]-Fluorodeoxyglucose (FDG)-PET from an integrated PET/MR system. We analyzed DWI distortions with a volume-preserving constraint in three different 3D-printed phantom models. A total of 12 PET/MR-DWI clinical datasets (bronchial carcinoma patients) were referenced to the T1 weighted-DIXON sequence. Back mapping of scatterplots and voxel-wise registration was performed and compared to the non-optimized datasets. Fusion was rated using a 5-point Likert scale. Using the 3D-elastic co-registration algorithm, geometric shapes were restored in phantom measurements; the measured ADC values did not change significantly (F = 1.12, p = 0.34). Reader assessment showed a significant improvement in fusion precision for DWI and morphological landmarks in the 3D-registered datasets (4.3 ± 0.2 vs. 4.6 ± 0.2, p = 0.009). Most pronounced differences were noted for the chest wall (p = 0.006), tumor (p = 0.007), and skin contour (p = 0.014). Co-registration increased the number of plausible ADC and SUV combinations by 25%. The volume-preserving elastic 3D registration of DWI significantly improved the precision of fusion with anatomical sequences in phantom and clinical datasets. The research software allowed for a voxel-wise analysis and visualization of [18F]FDG-PET/MR data as a “combined diffusivity–metabolic index” (cDMI). The clinical value of the optimized PET/MR biomarker can thus be tested in future PET/MR studies. Full article

The clinical significance of measuring vessel wall thickness is widely acknowledged. Recent advancements have enabled high-resolution 3D scans of arteries and precise segmentation of their lumens and outer walls; however, most existing methods for assessing vessel wall thickness are 2D. Despite being valuable, reproducibility and accuracy of 2D techniques depend on the extracted 2D slices. Additionally, these methods fail to fully account for variations in wall thickness in all dimensions. Furthermore, most existing approaches are difficult to be extended into 3D and their measurements lack spatial localization and are primarily confined to lumen boundaries. We advocate for a shift in perspective towards recognizing vessel wall thickness measurement as inherently a 3D challenge and propose adapting the Laplacian method as an outstanding alternative. The Laplacian method is implemented using convolutions, ensuring its efficient and rapid execution on deep learning platforms. Experiments using digital phantoms and vessel wall imaging data are conducted to showcase the accuracy, reproducibility, and localization capabilities of the proposed approach. The proposed method produce consistent outcomes that remain independent of centerlines and 2D slices. Notably, this approach is applicable in both 2D and 3D scenarios. It allows for voxel-wise quantification of wall thickness, enabling precise identification of wall volumes exhibiting abnormal wall thickness. Our research highlights the urgency of transitioning to 3D methodologies for vessel wall thickness measurement. Such a transition not only acknowledges the intricate spatial variations of vessel walls, but also opens doors to more accurate, localized, and insightful diagnostic insights. Full article

Background: The predictive value of thrombus perviousness in acute ischemic stroke (AIS), as measured by computed tomography (CT), has been intensively studied with conflicting results. In this study, we investigate the predictive potential of the novel concept of dynamic perviousness using three-dimensional (3D) volumetric evaluation of occlusive thrombi. Methods: The full thrombus volume in 65 patients with a hyperdense artery sign on non-contrast CT (NCCT), who underwent mechanical thrombectomy (MT), was segmented. Perviousness maps were computed voxel-wise for the entire thrombus volume as thrombus attenuation increase (TAI) between NCCT and CT angiography (CTA) as well as between CTA and late venous phase CT (CTV). Perviousness was analyzed for its association with NIHSS at admission, Thrombolysis In Cerebral Infarction (TICI) score, and number of MT passes. Results: The mean late-uptake TAI of thrombi with NIHSS scores greater than 21 at admission was approximately 100% higher than for lower scored NIHSS (p between 0.05 and 0.005). Concerning revascularization results, thrombi requiring less than four MT passes had ca. 80% higher group mean late-uptake TAI than clots requiring four or more passes (p = 0.03), and thrombi with TICI score III had ca. 95% higher group mean late-uptake TAI than thrombi with TICI II (p = 0.03). Standard perviousness showed no significant correlation with MT results. Conclusion: Standard thrombus perviousness of 3D clot volume is not associated with revascularization results in AIS. In contrast, dynamic perviousness assessed with a voxel-wise characterization of 3D thrombi volume may be a better predictor of MT outcomes than standard perviousness. Full article

