T1 Images Research Articles

Assessment of Y-90 Radioembolization Treatment Response for Hepatocellular Carcinoma Cases Using MRI Radiomics.

This study aimed to investigate the ability of radiomics features extracted from magnetic resonance imaging (MRI) images to differentiate between responders and non-responders for hepatocellular carcinoma (HCC) cases who received Y-90 transarterial radioembolization treatment. Thirty-six cases of HCC who underwent MRI scans after Y-90 radioembolization were included in this study. Tumors were segmented from MRI T2 images, and then 87 radiomic features were extracted through the LIFEx package software. Treatment response was determined 9 months after treatment through the modified response evaluation criteria in solid tumours (mRECIST). According to mRECIST, 28 cases were responders and 8 cases were non-responders. Two radiomics features, "Grey Level Size Zone Matrix (GLSZM)-Small Zone Emphasis" and "GLSZM-Normalized Zone Size Non-Uniformity", were the radiomics features that could predict treatment response with the area under curve (AUC)= 0.71, sensitivity= 0.93, and specificity= 0.62 for both features. Whereas the other 4 features (kurtosis, intensity histogram root mean square, neighbourhood gray-tone difference matrix strength, and GLSZM normalized grey level non-uniformity) have a relatively lower but acceptable discrimination ability range from AUC= 0.6 to 0.66. MRI radiomics analysis could be used to assess the treatment response for HCC cases treated with Y-90 radioembolization.

Use of multiparametric MRI to noninvasively assess iodinated contrast-induced acute kidney injury

PurposeTo gauge the utility of multiparametric MRI in characterizing pathologic changes after iodinated contrast-induced acute kidney injury (CI-AKI) in rats. MethodsWe randomly grouped 24 rats injected with 8 g iodine/kg of body weight (n = 6 each) and 6 rats injected with saline as controls. All rats underwent T1, T2 mapping and diffusion kurtosis imaging (DKI) after contrast injection at 0 (control), 1, 3, 7, 13 days. T1, T2, and mean kurtosis (MK) values were performed in renal outer/inner stripes of outer medulla (OSOM and ISOM) and cortex (CO), and their diagnosis performance for CI-AKI also been evaluated. Serum creatinine (SCr), insulin-like growth factor-binding protein 7 (IGFBP7), tissue inhibitor metalloproteinase 2 (TIMP-2), aquaporin-1 (AQP1), α-smooth muscle actin (α-SMA), and histologic indices were examined. ResultsCompared with controls, urinary concentrations of both TIMP-2 and IGFBP7 were obviously elevated from Day 1 to Day 13 (all p < 0.05). T2 values were significantly higher than control group for Days 1 and 3, and T1 and MK increased were more remarkable at all time points (Days 1–13) in CI-AKI (all p < 0.05) than control group. Changes in T1 and MK strongly correlated with renal injury scores of all anatomical compartments and with expression levels of AQP1 and moderately correlated with α-SMA. Changes in T2 values correlating moderately with renal scores of CO, ISOM and OSOM and AQP1. The MK obtained the highest area under the receiver operating characteristic (ROC) curve of 0.846 with a sensitivity of 70.8 % and specificity of 88.9 %. ConclusionsCombined use of multiparametric MRI could be a valid noninvasive method for comprehensive monitoring of CI-AKI. Among these parameters, MK may achieve the best diagnostic performance for CI-AKI.

Machine learning localization to identify the epileptogenic side in mesial temporal lobe epilepsy

BackgroundMesial temporal sclerosis (MTS) is the most common pathology associated with drug-resistant mesial temporal lobe epilepsy (mTLE) in adults.Most atrophic hippocampi can be identified using MRI based on standard epilepsy protocols; however, difficulties can arise in cases where sclerotic changes in the hippocampus are subtle or non-epilepsy-specific protocols have been implemented. In such cases, quantitative methods, such as T1-weighted axial series MRIs, are valuable additional tools to complement epilepsy-specific protocols. In the current study, we applied machine learning (ML) techniques to the analysis of brain regions of interest (ROIs), including the hippocampus, thalamus, and cortical areas, to enhance the accuracy of lesion lateralization in MRI. MethodsThis study included 104 patients diagnosed with mTLE, including 55 with lesions on the right side and 49 with lesions on the left side. FreeSurfer software was used to extract features from high-resolution T1-weighted axial brain MRI scans for use in computing lateralization indices (LI) for various brain regions. After using feature selection to pinpoint critical ROIs, the corresponding LI values were used as parameters in training the ML model. ResultsThe proposed ML model demonstrated exceptional performance in the lateralization of mTLE, achieving test accuracy of 92.38 % with an AUROC of 0.97. ConclusionThis study demonstrated the efficacy of ML in detecting instances of MTS from thin-slice T1 images. The proposed method provides valuable insights for surgical planning and treatment. Nonetheless, additional research will be required to enhance the robustness of the model and rigorously validate its effectiveness and applicability in clinical settings.

