Image Volume Research Articles

How Reproducible Are the Ultrasound Features of Adenomyosis Defined by the Revised MUSA Consensus?

Background/Objectives: The aim of this study is to assess the inter- and intra-observer reproducibility of the identification of direct and indirect ultrasonographic features of adenomyosis as defined by the revised Morphological Uterus Sonographic Assessment (MUSA) consensus (2022). Methods: A cohort of 74 women, aged 18 to 45, were recruited from the recurrent miscarriage and general gynaecology clinic at a university-based fertility centre. All the participants underwent 2D and 3D transvaginal Ultrasound scan (TVS) examination in the late follicular and early luteal phase. Conventional grey scale and power Doppler image volumes were acquired and stored. Subsequently, the stored 3D ultrasound images were independently re-evaluated offline by the two observers for the direct and indirect features of adenomyosis as outlined by the revised MUSA group. The intra- and the inter-observer reproducibility was estimated using Cohen’s Kappa coefficient. Results: The intra- and interobserver reproducibility (K −0.27, 95% CI 0.06–0.48 and K 0.13, 95% CI −0.10–0.37, respectively) for at least one direct feature of adenomyosis was only modest. Amongst the individual direct features, the interobserver variability of identifying myometrial cysts was fair (K 0.21, 95% CI −0.00–0.42), whereas the intra-observer variability was moderate (K 0.44, 95% CI 0.26–0.63). While hyperechogenic islands identification achieved a fair level of intra- (K 0.31, 95% CI 0.09–0.53) and interobserver (K 0.24, 95% CI 0.01–0.47) agreement, the reproducibility of reporting sub-endometrial lines/buds was fair for the intra-observer (K 0.22, 95% CI −0.02 0.47) and poor for the interobserver (K 0.00, 95% CI −0.20–0.19). The interobserver agreement for indirect features varied from poor to moderate, while the intra-observer agreement ranged between poor to good. Conclusions: The reporting of adenomyosis using direct features suggested by the revised MUSA group consensus showed only modest interobserver and intra-observer agreement. The definitions of ultrasound features for adenomyosis need further refining to enhance the reliability of diagnosis criteria of adenomyosis.

Sparse keypoint segmentation of lung fissures: efficient geometric deep learning for abstracting volumetric images.

Lung fissure segmentation on CT images often relies on 3D convolutional neural networks (CNNs). However, 3D-CNNs are inefficient for detecting thin structures like the fissures, which make up a tiny fraction of the entire image volume. We propose to make lung fissure segmentation more efficient by using geometric deep learning (GDL) on sparse point clouds. We abstract image data with sparse keypoint (KP) clouds. We train GDL models to segment the point cloud, comparing three major paradigms of models (PointNets, graph convolutional networks (GCNs), and PointTransformers). From the sparse point segmentations, 3D meshes of the objects are reconstructed to obtain a dense surface. The state-of-the-art Poisson surface reconstruction (PSR) makes up most of the time in our pipeline. Therefore, we propose an efficient point cloud to mesh autoencoder (PC-AE) that deforms a template mesh to fit a point cloud in a single forward pass. Our pipeline is evaluated extensively and compared to the 3D-CNN gold standard nnU-Net on diverse clinical and pathological data. GCNs yield the best trade-off between inference time and accuracy, being faster with only increased error over the nnU-Net. Our PC-AE also achieves a favorable trade-off, being faster at the error compared to the PSR. We present a KP-based fissure segmentation pipeline that is more efficient than 3D-CNNs and can greatly speed up large-scale analyses. A novel PC-AE for efficient mesh reconstruction from sparse point clouds is introduced, showing promise not only for fissure segmentation. Source code is available on https://github.com/kaftanski/fissure-segmentation-IJCARS.

RegBoost: Enhancing mouse brain image registration using geometric priors and Laplacian interpolation.

We show in this work that incorporating geometric features and geometry processing algorithms for mouse brain image registration broadens the applicability of registration algorithms and improves the registration accuracy of existing methods. We introduce the preprocessing and postprocessing steps in our proposed framework as RegBoost. We develop a method to align the axis of 3D image stacks by detecting the central planes that pass symmetrically through the image volumes. We then find geometric contours by defining external and internal structures to facilitate image correspondences. We establish Dirichlet boundary conditions at these correspondences and find the displacement map throughout the volume using Laplacian interpolation. We discuss the challenges in our standalone framework and demonstrate how our new approaches can improve the results of existing image registration methods. We expect our new approach and algorithms will have critical applications in brain mapping projects.

