International Journal of Biomedical Imaging

We present a deep learning-based method that corrects motion artifacts and thus accelerates data acquisition and reconstruction of magnetic resonance images. The novel model, the Motion Artifact Correction by Swin Network (MACS-Net), uses a Swin transformer layer as the fundamental block and the Unet architecture as the neural network backbone. We employ a hierarchical transformer with shifted windows to extract multiscale contextual features during encoding. A new dual upsampling technique is employed to enhance the spatial resolutions of feature maps in the Swin transformer-based decoder layer. A raw magnetic resonance imaging dataset is used for network training and testing; the data contain various motion artifacts with ground truth images of the same subjects. The results were compared to six state-of-the-art MRI image motion correction methods using two types of motions. When motions were brief (within 5 s), the method reduced the average normalized root mean square error (NRMSE) from 45.25% to 17.51%, increased the mean structural similarity index measure (SSIM) from 79.43% to 91.72%, and increased the peak signal-to-noise ratio (PSNR) from 18.24 to 26.57 dB. Similarly, when motions were extended from 5 to 10 s, our approach decreased the average NRMSE from 60.30% to 21.04%, improved the mean SSIM from 33.86% to 90.33%, and increased the PSNR from 15.64 to 24.99 dB. The anatomical structures of the corrected images and the motion-free brain data were similar.

Brain tumors are critical neurological ailments caused by uncontrolled cell growth in the brain or skull, often leading to death. An increasing patient longevity rate requires prompt detection; however, the complexities of brain tissue make early diagnosis challenging. Hence, automated tools are necessary to aid healthcare professionals. This study is particularly aimed at improving the efficacy of computerized brain tumor detection in a clinical setting through a deep learning model. Hence, a novel thresholding-based MRI image segmentation approach with a transfer learning model based on contour (ContourTL-Net) is suggested to facilitate the clinical detection of brain malignancies at an initial phase. The model utilizes contour-based analysis, which is critical for object detection, precise segmentation, and capturing subtle variations in tumor morphology. The model employs a VGG-16 architecture priorly trained on the “ImageNet” collection for feature extraction and categorization. The model is designed to utilize its ten nontrainable and three trainable convolutional layers and three dropout layers. The proposed ContourTL-Net model is evaluated on two benchmark datasets in four ways, among which an unseen case is considered as the clinical aspect. Validating a deep learning model on unseen data is crucial to determine the model’s generalization capability, domain adaptation, robustness, and real-world applicability. Here, the presented model’s outcomes demonstrate a highly accurate classification of the unseen data, achieving a perfect sensitivity and negative predictive value (NPV) of 100%, 98.60% specificity, 99.12% precision, 99.56% -score, and 99.46% accuracy. Additionally, the outcomes of the suggested model are compared with state-of-the-art methodologies to further enhance its effectiveness. The proposed solution outperforms the existing solutions in both seen and unseen data, with the potential to significantly improve brain tumor detection efficiency and accuracy, leading to earlier diagnoses and improved patient outcomes.

A challenge in accurately identifying and classifying left ventricular hypertrophy (LVH) is distinguishing it from hypertrophic cardiomyopathy (HCM) and Fabry disease. The reliance on imaging techniques often requires the expertise of multiple specialists, including cardiologists, radiologists, and geneticists. This variability in the interpretation and classification of LVH leads to inconsistent diagnoses. LVH, HCM, and Fabry cardiomyopathy can be differentiated using T1 mapping on cardiac magnetic resonance imaging (MRI). However, differentiation between HCM and Fabry cardiomyopathy using echocardiography or MRI cine images is challenging for cardiologists. Our proposed system named the MRI short-axis view left ventricular hypertrophy classifier (MSLVHC) is a high-accuracy standardized imaging classification model developed using AI and trained on MRI short-axis (SAX) view cine images to distinguish between HCM and Fabry disease. The model achieved impressive performance, with an -score of 0.846, an accuracy of 0.909, and an AUC of 0.914 when tested on the Taipei Veterans General Hospital (TVGH) dataset. Additionally, a single-blinding study and external testing using data from the Taichung Veterans General Hospital (TCVGH) demonstrated the reliability and effectiveness of the model, achieving an -score of 0.727, an accuracy of 0.806, and an AUC of 0.918, demonstrating the model’s reliability and usefulness. This AI model holds promise as a valuable tool for assisting specialists in diagnosing LVH diseases.