Specialized gray matter segmentation via a generative adversarial network: application on brain white matter hyperintensities classification.

White matter hyperintensities (WMH) observed in T2 fluid-attenuated inversion recovery (FLAIR) images have emerged as potential markers of neurodegenerative diseases like Multiple Sclerosis (MS). Lacking comprehensive automated WMH classification systems in current research, there is a need to develop accurate detection and classification methods for WMH that will benefit the diagnosis and monitoring of brain diseases. Juxtacortical WMH (JCWMH) is a less explored subtype of WMH, primarily due to the hard definition of the cortex in FLAIR images, which is escalated by the presence of lesions to obtain appropriate gray matter (GM) masks. In this study, we present a method to perform a specialized GM segmentation developed for the classification of WMH, especially JCWMH. Using T1 and FLAIR images, we propose a pipeline to integrate masks of white matter, cerebrospinal fluid, ventricles, and WMH to create a unique mask to refine the primary GM map. Subsequently, we utilize this pipeline to generate paired data for training a conditional generative adversarial network (cGAN) to substitute the pipeline and reduce the inputs to only FLAIR images. The classification of WMH is then based on the distances between WMH and ventricular and GM masks. Due to the lack of multi-class labeled WMH datasets and the need for extensive data for training deep learning models, we attempted to collect a large local dataset and manually segment and label some data for WMH and ventricles. In JCWMH classification, the proposed method exhibited a Dice similarity coefficient, precision, and sensitivity of 0.76, 0.69, and 0.84, respectively. With values of 0.66, 0.55, and 0.81, the proposed method clearly outperformed the approach commonly used in the literature, which uses extracted GM masks from registered T1 images on FLAIR. After training, the method proves its efficiency by providing results in less than one second. In contrast, the usual approach would require at least two minutes for registration and segmentation alone. The proposed method is automated and fast and requires no initialization as it works exclusively with FLAIR images. Such innovative methods will undoubtedly facilitate accurate and meaningful analysis of WMH in clinical practice by reducing complexity and increasing efficiency.

Deep-Dixon: Deep-Learning frameworks for fusion of MR T1 images for fat and water extraction

Deep-Dixon: Deep-Learning frameworks for fusion of MR T1 images for fat and water extraction

Non-traumatic brachial plexopathy identification from routine MRIs: Retrospective studies with deep learning networks

PurposeThis study aims to seek an optimized deep learning model for differentiating non-traumatic brachial plexopathy from routine MRI scans. Materials and methodsThis retrospective study collected patients through the electronic medical records (EMR) or pathological reports at Mayo Clinic and underwent BP MRI from January 2002 to December 2022. Using sagittal T1, fluid-sensitive and post-gadolinium images, a radiology panel selected BP’s region of interest (ROI) to form 3 dimensional volumes for this study. We designed six deep learning schemes to conduct BP abnormality differentiation across three MRI sequences. Utilizing five prestigious deep learning networks as the backbone, we trained and validated these models by nested five-fold cross-validation schemes. Furthermore, we defined a ’method score’ derived from the radar charts as a quantitative indicator as the guidance of the preference of the best model. ResultsThis study selected 196 patients from initial 267 candidates. A total of 256 BP MRI series were compiled from them, comprising 123 normal and 133 abnormal series. The abnormal series included 4 sub-categories, et al. breast cancer (22.5 %), lymphoma (27.1 %), inflammatory conditions (33.1 %) and others (17.2 %). The best-performing model was produced by feature merging mode with triple MRI joint strategy (AUC, 92.2 %; accuracy, 89.5 %) exceeding the multiple channel merging mode (AUC, 89.6 %; accuracy, 89.0 %), solo channel volume mode (AUC, 89.2 %; accuracy, 86.7 %) and the remaining. Evaluated by method score (maximum 2.37), the feature merging mode with backbone of VGG16 yielded the highest score of 1.75 under the triple MRI joint strategy. ConclusionDeployment of deep learning models across sagittal T1, fluid-sensitive and post-gadolinium MRI sequences demonstrated great potential for brachial plexopathy diagnosis. Our findings indicate that utilizing feature merging mode and multiple MRI joint strategy may offer satisfied deep learning model for BP abnormalities than solo-sequence analysis.