Experimental Study on the Damage Evolution Mechanism of Siltstone by Water Content and Confining Pressure

ABSTRACTProlonged exposure of deep coal mines to erosion from groundwater results in a gradual accumulation of rock mass damage, which can lead to geological hazards such as deformation and instability. These challenges significantly impact the safe operation of deep coal mines. To understand the mechanisms behind siltstone damage progression related to water content and confining pressure, this study explores the influence of these factors on the deformation and damage evolution of siltstone, employing a combination of rock mechanics testing, numerical simulation, and CT scanning techniques. Results demonstrate that increasing water content reduces the compressive strength of rock, leading to more complex failure modes. In contrast, higher confining pressure strengthens the compressive capacity, thereby suppressing the formation and growth of transverse cracks under compression. Using Avizo software, a three‐dimensional model of siltstone was developed to visualize the distribution of fractures in a three‐dimensional field. In the MATLAB platform, a box dimension algorithm based on three‐dimensional digital volume imaging was developed, employing box dimension theory and digital image storage methods. Fractal analysis reveals that the fractal dimension of internal fractures in loaded samples increases linearly with water content, indicating more extensive fracture development and greater specimen damage. Applying the box dimension from three‐dimensional digital volume images as a metric facilitates characterizing the damage evolution in siltstone under different water content conditions. This approach provides a new means to quantitatively evaluate the growth and complexity of internal fractures in siltstone, offering insights into rock damage progression under varying moisture conditions.

Development of an Open-Source Software Tool for Microstructure Analysis of Materials Using Artificial Intelligence

Investigating the microstructures of materials with microscopy is a key task in quality assurance, the development of new materials, and the optimization of manufacturing processes. However, conventional image analysis often demands significant time for analysis and a large volume of images, and the predictions produced are commonly constrained. Applying deep learning, models can be trained to analyze material microstructures quickly and with greater accuracy. The objective of this study is to provide a method for the automatic segmentation of microstructural images obtained from microscopes or scanning electron microscopes using Convolutional Neural Networks. For this purpose, two software scripts were developed in Python employing OpenCV and the fastai library. The first script is designed to generate reference images, while the second is utilized for training a model and predicting the microstructure in an image. The test of the microstructural analysis using the developed software tools demonstrates that robust prediction results are attainable by using high-quality reference images. This tool has been made available as an open-source on GitHub for public use in materials analysis and can be enhanced and further developed if required.

Associations between Growth Differentiation Factor 15, Cardiac Troponin T, and N-terminal pro-B-type Natriuretic Peptide, and Future Myocardial Fibrosis Assessed by Cardiac Magnetic Resonance Imaging: Data from the Akershus Cardiac Examination 1950 Study.

Myocardial fibrosis is associated with a poor outcome for patients with cardiovascular disease (CVD). Growth differentiation factor 15 (GDF-15) concentrations predict the risk of death in patients with CVD, but the underlying pathophysiological mechanisms are poorly understood. We aimed to assess the associations between biomarkers of cellular stress and inflammation (GDF-15), cardiac injury (cardiac troponin T [cTnT]), and stretch (N-terminal pro-B-type natriuretic peptide [NT-proBNP]), and subsequent focal and diffuse myocardial fibrosis assessed by cardiac magnetic resonance (CMR) imaging. We measured GDF-15, cTnT, and NT-proBNP in 200 study participants without known coronary artery disease or renal dysfunction from the population-based Akershus Cardiac Examination 1950 Study at baseline in 2012 to 2015. Focal myocardial scars and diffuse fibrosis were assessed by late gadolinium enhancement imaging and septal extracellular volume fraction (ECV) by CMR 4 to 7 years later. The relationships between cardiac biomarkers and CMR parameters were assessed by logistic regression analysis adjusted for common cardiovascular risk factors. The median age was 63.9 (interquartile range 63.4-64.5) years and 49% were women. GDF-15 (adjusted odds ratio [aOR] 4.40, 95% CI 1.09-17.72) and cTnT (aOR 1.59, 95% CI 1.01-2.50) were associated with nonischemic scars in the fully adjusted model. cTnT (aOR 2.45, 95% CI 1.41-4.25) and NT-proBNP (aOR 3.12, 95% CI 1.55-6.28) were associated with ischemic scars. None of the biomarkers were significantly associated with elevated ECV. In a general population cohort, GDF-15, an emerging biomarker of cellular stress and inflammation, associates with nonischemic scars. Biomarkers of myocardial injury and stretch associate with ischemic scars, while no biomarker was associated with diffuse fibrosis as assessed by CMR.