Purpose. The Gram-positive Staphylococcus aureus bacterium is one of the leading causes of infection in humans. The lack of specific noninvasive techniques for diagnosis of staphylococcal infection together with the severity of its associated complications support the need for new specific and selective diagnostic tools. This work presents the successful synthesis of an immunotracer that targets the α-toxin released by S. aureus. Methods. [⁸⁹Zr]Zr-DFO-ToxAb was synthesized based on radiolabeling an anti-α-toxin antibody with zirconium-89. The physicochemical characterization of the immunotracer was performed by high-performance liquid chromatography (HPLC), radio-thin layer chromatography (radio-TLC), and electrophoretic analysis. Its diagnostic ability was evaluated in vivo by positron emission tomography/computed tomography (PET/CT) imaging in an animal model of local infection-inflammation (active S. aureus vs. heat-killed S. aureus) and infective osteoarthritis. Results. Chemical characterization of the tracer established the high radiochemical yield and purity of the tracer while maintaining antibody integrity. In vivo PET/CT image confirmed the ability of the tracer to detect active foci of S. aureus. Those results were supported by ex vivo biodistribution studies, autoradiography, and histology, which confirmed the ability of [⁸⁹Zr]Zr-DFO-ToxAb to detect staphylococcal infectious foci, avoiding false-positives derived from inflammatory processes. Conclusions. We have developed an immuno-PET tracer capable of detecting S. aureus infections based on a radiolabeled antibody specific for the staphylococcal alpha toxins. The in vivo assessment of [⁸⁹Zr]Zr-DFO-ToxAb confirmed its ability to selectively detect staphylococcal infectious foci, allowing us to discern between infectious and inflammatory processes.

Background. Ferumoxytol (Ferahame, AMAG Pharmaceuticals, Waltham, MA) is increasingly used off-label as an MR contrast agent due to its relaxivity and safety profiles. However, its potent T2 relaxivity limits achievable T1-weighted positive contrast and leads to artifacts in standard MRI protocols. Optimization of protocols for ferumoxytol deployment is necessary to realize its potential. Methods. We present first-in-human clinical results of the Quantitative Ultrashort Time-to-Echo Contrast Enhanced (QUTE-CE) MRA technique using the superparamagnetic iron oxide nanoparticle agent ferumoxytol for vascular imaging of the head/brain in 15 subjects at 3.0T. The QUTE-CE MRA method was implemented on a 3T scanner using a stack-of-spirals 3D Ultrashort Time-to-Echo sequence. Time-of-flight MRA and standard TE T1-weighted (T1w) images were also collected. For comparison, gadolinium-enhanced blood pool phase images were obtained retrospectively from clinical practice. Signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and intraluminal signal heterogeneity (ISH) were assessed and compared across approaches with Welch’s two-sided -test. Results. Fifteen volunteers ( years old, 9 women) participated. QUTE-CE MRA provided high-contrast snapshots of the arterial and venous networks with lower intraluminal heterogeneity. QUTE-CE demonstrated significantly higher SNR (), blood-tissue CNR (), and lower ISH () compared to ferumoxytol T1-weighted (; ; , respectively) and time-of-flight (; ; , respectively), with in each comparison. The high CNR increased the depth of vessel visualization. Vessel lumina were captured with lower heterogeneity. Conclusion. Quantitative Ultrashort Time-to-Echo Contrast-Enhanced MR angiography provides approximately 5-fold superior contrast with fewer artifacts compared to other contrast-enhanced vascular imaging techniques using ferumoxytol or gadolinium, and to noncontrast time-of-flight MR angiography, for clinical vascular imaging. This trial is registered with NCT03266848.

The kidney’s microstructure, which comprises a highly convoluted tubular and vascular network, can only be partially revealed using classical 2D histology. Considering that the kidney’s microstructure is closely related to its function and is often affected by pathologies, there is a need for powerful and high-resolution 3D imaging techniques to visualize the microstructure. Here, we present how cryogenic contrast-enhanced microCT (cryo-CECT) allowed 3D visualization of glomeruli, tubuli, and vasculature. By comparing different contrast-enhancing staining agents and freezing protocols, we found that the preferred sample preparation protocol was the combination of staining with 1:2 hafnium(IV)-substituted Wells-Dawson polyoxometalate and freezing by submersion in isopentane at −78°C. This optimized protocol showed to be highly sensitive, allowing to detect small pathology-induced microstructural changes in a mouse model of mild trauma-related acute kidney injury after thorax trauma and hemorrhagic shock. In summary, we demonstrated that cryo-CECT is an effective 3D histopathological tool that allows to enhance our understanding of kidney tissue microstructure and their related function.