Evaluation of the impact of ultra-trail running on knee cartilage using magnetic resonance imaging t2 mapping.

Ultra-marathon trails involve a combination of specific physiological and mechanical constraints and raise new questions regarding the osteoarticular impact on the knees and the long-term risk of osteoarthritis. Magnetic resonance imaging (MRI) T2 relaxation time measurement has shown the ability to determine cartilage response to loading. Higher T2 measurements correspond with cartilage damage. The aim of this study was to quantify the changes in MRI T2 relaxation times of knee articular cartilage after an ultra-trail run and determine knee's consequences of regular practice. Twenty participants in a 55-km race involving total elevation changes of 2600m had 1.5-T knee MRI prior to the race (V0), immediately after (V1) and one month after the race (V2) for T2 relaxation times measurement and morphological sequences (T1, T2 & T2 Fast-Spin Echo (FSE)). T2 measurements were significantly increased in V1 from V0 and remained so one month after the race (V2), despite a significant reduction from V1. Morphological sequences revealed that 65% of the participant had cartilage damage and 65% meniscal damage, 100% of which affected the posterior horn of the medial meniscus. Only one subject (5%) presented no anomaly whatsoever. Damage appeared to be stable between the assessments. Ultra-trail running leads to modifications in the knee cartilage ultrastructure, which persists for at least one month after the event. Furthermore, regular ultra-trail runners present a high number of low-grade cartilage and meniscus lesions.

Single-point macromolecular proton fraction mapping using a 0.3T permanent magnet MRI system: phantom and healthy volunteer study.

In a 0.3T permanent-magnet magnetic resonance imaging (MRI) system, quantifying myelin content is challenging owing to long imaging times and low signal-to-noise ratio. macromolecular proton fraction (MPF) offers a quantitative assessment of myelin in the nervous system. We aimed to demonstrate the practical feasibility of MPF mapping in the brain using a 0.3T MRI. Both 0.3T and 3.0T MRI systems were used. The MPF-mapping protocol used a standard 3D fast spoiled gradient-echo sequence based on the single-point reference method. Proton density, T1, and magnetization transfer-weighted images were obtained from a protein phantom at 0.3T and 3.0T to calculate MPF maps. MPF was measured in all phantom sections to assess its relationship to protein concentration. We acquired MPF maps for 16 and 8 healthy individuals at 0.3T and 3.0T, respectively, measuring MPF in nine brain tissues. Differences in MPF between 0.3T and 3.0T, and between 0.3T and previously reported MPF at 0.5T, were investigated. Pearson's correlation coefficient between protein concentration and MPF at 0.3T and 3.0T was 0.92 and 0.90, respectively. The 0.3T MPF of brain tissue strongly correlated with 3.0T MPF and literature values measured at 0.5T. The absolute mean differences in MPF between 0.3T and 0.5T were 0.42% and 1.70% in white and gray matter, respectively. Single-point MPF mapping using 0.3T permanent-magnet MRI can effectively assess myelin content in neural tissue.

Prediction of early recurrence of adult-type diffuse gliomas following radiotherapy using multi-modal magnetic resonance images.