Accurate and fast segmentation of filaments and membranes in micrographs and tomograms with TARDIS.

It is now possible to generate large volumes of high-quality images of biomolecules at near-atomic resolution and in near-native states using cryogenic electron microscopy/electron tomography (Cryo-EM/ET). However, the precise annotation of structures like filaments and membranes remains a major barrier towards applying these methods in high-throughput. To address this, we present TARDIS (Transformer-based Rapid Dimensionless Instance Segmentation), a machine-learning framework for fast and accurate annotation of micrographs and tomograms. TARDIS combines deep learning for semantic segmentation with a novel geometric model for precise instance segmentation of various macromolecules. We develop pre-trained models within TARDIS for segmenting microtubules and membranes, demonstrating high accuracy across multiple modalities and resolutions, enabling segmentation of over 13,000 tomograms from the CZI Cryo-Electron Tomography data portal. As a modular framework, TARDIS can be extended to new structures and imaging modalities with minimal modification. TARDIS is open-source and freely available at https://github.com/SMLC-NYSBC/TARDIS, and accelerates analysis of high-resolution biomolecular structural imaging data.

Deep learning model for low-dose CT late iodine enhancement imaging and extracellular volume quantification.

To develop and validate deep learning (DL)-models that denoise late iodine enhancement (LIE) images and enable accurate extracellular volume (ECV) quantification. This study retrospectively included patients with chest discomfort who underwent CT myocardial perfusion + CT angiography + LIE from two hospitals. Two DL models, residual dense network (RDN) and conditional generative adversarial network (cGAN), were developed and validated. 423 patients were randomly divided into training (182 patients), tuning (48 patients), internal validation (92 patients) and external validation group (101 patients). LIEsingle (single-stack image), LIEaveraging (averaging multiple-stack images), LIERDN (single-stack image denoised by RDN) and LIEGAN (single-stack image denoised by cGAN) were generated. We compared image quality score, signal-to-noise (SNR) and contrast-to-noise (CNR) of four LIE sets. The identifiability of denoised images for positive LIE and increased ECV (> 30%) was assessed. The image quality of LIEGAN (SNR: 13.3 ± 1.9; CNR: 4.5 ± 1.1) and LIERDN (SNR: 20.5 ± 4.7; CNR: 7.5 ± 2.3) images was markedly better than that of LIEsingle (SNR: 4.4 ± 0.7; CNR: 1.6 ± 0.4). At per-segment level, the area under the curve (AUC) of LIERDN images for LIE evaluation was significantly improved compared with those of LIEGAN and LIEsingle images (p = 0.040 and p < 0.001, respectively). Meanwhile, the AUC and accuracy of ECVRDN were significantly higher than those of ECVGAN and ECVsingle at per-segment level (p < 0.001 for all). RDN model generated denoised LIE images with markedly higher SNR and CNR than the cGAN-model and original images, which significantly improved the identifiability of visual analysis. Moreover, using denoised single-stack images led to accurate CT-ECV quantification. Question Can the developed models denoise CT-derived late iodine enhancement high images and improve signal-to-noise ratio? Findings The residual dense network model significantly improved the image quality for late iodine enhancement and enabled accurate CT- extracellular volume quantification. Clinical relevance The residual dense network model generates denoised late iodine enhancement images with the highest signal-to-noise ratio and enables accurate quantification of extracellular volume.

Preliminary study: Evaluating the reliability of cone-beam computed tomography images for upper airway volume in orthodontics