(1) Background: Minimal hepatic encephalopathy (MHE) is an important complication of decompensated cirrhosis. Previous studies have demonstrated spontaneous brain activity alterations in cirrhotic patients with MHE. However, the reported results are inconsistent, which has limited our understanding of the potential neural mechanisms. Thus, we conducted a quantitative meta-analysis of resting-state functional imaging studies to identify the regional activity alterations consistently involved in MHE. (2) Methods: We searched six databases to include resting-state functional imaging studies and compared spontaneous brain activity patterns between MHE patients and healthy controls (HCs), and between cirrhotic patients without minimal hepatic encephalopathy (NMHE) and HCs. Then, a separate whole-brain voxel-wise meta-analysis between MHE or NMHE patients and HCs was conducted using seed-based d mapping with permutation of subject images. We further conducted the conjunction analysis to assess the distinct regional activity alterations between MHE and NMHE patients as compared to HCs. (3) Results: Thirteen studies with twenty datasets were included in this meta-analysis. Compared with HCs, MHE patients showed decreased spontaneous brain activity in the left superior frontal gyrus, left median cingulate/paracingulate gyri, and right precuneus. Compared with NMHE patients, MHE patients indicated decreased spontaneous brain activity in the left superior frontal gyrus, left median cingulate/paracingulate gyri, and right precuneus. (4) Conclusions: MHE is associated with spontaneous brain activity alterations involving the left superior frontal gyrus and median cingulate/paracingulate gyri, which may implicate primarily in spatial working memory and emotional disorders. These findings may contribute to a better understanding of the potential neural mechanisms, and guide further research. Full article

The aim of this study is to investigate the applicability of the dual-venc (DV) 4D flow magnetic resonance imaging (MRI) to quantify the complex flow patterns in type B aortic dissection (TBAD). One GRAPPA-accelerated single-venc (SV) and one compressed-sensing (CS) accelerated DV 4D flow MRI sequences are used to scan all subjects, including twelve chronic TBAD patients and two volunteers. The scans are performed twice for the reproducibility assessment of the scan protocols. Voxelwise quantitative flow parameters including kinetic energy (KE), peak velocity (PV), forward and reverse flows (FF, RF) and stasis are calculated. High-venc (HV) data from the DV acquisition are separately analyzed. The scan time reduction by the CS-accelerated DV 4D flow MRI acquisition is 46.4% compared with the SV acquisition. The DV velocity-to-noise ratio (VNR) is higher compared with HV (p = 0.000). No true lumen (TL) parameter shows a significant difference among the acquisition types (p > 0.05). The false lumen (FL) RF is higher in SV compared with the DV acquisition (p = 0.009). The KE is higher (p = 0.038) and stasis is lower (p = 0.01) in HV compared with SV acquisition. All FL parameters except stasis are higher and stasis is lower in HV compared with DV acquisition (p < 0.05). Positive Pearson correlations among the acquisition types in TL and high agreements between the two scans for all acquisition types are observed except HV RF in the FL, which demonstrates a moderate agreement. The CS-accelerated DV 4D flow MRI may have utility in the clinical daily routine with shortened scan times and improved velocity measurements while providing high VNR in TBAD. The observed hemodynamic flow trends are similar between GRAPPA-accelerated SV and CS-accelerated DV 4D flow MRI acquisitions; however, parameters are more impacted by CS-accelerated HV protocol in FL, which may be secondary to the CS regularization effects. Full article