Adult-type diffuse gliomas are among the central nervous system's most aggressive malignant primary neoplasms. Despite advancements in systemic therapies and technological improvements in radiation oncology treatment delivery, the survival outcome for these patients remains poor. Fast and accurate assessment of tumor response to oncologic treatments is crucial, as it can enable the early detection of recurrent or refractory gliomas, thereby allowing timely intervention with life-prolonging salvage therapies. Radiomics is a developing field with great potential to improve medical image interpretation. This study aims to apply a radiomics-based predictive model for classifying response to radiotherapy within the first 3 months post-treatment. Ninety-five patients were selected from the Burdenko Glioblastoma Progression Dataset. Tumor regions were delineated in the axial plane on contrast-enhanced T1(CE T1W) and T2 fluid-attenuated inversion recovery (T2_FLAIR) magnetic resonance imaging (MRI). Hand-crafted radiomic (HCR) features, including first- and second-order features, were extracted using PyRadiomics (3.7.6) in Python (3.10). Then, recursive feature elimination with a random forest (RF) classifier was applied for feature dimensionality reduction. RF and support vector machine (SVM) classifiers were built to predict treatment outcomes using the selected features. Leave-one-out cross-validation was employed to tune hyperparameters and evaluate the models. For each segmented target, 186 HCR features were extracted from the MRI sequence. Using the top-ranked radiomic features from a combination of CE T1W and T2_FLAIR, an optimized classifier achieved the highest averaged area under the curve (AUC) of 0.829±0.075 using the RF classifier. The HCR features of CE T1W produced the worst outcomes among all models (0.603±0.024 and 0.615±0.075 for RF and SVM classifiers, respectively). We developed and evaluated a radiomics-based predictive model for early tumor response to radiotherapy, demonstrating excellent performance supported by high AUC values. This model, harnessing radiomic features from multi-modal MRI, showed superior predictive performance compared to single-modal MRI approaches. These results underscore the potential of radiomics in clinical decision support for this disease process.

Verification of image quality improvement by deep learning reconstruction to 1.5T MRI in T2-weighted images of the prostate gland.

This study aimed to evaluate whether the image quality of 1.5T magnetic resonance imaging (MRI) of the prostate is equal to or higher than that of 3T MRI by applying deep learning reconstruction (DLR). To objectively analyze the images from the 13 healthy volunteers, we measured the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the images obtained by the 1.5T scanner with and without DLR, as well as for images obtained by the 3T scanner. In the subjective, T2W images of the prostate were visually evaluated by two board-certified radiologists. The SNRs and CNRs in 1.5T images with DLR were higher than that in 3T images. Subjective image scores were better for 1.5T images with DLR than 3T images. The use of the DLR technique in 1.5T MRI substantially improved the SNR and image quality of T2W images of the prostate gland, as compared to 3T MRI.

Improved Dementia Prediction in Cerebral Small Vessel Disease Using Deep Learning-Derived Diffusion Scalar Maps From T1.

Cerebral small vessel disease is the most common pathology underlying vascular dementia. In small vessel disease, diffusion tensor imaging is more sensitive to white matter damage and better predicts dementia risk than conventional magnetic resonance imaging sequences, such as T1 and fluid attenuation inversion recovery, but diffusion tensor imaging takes longer to acquire and is not routinely available in clinical practice. As diffusion tensor imaging-derived scalar maps-fractional anisotropy (FA) and mean diffusivity (MD)-are frequently used in clinical settings, one solution is to synthesize FA/MD from T1 images. We developed a deep learning model to synthesize FA/MD from T1. The training data set consisted of 4998 participants with the highest white matter hyperintensity volumes in the UK Biobank. Four external validations data sets with small vessel disease were included: SCANS (St George's Cognition and Neuroimaging in Stroke; n=120), RUN DMC (Radboud University Nijmegen Diffusion Tensor and Magnetic Resonance Imaging Cohort; n=502), PRESERVE (Blood Pressure in Established Cerebral Small Vessel Disease; n=105), and NETWORKS (n=26), along with 1000 normal controls from the UK Biobank. The synthetic maps resembled ground-truth maps (structural similarity index >0.89 for MD maps and >0.80 for FA maps across all external validation data sets except for SCANS). The prediction accuracy of dementia using whole-brain median MD from the synthetic maps is comparable to the ground truth (SCANS ground-truth c-index, 0.822 and synthetic, 0.821; RUN DMC ground truth, 0.816 and synthetic, 0.812) and better than white matter hyperintensity volume (SCANS, 0.534; RUN DMC, 0.710). We have developed a fast and generalizable method to synthesize FA/MD maps from T1 to improve the prediction accuracy of dementia in small vessel disease when diffusion tensor imaging data have not been acquired.