Objectives: The primary aim of this study was to evaluate the reliability of cone-beam computed tomography (CBCT) in measuring upper airway volume in orthodontics. The secondary aim was to compare the upper airway volumes between laypeople and military personnel (physically fit individuals). Material and Methods: This study included 36 male participants aged 21–29 years who had no upper airway pathology or history of upper airway surgery. CBCT scans had been previously obtained as part of routine dental checkups using a standardized protocol by the same operator. The digital imaging and communications in the medicine files of the subjects were imported into the 3D analysis software SimPlant® 17 Pro (Materialize Dental, Leuven, Belgium) to facilitate the measurement of upper airway volume. The upper airway volume was divided into three sections and measured on days 0 and 14 by the same operator. The intrarater intra-class correlation coefficient (ICC) and Bland-Altman plot were used to assess reliability. The upper airway volumes of the laypeople and military personnel were compared using Tukey’s pairwise test. Results: The variables demonstrated excellent reliability across all sections of the upper airway; the Superior part of the upper airway has an ICC of 0.993, the middle part of the upper airway has an ICC of 0.995, and the inferior part of the upper airway has an ICC of 0.989. The ICC for the entire upper airway was 0.987. The intrarater plot for superior airway measurement showed that 8 out of 8 measurements (100%) fell within the limits of agreement. Similarly, the intrarater plot for the middle airway measurements indicated that 8 out of 8 measurements (100%) fell within the limits of agreement. For inferior airway measurement, the intrarater plot revealed that 7 out of 8 measurements (87.5%) fell within the limits of agreement. No significant differences were found between the upper airway volumes of the laypeople and military personnel. Conclusion: Preliminary evidence suggests that CBCT using a standardized protocol is reliable for measuring upper airway volume. Separating the upper airway into three different parts also demonstrated high reliability and reproducibility among the raters. The ability to accurately measure the upper airway enhances the evaluation of the effects of orthodontic treatment by comparing the pre-and post-upper airway volume.

Optimized radiofrequency shimming using low-heating B1+-mapping in the presence of deep brain stimulation implants: Proof of concept.

MRI of patients with Deep Brain Stimulation (DBS) implants is constrained due to radiofrequency (RF) heating of the implant lead. However, "RF-shimming" parallel transmission (PTX) has the potential to reduce DBS heating during MRI. As part of using PTX in such a "safe mode", maps of the RF transmission field (B1+) are typically acquired for calibration purposes, with each transmit coil excited individually. These maps often have large zones of low signal intensity distant from the specific coil that is being excited, raising concerns that low signal-to-noise ratio (SNR) in these zones might negatively impact the ability of the optimized RF shim settings to suppress heating in safe mode. One way to improve SNR would be to increase RF transmission power during B1+ mapping, but this also raises heating concerns especially for coil elements proximal to the implant. Acting with an abundance of caution, it would be useful to investigate methods that permit B1+ mapping with low localized heating while producing high SNR measurements that lead to safe PTX RF shim settings. The present work addresses this issue in proof of concept using electromagnetic simulations and experimental PTX MRI. A two-step optimization algorithm is proposed and examined for a cylindrical phantom with an implanted wire to enable 1) robust B1+ mapping with low localized heating; and 2) robust RF shimming PTX with low localized heating and good B1+ homogeneity over a large imaging volume. Simulation and experimental outcomes were compared with those obtained using an existing simulation-driven workflow for obtaining safe mode RF shim settings, and for quadrature RF transmission using a circularly polarized (CP) birdcage head coil. Experimental results showed that although both existing and proposed safe-mode workflows effectively suppressed localized heating at the wire tip in comparison to the CP coil results, the proposed workflow produced much smaller temperature elevations and much improved signal uniformity. These promising results support continued investigation and refinement of the proposed workflow, involving more realistic scenarios toward ultimate implementations in DBS patients.

Multi-slice electron ptychographic tomography for three-dimensional phase-contrast microscopy beyond the depth of focus limits

Abstract Electron ptychography is a powerful computational method for atomic-resolution imaging with high contrast for weakly and strongly scattering elements. Modern algorithms coupled with fast and efficient detectors allow imaging specimens with tens of nanometers thicknesses with sub-0.5 Ångstrom lateral resolution. However, the axial resolution in these approaches is currently limited to a few nanometers, limiting their ability to solve novel atomic structures ab initio. Here, we experimentally demonstrate multi-slice ptychographic electron tomography, which allows atomic resolution three-dimensional phase-contrast imaging in a volume surpassing the depth of field limits. We reconstruct tilt-series 4D-STEM measurements of a Co 3 O 4 nanocube, yielding 2 Å axial and 0.7 Å transverse resolution in a reconstructed volume of ( 18.2 nm ) 3 . Our results demonstrate a 13.5-fold improvement in axial resolution compared to multi-slice ptychography while retaining the atomic lateral resolution and the capability to image volumes beyond the depth of field limit. Multi-slice ptychographic electron tomography significantly expands the volume of materials accessible using high-resolution electron microscopy. We discuss further experimental and algorithmic improvements necessary to also resolve single weakly scattering atoms in 3D.