Voxel-wise quantitative assessment of typical characteristics in three-dimensional (3D) multiphase computed tomography (CT) imaging, especially arterial phase hyperenhancement (APHE) and subsequent washout (WO), is crucial for the diagnosis and therapy of hepatocellular carcinoma (HCC). However, this process is still missing in practice. Radiologists often visually estimate these features, which limit the diagnostic accuracy due to subjective interpretation and qualitative assessment. Quantitative assessment is one of the solutions to this problem. However, performing voxel-wise assessment in 3D is difficult due to the misalignments between images caused by respiratory and other physiological motions. In this paper, based on the Liver Imaging Reporting and Data System (v2018), we propose a registration-based quantitative model for the 3D voxel-wise assessment of image characteristics through multiple CT imaging phases. Specifically, we selected three phases from sequential CT imaging phases, i.e., pre-contrast phase (Pre), arterial phase (AP), delayed phase (DP), and then registered Pre and DP images to the AP image to extract and assess the major imaging characteristics. An iterative reweighted local cross-correlation was applied in the proposed registration model to construct the fidelity term for comparison of intensity features across different imaging phases, which is challenging due to their distinct intensity appearance. Experiments on clinical dataset showed that the means of dice similarity coefficient of liver were $98.6\%$ and $98.1\%$, those of surface distance were $0.38$ and $0.54$ mm, and those of Hausdorff distance were $4.34$ and $6.16$ mm, indicating that quantitative estimation can be accomplished with high accuracy. For the classification of APHE, the result obtained by our method was consistent with those acquired by experts. For the WO, the effectiveness of the model was verified in terms of WO volume ratio. Full article

Diffusion tensor imaging (DTI) allows the in vivo imaging of pathological white matter alterations, either with unbiased voxel-wise or hypothesis-guided tract-based analysis. Alterations of diffusion metrics are indicative of the cerebral status of patients with amyotrophic lateral sclerosis (ALS) at the individual level. Using machine learning (ML) models to analyze complex and high-dimensional neuroimaging data sets, new opportunities for DTI-based biomarkers in ALS arise. This review aims to summarize how different ML models based on DTI parameters can be used for supervised diagnostic classifications and to provide individualized patient stratification with unsupervised approaches in ALS. To capture the whole spectrum of neuropathological signatures, DTI might be combined with additional modalities, such as structural T1w 3-D MRI in ML models. To further improve the power of ML in ALS and enable the application of deep learning models, standardized DTI protocols and multi-center collaborations are needed to validate multimodal DTI biomarkers. The application of ML models to multiparametric MRI/multimodal DTI-based data sets will enable a detailed assessment of neuropathological signatures in patients with ALS and the development of novel neuroimaging biomarkers that could be used in the clinical workup. Full article

Traumatic brain injury (TBI) is currently a problematic issue of public health due to its frequency, and many of the mild cases often remain undiagnosed despite the possible predisposition to prolonged or persistent post-concussive symptomatology. It was shown here that the severity and persistence of grey matter (GM) changes following TBI could predict disease outcomes. Our aim was to conduct a voxel-wise meta-analysis to detect significant GM changes following mild TBI (mTBI) and to investigate whether these changes are associated with the duration and severity of post-concussion syndrome (PCS). A voxel-wise meta-analysis was conducted regarding the GM and white matter (WM) changes in mTBI adult patients versus healthy controls, and Seed-based d Mapping was used to correlate the data. Standard meta-analysis statistical processing was used to assess heterogeneity and publication bias. Our analysis showed significant GM volume increases in the left medial cingulate/paracingulate gyri, the middle frontal gyrus, and the right caudate nucleus of the mTBI patients and significant volume loss in the thalamus, the frontal lobe, and the temporal lobe. These changes could potentially be associated with PCS that some mTBI later patients develop as a result to the injury or other compensatory changes. Additional studies considering long-term GM changes in mTBI patients and their potential relationship to PCS could provide further insight into the pathophysiological similarities and correlations between mTBI and PCS. Full article

The present study aimed to investigate the association between the functional connectivity (FC) of the olfactory cortex and olfactory performance in Parkinson’s disease (PD). Eighty-two early PD patients and twenty-one healthy controls underwent structural and resting-state functional MRI scans, as well as neuropsychological assessments from the Parkinson’s Progression Markers Initiative database. A whole brain voxel-wise regression analysis was conducted to evaluate the relationship between the FC of the entorhinal cortex (EC-FC) and olfactory performance. Then, a one-way ANCOVA, based on the regions of interest, was performed with SPSS to investigate the group differences and correlation analysis that were used to analyze the relationships between the FC and neuropsychological assessments. In addition, regression models were used to evaluate the risk factors for the decreased olfactory function. A significantly negative correlation was observed between the olfactory performance and the left EC-FC in the right dorsal cingulate gyrus (dCC) in patients. The PD patients with anosmia exhibited significantly higher FC values than the PD patients with normal olfaction or the PD patients with mild to moderate microsomia. Except for the olfactory performance, no significant correlation was detected between the neuropsychological assessments and the FC values. A linear regression analysis revealed that the increased FC and Geriatric Depression Scale are independently associated with lower the University of Pennsylvania Smell Identification Test scores. The current findings enhanced the understanding of olfactory dysfunction-related pathophysiological mechanisms in early PD and suggested that the left EC-FC in the right dCC may be a potential neuroimaging biomarker for olfactory performance. Full article