Importance of gap evaluation in the ossification of posterior longitudinal ligament lesions using 3-dimensional computed tomography

BACKGROUND CONTEXTEvaluating the gaps within the ossification of the posterior longitudinal ligament (OPLL) lesions, which may contribute to neurological symptoms, using conventional imaging techniques is challenging. OBJECTIVEThis study aimed to investigate the importance of evaluating gaps using 3-dimensional computed tomography (3D-CT) and their association with the occurrence of magnetic resonance imaging (MRI) T2 high intensity in the spinal cord. STUDY DESIGN/SETTINGRetrospective cohort study. PATIENT SAMPLERetrospective analysis of 116 patients diagnosed with cervical OPLL. OUTCOME MEASURESPresence of gaps in OPLL, presence of T2 high intensity in the cervical spinal cord, and OPLL thickness were evaluated. METHODSLateral X-ray, CT, and reconstructed 3D-CT images were reviewed to assess lesion characteristics and the presence of gaps. MRI was used to evaluate the change in spinal cord signal intensity. The relationship among gap presence, lesion morphology, and MRI T2 high intensity in the spinal cord was examined. RESULTSA significant difference in gap detection accuracy was observed between CT and 3D-CT (p=.0054). CT demonstrated false-positive results in the detection of gaps as compared with 3D-CT. The presence of gaps was significantly associated with an increased likelihood of MRI T2 high intensity in the spinal cord (p=.037). Patients with thicker lesions and smaller space available for the spinal cord (SAC) were more likely to exhibit T2 high intensity. Meanwhile, patients with gaps co-occurring with T2 high intensity exhibited significantly thinner lesions (p=.011) and larger SACs (p=.0002). Patients with gaps had a significantly lower JOA scores (p=.0035), which indicates that patient with gaps are likely to exhibit more severe clinical neurological symptoms. CONCLUSION3D-CT showed superiority in accurately identifying gaps within OPLL lesions, while CT demonstrated false-positive results in the detection of gaps. Furthermore, the gap presence was a risk factor for MRI T2 high intensity in the spinal cord, independent of lesion thickness. In addition, gaps are related to more severe clinical symptoms. This study highlighted the importance of evaluating gaps within OPLL lesions using 3D-CT to clarify neurological pathogenesis.

Evaluation of Myocarditis with a Free-breathing 3D Isotropic Whole-Heart Joint T1 and T2 Mapping Sequence

BackgroundThe diagnosis of myocarditis by CMR requires the use of T2 and T1 weighted imaging, ideally incorporating parametric mapping. Current 2D mapping sequences are acquired sequentially and involve multiple breath-holds resulting in prolonged scan times and anisotropic image resolution. We developed an isotropic free-breathing 3D whole-heart sequence which allows simultaneous T1 and T2 mapping and validated it in patients with suspected acute myocarditis. MethodsEighteen healthy volunteers and 28 patients with suspected myocarditis underwent conventional 2D T1 and T2 mapping with whole heart coverage and 3D joint T1/T2 mapping on a 1.5T scanner. Acquisition time, image quality, and diagnostic performance were compared. Qualitative analysis was performed using a 4-point Likert scale. Bland-Altman plots were used to assess the quantitative agreement between 2D and 3D sequences. ResultsThe 3D T1/T2 sequence was acquired in 8mins 26s under free breathing, whereas 2D T1 and T2 sequences were acquired with breath holds in 11mins 44s (p=0.0001). All 2D images were diagnostic. For 3D images, 89% of T1 and 96% of T2 images were diagnostic with no significant difference in the proportion of diagnostic images for the 3D and 2D T1 (p=0.2482) and T2 maps (p=1.0000). Systematic bias in T1 was noted with biases of 102ms, 115ms, and 152ms for basal-apical segments, with a larger bias for higher T1 values. Good agreement between T2 values for 3D and 2D techniques was found (bias of 1.8ms, 3.9ms, and 3.6ms for basal-apical segments). The sensitivity and specificity of the 3D sequence for diagnosing acute myocarditis was 74% (95% confidence interval [CI] 49-91%) and 83% (36-100%) respectively, with an estimated c-statistic (95% CI) of 0.85 (0.79-0.91) and no statistically significant difference between the 2D and 3D sequences for the detection of acute myocarditis for T1 (p=0.2207) or T2 (p=1.0000). ConclusionFree-breathing whole heart 3D joint T1/T2 mapping was comparable to 2D mapping sequences with respect to diagnostic performance, but with the added advantages of free-breathing, and shorter scan times. Further work is required to address the bias noted at high T1 values, but this did not significantly impact on diagnostic accuracy.