Deep Learning Methods and UAV Technologies for Crop Disease Detection

The paper underscores the significant advancements in plant disease diagnostics achieved through the integration of remote sensing technologies and deep learning algorithms, particularly in aerial imagery interpretation. It focuses on evaluating deep learning techniques and unmanned aerial vehicles for crop disease detection. (Research purpose) The study aims to review and systemize scientific literature on the application of unmanned aerial vehicles, remote sensing technologies and deep learning 24 methods for the early detection and prediction of crop diseases. (Materials and methods) The paper presents various technologies employing unmanned aerial vehicles and sensors for monitoring plant condition, with an emphasis on modern computer vision tools designed to improve the accuracy of plant pathology identification. (Results and discussion) The analysis encompasses scientific publications from 2010 to 2023, with a primary focus on comparing the effectiveness of deep learning algorithms, such as convolutional neural networks (CNN), against traditional methods, including support vector machines (SVMs) and random forest classifiers. The findings demonstrate that deep learning algorithms offer more accurate and earlier detection of diseases, highlighting their potential for application in plant growing. The paper also addresses challenges associated with the use of unmanned aerial vehicles, such as data quality limitations, the complexity of processing large volumes of images, and the need for the development of more advanced models. The paper proposes solutions to these issues, including algorithm optimization and improved data preprocessing techniques. (Conclusions) The integration of unmanned aerial vehicles and deep learning provides new prospects for enhancing the efficiency of agricultural production. These technologies enable precise early-stage diagnosis of plant diseases and facilitate the prediction of their progression, allowing for timely implementation of crop protection measures. The combination of intelligent computer vision systems with unmanned aerial vehicles presents significant opportunities for advancing monitoring methods and improving plant health management.

Weakly supervised brain tumour segmentation with label propagation and level set loss

AbstractEarly diagnosis of brain tumors significantly enhances treatment success. However, accurate detection and segmentation of tumors, essential for diagnosis, rely heavily on costly manual annotation by experts. To mitigate these costs, weakly supervised methods have gained traction. This paper introduces a novel weakly supervised brain tumor segmentation approach utilizing point and scribble supervision. Experts annotate only the slice with the largest tumor area by marking a single point near the tumor center or drawing a scribble within the tumor region. The method operates in two phases. First, labels are propagated to unlabelled pixels, generating a pseudo‐ground‐truth with three labels: tumor, non‐tumor, and marginal pixels (unlabelled pixels surrounding the initial segmentation). Second, a segmentation model is trained using the pseudo‐ground‐truth and a loss function combining level‐set and binary cross‐entropy losses. Marginal pixels contribute to level‐set loss computation, refining the segmentation process. The approach is validated on 3D magnetic resonance imaging (MRI) volumes from BraTS2020, BraTS2021, and BraTS2023 benchmark datasets. Experimental results show Dice scores comparable to fully supervised methods for whole tumor segmentation, demonstrating the effectiveness of the proposed weakly supervised strategy. This method reduces annotation effort while maintaining competitive segmentation performance, making it valuable for clinical applications.

The CT collateral map: collateral perfusion estimation and baseline lesion assessment after acute anterior circulation ischemic stroke.

To investigate the clinical feasibility of a CT collateral map compared with an MRA collateral map, focusing on collateral perfusion (CP) estimation and baseline lesion assessment in acute ischemic stroke (AIS). This retrospective analysis used selected data from a prospectively collected database. We generated CT collateral maps derived from CT perfusion, encompassing images of arterial, capillary, early venous (CMEV), late venous, and delay phases. Three raters assessed CP scores from MRA and CT collateral maps and CMEV lesion volumes. Lesion volumes of baseline diffusion-weighted imaging (bDWI) and cerebral blood flow rate (CBF) < 30% were automatically measured by the software. The agreement between MRA and CT collateral maps in CP estimation and the correlation between lesion volumes with a CBF < 30% and the CMEV for bDWI lesion volumes were analyzed. One-hundred ten patients (mean age ± standard deviation, 71 ± 14; 60 women) with AIS due to steno-occlusion of the internal carotid and/or middle cerebral arteries were included. The agreement between the MRA and CT collateral maps in CP grading was excellent (weighted κ = 0.93; 95% CI, 0.90-0.97). The concordance correlation coefficients (CCCs) of the CBF < 30% and CMEV for bDWI lesion volumes were 0.76 (95% CI, 0.60-0.91) and 0.97 (0.95-0.98), respectively. The clinical feasibility of the CT collateral map is demonstrated by its significant correlation with the MRA collateral map in CP estimation and baseline lesion assessment.