Radiation-induced lung damage (RILD) is a common side effect of radiotherapy (RT). The ability to automatically segment, classify, and quantify different types of lung parenchymal change is essential to uncover underlying patterns of RILD and their evolution over time. A RILD dedicated tissue classification system was developed to describe lung parenchymal tissue changes on a voxel-wise level. The classification system was automated for segmentation of five lung tissue classes on computed tomography (CT) scans that described incrementally increasing tissue density, ranging from normal lung (Class 1) to consolidation (Class 5). For ground truth data generation, we employed a two-stage data annotation approach, akin to active learning. Manual segmentation was used to train a stage one auto-segmentation method. These results were manually refined and used to train the stage two auto-segmentation algorithm. The stage two auto-segmentation algorithm was an ensemble of six 2D Unets using different loss functions and numbers of input channels. The development dataset used in this study consisted of 40 cases, each with a pre-radiotherapy, 3-, 6-, 12-, and 24-month follow-up CT scans (n = 200 CT scans). The method was assessed on a hold-out test dataset of 6 cases (n = 30 CT scans). The global Dice score coefficients (DSC) achieved for each tissue class were: Class (1) 99% and 98%, Class (2) 71% and 44%, Class (3) 56% and 26%, Class (4) 79% and 47%, and Class (5) 96% and 92%, for development and test subsets, respectively. The lowest values for the test subsets were caused by imaging artefacts or reflected subgroups that occurred infrequently and with smaller overall parenchymal volumes. We performed qualitative evaluation on the test dataset presenting manual and auto-segmentation to a blinded independent radiologist to rate them as ‘acceptable’, ‘minor disagreement’ or ‘major disagreement’. The auto-segmentation ratings were similar to the manual segmentation, both having approximately 90% of cases rated as acceptable. The proposed framework for auto-segmentation of different lung tissue classes produces acceptable results in the majority of cases and has the potential to facilitate future large studies of RILD. Full article

Corticobasal syndrome (CBS) is a rare neurodegenerative condition characterized by four-repeat tau aggregation in the cortical and subcortical brain regions and accompanied by severe atrophy. The aim of this study was to evaluate partial volume effect correction (PVEC) in patients with CBS compared to a control cohort imaged with the 18-kDa translocator protein (TSPO) positron emission tomography (PET) tracer [¹⁸F]GE-180. Eighteen patients with CBS and 12 age- and sex-matched healthy controls underwent [¹⁸F]GE-180 PET. The cortical and subcortical regions were delineated by deep nuclei parcellation (DNP) of a 3D-T1 MRI. Region-specific subcortical volumes and standardized uptake values and ratios (SUV and SUVr) were extracted before and after region-based voxel-wise PVEC. Regional volumes were compared between patients with CBS and controls. The % group differences and effect sizes (CBS vs. controls) of uncorrected and PVE-corrected SUVr data were compared. Single-region positivity in patients with CBS was assessed by a >2 SD threshold vs. controls and compared between uncorrected and PVE-corrected data. Smaller regional volumes were detected in patients with CBS compared to controls in the right ventral striatum (p = 0.041), the left putamen (p = 0.005), the right putamen (p = 0.038) and the left pallidum (p = 0.015). After applying PVEC, the % group differences were distinctly higher, but the effect sizes of TSPO uptake were only slightly stronger due to the higher variance after PVEC. The single-region positivity of TSPO PET increased in patients with CBS after PVEC (100 vs. 83 regions). PVEC in the cortical and subcortical regions is valuable for TSPO imaging of patients with CBS, leading to the improved detection of elevated [¹⁸F]GE-180 uptake, although the effect sizes in the comparison against the controls did not improve strongly. Full article