Development and validation of a prediction model for malignant sinonasal tumors based on MR radiomics and machine learning.

This study aimed to utilize MR radiomics-based machine learning classifiers on a large-sample, multicenter dataset to develop an optimal model for predicting malignant sinonasal tumors and tumor-like lesions. This study included 1711 adult patients (875 benign and 836 malignant) with sinonasal tumors or tumor-like lesions from three institutions. Patients from institution 1 (n = 1367) constituted both the training and validation cohorts, while those from institution 2 and 3 (n = 158/186) made up the test cohorts. Manual segmentation of the region of interest of the tumor was performed on T1WI, T2WI, and contrast-enhanced T1WI (CE-T1WI). Data normalization, dimensional reductions, feature selection, and classifications were performed using ten machine-learning classifiers. Four fusion models, namely T1WI + T2WI, T1WI + CE-T1WI, T2WI + CE-T1WI, and T1WI + T2WI + CE-T1WI, were constructed using the top ten features with the highest contribution in feature selection in the optimal models of T1WI, T2WI, and CE-T1WI. The Delong test compared areas under the curve (AUC) between models. The AUCs of training/validation/test1/test2 datasets for T1WI, T2WI, and CE-T1WI were 0.900/0.842/0.872/0.839, 0.876/0.789/0.842/0.863, and 0.899/0.824/0.831/0.707, respectively. The fusion model from T1WI + T2WI + CE-T1WI had the highest AUC. The AUCs of training/validation/test1/test2 datasets were 0.947/0.849/0.871/0.887. The T1WI + T2WI + CE-T1WI model demonstrated a significantly higher AUC than the T2WI + CE-T1WI model in both cohorts (p < 0.05) and outperformed the T2WI model in test 1 (p = 0.008) and the T1WI model in test 2 (p = 0.006). This fusion model based on radiomics from T1WI + T2WI + CE-T1WI images and machine learning can improve the power in predicting malignant sinonasal tumors with high accuracy, resilience, and robustness. Our study proposes a radiomics-based machine learning fusion model from T1- and T2-weighted images and contrast-enhanced T1-weighted images, which can non-invasively identify the nature of sinonasal tumors and improve the performance in predicting malignant sinonasal tumors. Differentiating benign and malignant sinonasal tumors is difficult due to similar clinical presentations. A radiomics model from T1 + T2 + contrast-enhanced T1 images can identify the nature of sinonasal tumors. This model can help distinguish benign and malignant sinonasal tumors.

Synthetic CT for gamma knife radiosurgery dose calculation: A feasibility study

PurposeTo determine if MRI-based synthetic CTs (sCT), generated with no predefined pulse sequence, can be used for inhomogeneity correction in routine gamma knife radiosurgery (GKRS) treatment planning dose calculation. MethodsTwo sets of sCTs were generated from T1post and T2 images using cycleGAN. Twenty-eight patients (18 training, 10 validation) were retrospectively selected. The image quality of the generated sCTs was compared with the original CT (oCT) regarding the HU value preservation using histogram comparison, RMSE and MAE, and structural integrity. Dosimetric comparisons were also made among GKRS plans from 3 calculation approaches: TMR10 (oCT), and convolution (oCT and sCT), at four locations: original disease site, bone/tissue interface, air/tissue interface, and mid-brain. ResultsThe study showed that sCTs and oCTs’ HU were similar, with T2-sCT performing better. TMR10 significantly underdosed the target by a mean of 5.4% compared to the convolution algorithm. There was no significant difference in convolution algorithm shot time between the oCT and sCT generated with T2. The highest and lowest dosimetric differences between the two CTs were observed in the bone and air interface, respectively. Dosimetric differences of 3.3% were observed in sCT predicted from MRI with stereotactic frames, which was not included in the training sets. ConclusionsMRI-based sCT can be utilized for GKRS convolution dose calculation without the unnecessary radiation dose, and sCT without metal artifacts could be generated in framed cases. Larger datasets inclusive of all pulse sequences can improve the training set. Further investigation and validation studies are needed before clinical implementation.