Impact of Discordance Between Magnetic Resonance Imaging and Ultrasound Volume Measurements on Prostate Fusion Biopsy Outcomes.

To determine whether the difference between MRI-based and ultrasound (US)-based volume measurements are associated with MRI/US-targeted fusion-guided biopsy outcomes. This retrospective, single-center study involved 4177 consecutive patients biopsied between 2010 and 2023 using both MRI/US-targeted fusion and systematic biopsy. Biopsies were indicated because of elevated PSA levels or abnormal multiparametric MRI results. US volume measurements were calculated using the triplane ellipsoid formula, and MRI volumes were obtained by semiautomatic planimetric segmentation. Performance of fusion biopsy compared with systematic biopsy was analyzed with respect to the discordance between MRI and US volume measurements. In 2736 patients (66%), biopsy detected prostate cancer. In cases where both techniques yielded prostate cancers (1695/2736 [62%]), a statistically higher proportion of patients had higher Gleason scores on MRI/US-targeted fusion biopsy compared with systematic biopsy (343 patients [20.2%] vs 137 patients [8.1%], P < .001). MRI volume measurements were significantly smaller compared with US volume measurements (median [IQR] 54 mL [39-77], 56 mL [40-80], respectively, P < .001). Beyond 5 mL volume discordance, MRI/US-targeted fusion biopsy gradually showed less added diagnostic benefit compared with systematic biopsy. In the ≤ 5 mL cohort, MRI/US-targeted fusion biopsy detected more aggressive tumors in 4 times as many patients than systematic biopsy (136 vs 32 patients, P < .001). Although MRI/US-targeted fusion biopsy detected more prostate cancers than systematic biopsy, the performance of MRI/US-targeted fusion biopsy declined with more discordance between volumes measured in MRI vs US. Awareness of volume discordance in MRI and US-based volume measurements should alert the operator about the possibility of reduced performance of MRI/US-targeted fusion biopsy.

Advancing yeast cell analysis: A cryomethod for serial block-face scanning electron microscopy imaging in mitochondrial morphology studies.

Conventional Transmission Electron Microscopy analysis of biological samples often provides limited insights due to its inherent two-dimensional (2D) nature. This limitation hampers a comprehensive understanding of the three-dimensional (3D) complexity of cellular structures, occasionally leading to misinterpretations. Serial block-face scanning electron microscopy emerges as a powerful method for acquiring high-resolution 3D images of cellular volumes. By iteratively removing ultrathin sample sections and capturing images of each newly exposed surface, Serial block-face scanning electron microscopy allows for the meticulous reconstruction of a comprehensive 3D volume. In this study, we investigate the 3D architecture of altered mitochondrial morphologies in Saccharomyces cerevisiae using Serial block-face scanning electron microscopy imaging. We have developed a novel cryomethod based on plunge freezing and a dedicated freeze-substitution protocol. This protocol enhances ultrastructural preservation enabling a more accurate understanding of mitochondrial defects observed in 2D electron microscopy. Our findings underscore the utility of Serial block-face scanning electron microscopy coupled with optimized sample preparation techniques in elucidating complex cellular structures in 3D.

MerQuaCo: a computational tool for quality control in image-based spatial transcriptomics.

Image-based spatial transcriptomics platforms are powerful tools often used to identify cell populations and describe gene expression in intact tissue. Spatial experiments return large, high-dimension datasets and several open-source software packages are available to facilitate analysis and visualization. Spatial results are typically imperfect. For example, local variations in transcript detection probability are common. Software tools to characterize imperfections and their impact on downstream analyses are lacking so the data quality is assessed manually, a laborious and often a subjective process. Here we describe imperfections in a dataset of 641 fresh-frozen adult mouse brain sections collected using the Vizgen MERSCOPE. Common imperfections included the local loss of tissue from the section, tissue outside the imaging volume due to detachment from the coverslip, transcripts missing due to dropped images, varying detection probability through space, and differences in transcript detection probability between experiments. We describe the incidence of each imperfection and the likely impact on the accuracy of cell type labels. We develop MerQuaCo, open-source code that detects and quantifies imperfections without user input, facilitating the selection of sections for further analysis with existing packages. Together, our results and MerQuaCo facilitate rigorous, objective assessment of the quality of spatial transcriptomics results.