Development and validation of a perivascular space segmentation method in multi-center datasets

BackgroundPerivascular spaces (PVS) visible on magnetic resonance imaging (MRI) are significant markers associated with various neurological diseases. Although quantitative analysis of PVS may enhance sensitivity and improve consistency across studies, the field lacks a universally validated method for analyzing images from multi-center studies. MethodsWe annotated PVS on multi-center 3D T1-weighted (T1w) images acquired using scanners from three major vendors (Siemens, General Electric, and Philips). A neural network, mcPVS-Net (multi-center PVS segmentation network), was trained using data from 40 subjects and then tested in a separate cohort of 15 subjects. We assessed segmentation accuracy against ground truth masks tailored for each scanner vendor. Additionally, we evaluated the agreement between segmented PVS volumes and visual scores for each scanner. We also explored correlations between PVS volumes and various clinical factors such as age, hypertension, and white matter hyperintensities (WMH) in a larger sample of 1020 subjects. Furthermore, mcPVS-Net was applied to a new dataset comprising both T1w and T2-weighted (T2w) images from a United Imaging scanner to investigate if PVS volumes could discriminate between subjects with differing visual scores. We also compared the mcPVS-Net with a previously published method that segments PVS from T1 images. ResultsIn the test dataset, mcPVS-Net achieved a mean DICE coefficient of 0.80, with an average Precision of 0.81 and Recall of 0.79, indicating good specificity and sensitivity. The segmented PVS volumes were significantly associated with visual scores in both the basal ganglia (r = 0.541, p < 0.001) and white matter regions (r = 0.706, p < 0.001), and PVS volumes were significantly different among subjects with varying visual scores. Segmentation performance was consistent across different scanner vendors. PVS volumes exhibited significant associations with age, hypertension, and WMH. In the United Imaging scanner dataset, PVS volumes showed good associations with PVS visual scores evaluated on either T1w or T2w images. Compared to a previously published method, mcPVS-Net showed a higher accuracy and improved PVS segmentation in the basal ganglia region. ConclusionThe mcPVS-Net demonstrated good accuracy for segmenting PVS from 3D T1w images. It may serve as a useful tool for future PVS research.

Trapped in Endosome PEGylated Ultra-Small Iron Oxide Nanoparticles Enable Extraordinarily High MR Imaging Contrast for Hepatocellular Carcinomas.

The early diagnosis of hepatocellular carcinomas (HCCs) remains challenging in the clinic. Primovist-enhanced magnetic resonance imaging (MRI) aids HCC diagnosis but loses sensitivity for tumors <2cm. Therefore, developing advanced MRI contrast agents is imperative for improving the diagnostic accuracy of HCCs in very-early-stage. To address this challenge, PEGylated ultra-small iron oxide nanoparticles (PUSIONPs) are synthesized and employed as liver-specific T1 MRI contrast agents. Intravenous delivery produces simultaneous hyperintense HCC and hypointense hepatic parenchyma signals on T1 imaging, creating an extraordinarily high tumor-to-liver contrast. Systematic studies uncover PUSIONP distribution in hepatic parenchyma, HCC lesions at the organ, tissue, cellular, and subcellular levels, revealing endosomal confinement of PUSIONP without aggregation. By mimicking such situations, the dependency of relaxometric properties on local PUSIONP concentration is investigated, emphasizing the key role of different endosomal concentrations in liver and tumor cells for high tumor-to-liver contrast and clear tumor boundaries. These findings offer exceptional imaging capabilities for early HCC diagnosis, potentially benefiting real HCC patients.

Progression predictors of clinically isolated syndrome to multiple sclerosis: A prospective study in China

ObjectivesClinically isolated syndrome (CIS) is a preclinical phase of multiple sclerosis (MS). The progression rate of CIS to clinical definite MS (CDMS) varies significantly across different populations, and identifying predictors of progression is crucial for early diagnosis and treatment. We aimed to investigate predictors of progression from CIS to CDMS in a Chinese cohort. MethodsA single-center cohort study was conducted with newly diagnosed patients with CIS in China between 2018 and 2021. All patients underwent a comprehensive clinical evaluation, including neurological examination, magnetic resonance imaging, and laboratory tests. Follow-up assessments were conducted at regular intervals to monitor disease progression. Progression to CDMS was defined according to the 2017 McDonald criteria. Age, sex, Expanded Disability Status Scale (EDSS) score, number of patients with magnetic resonance imaging gadolinium-enhancing (Gd+) lesions, T2 lesions and Gd+ lesions count, CSF cell count, CSF total protein, CSF and serum neurofilament light chain (NfL), progranulin (PGRN) and Th17-related cytokines (IL-6, IL-17, IL-21, IL-22, IL-23 and TGF-β) were measured for association with risk of progression to CDMS. ResultsA total of 96 CIS patients were recruited in the study. During the at least 24 months follow-up period, 57 (59.38 %) CIS patients progressed to CDMS, while 39 (40.62 %) patients without progression remained stable as CIS. Multivariate analysis revealed that younger age at onset (OR= 43.43, 95 % CI= 1.76–1071.68, p<0.021), higher CSF elevated protein (OR=58.64, 95 % CI=2.72–1264.51, p=0.009), higher CSF NfL levels (OR= 97.00, 95 % CI= 4.68–2012.99, p=0.003) and higher CSF IL-23 levels (OR= 412.02, 95 % CI=6.56–25869.60, p=0.004) were associated with high risk of progression to CDMS. ConclusionYounger age at onset, elevated CSF NfL, IL-23 and protein levels might be progression predictors of CIS to CDMS in Chinese population.

Accuracy of Bathymetric Depth Change Maps Using Multi-Temporal Images and Machine Learning

Most work to date on satellite-derived bathymetry (SDB) depth change estimates water depth at individual times t1 and t2 using two separate models and then differences the model estimates. An alternative approach is explored in this study: a multi-temporal Sentinel-2 image is created by “stacking” the bands of the times t1 and t2 images, geographically coincident reference data for times t1 and t2 allow for “true” depth change to be calculated for the pixels of the multi-temporal image, and this information is used to fit a single model that estimates depth change directly rather than indirectly as in the model-differencing approach. The multi-temporal image approach reduced the depth change RMSE by about 30%. The machine learning modelling method (categorical boosting) outperformed linear regression. Overfitting of models was limited even for the CatBoost models having the maximum number of variables examined. The visible Sentinel-2 spectral bands contributed most to the model predictions. Though the multi-temporal stacked image approach produced clearly superior depth change estimates compared to the conventional approach, it is limited only to those areas for which geographically coincident multi-temporal reference/“true” depth data exist.

Deep learning-based quantification of brain atrophy using 2D T1-weighted MRI for Alzheimer's disease classification.

Determining brain atrophy is crucial for the diagnosis of neurodegenerative diseases. Despite detailed brain atrophy assessments using three-dimensional (3D) T1-weighted magnetic resonance imaging, their practical utility is limited by cost and time. This study introduces deep learning algorithms for quantifying brain atrophy using a more accessible two-dimensional (2D) T1, aiming to achieve cost-effective differentiation of dementia of the Alzheimer's type (DAT) from cognitively unimpaired (CU), while maintaining or exceeding the performance obtained with T1-3D individuals and to accurately predict AD-specific atrophy similarity and atrophic changes [W-scores and Brain Age Index (BAI)]. Involving 924 participants (478 CU and 446 DAT), our deep learning models were trained on cerebrospinal fluid (CSF) volumes from 2D T1 images and compared with 3D T1 images. The performance of the models in differentiating DAT from CU was assessed using receiver operating characteristic analysis. Pearson's correlation analyses were used to evaluate the relations between 3D T1 and 2D T1 measurements of cortical thickness and CSF volumes, AD-specific atrophy similarity, W-scores, and BAIs. Our deep learning models demonstrated strong correlations between 2D and 3D T1-derived CSF volumes, with correlation coefficients r ranging from 0.805 to 0.971. The algorithms based on 2D T1 accurately distinguished DAT from CU with high accuracy (area under the curve values of 0.873), which were comparable to those of algorithms based on 3D T1. Algorithms based on 2D T1 image-derived CSF volumes showed high correlations in AD-specific atrophy similarity (r = 0.915), W-scores for brain atrophy (0.732 ≤ r ≤ 0.976), and BAIs (r = 0.821) compared with those based on 3D T1 images. Deep learning-based analysis of 2D T1 images is a feasible and accurate alternative for assessing brain atrophy, offering diagnostic precision comparable to that of 3D T1 imaging. This approach offers the advantage of the availability of T1-2D imaging, as well as reduced time and cost, while maintaining diagnostic precision comparable to T1-3D.