Use of Three-Dimensional (3D) Airway Modeling and Virtual Reality for Diagnosis, Communication, and Surgical Planning of Complex Airway Stenosis.

We hypothesized that a method to segment human airways from clinical cases and import them into a case presentation environment in Virtual Reality (VR)could be developed to model and visualize complex airway stenosis for efficient surgical planning. One normal and two pathological airways modeled from CT scans at a slice thickness of 0.625 mm were processed. A multidisciplinary team composed of airway surgeons, VR engineers, educators, and radiologists collaborated to create a clinically relevant VR rendering and explanatory Narrations of the three clinical cases. Segmentation and postprocessing were completed in the Mimics Innovation Suite v24 from Materialize. Structures were segmented from the level of bifurcation of common carotid arteries to the level of bifurcation of the main bronchi, including cartilaginous and bony airway structures, vessels, and soft tissues. They were then postprocessed into 3D image volumes and imported into syGlass (IstoVisio Inc.), a VR software. Direct visualization and free manipulation of these 3D airway models within the VR environment provided improved geometrical and anatomical details compared to traditional two-dimensional (2D) CT. Then, specialized presentation and active learning tools developed for scientific communication using the VR environment permitted the creation of VR Narrations to explain pathological cases. The method to segment human airways from clinical cases used in this paper, combined with intuitive VR tools to overlay segmentation and image data in an active learning environment, shows potential in the use of 3D airway modeling and VR in clinical practice for the description and surgical planning of complex airways. Further work is needed to validate the use of these models in clinical practice and patient education. Level 4.

Safety and therapeutic potential of allogeneic adipose-derived stem cell spray transplantation in ischemic cardiomyopathy: a phase I clinical trial

BackgroundIschemic cardiomyopathy, characterized by coronary artery atherosclerosis, impairs the myocardial tissue. Coronary artery bypass grafting (CABG) is commonly used to revascularize affected areas and improve patient survival rates; however, it can fail to enhance cardiac function. Impaired capillary blood flow may obstruct functional recovery, prompting interest in treatments, such as angiogenic factor administration. Adipose-derived stem cells (ADSCs), which are known for immune evasion, have shown the potential to construct capillary networks and improve myocardial function. This clinical trial aimed to evaluate the safety and efficacy of ADSC spray therapy combined with CABG.MethodsThis single-center, randomized, double-blind study involved patients with ischemic cardiomyopathy who were scheduled for CABG and who had a left ventricular ejection fraction ≤ 40%. The participants were randomized to receive CABG as well as ADSC spray therapy or placebo. The primary endpoints were safety, changes in late gadolinium-enhanced (LGE) magnetic resonance imaging (MRI) volumes, and feasibility. The secondary endpoints included left ventricular function, exercise tolerance, and heart failure symptoms.ResultsSeven patients were enrolled; of them, six were randomized to receive ADSC therapy (n = 3) or placebo (n = 3). The procedure was successfully completed with minimal adverse events. One patient in the ADSC group developed pleural effusion that was resolved with drainage. The LGE-MRI volumes decreased in the ADSC group but remained unchanged in the placebo group. Improvements in left ventricular function and exercise tolerance were noted in the ADSC group, with heart failure symptoms improving to New York Heart Association class I. In contrast, the placebo group showed no significant changes, with one patient experiencing worsening symptoms.ConclusionsADSC spray therapy combined with CABG demonstrated safety and efficacy at enhancing cardiac function. ADSC likely contributes to capillary network reconstruction, thereby augmenting the benefits of CABG. Future phase II and III trials are warranted to confirm its therapeutic efficacy and long-term outcomes. This novel approach represents a significant advancement in the treatment of ischemic cardiomyopathy and offers a viable strategy for improving myocardial function and patient prognosis.Trial registration This study was registered with the Japan Registry of Clinical Trials (jRCT2053190103) and ClinicalTrials.gov (NCT04695522)Graphical

A depth-based land-use semantic extraction method for landscape images

Integrated aerial and ground-based observations are essential for rapid land-use classification in modern land surveys, requiring efficient analysis of large volumes of landscape images. These images often contain background noise that impedes classification accuracy and exhibit spatial compression of the main scene, which causes semantic spatial distortion as depth increases. To address these challenges, this paper proposes a depth-based land-use semantic extraction (DLSE) method that effectively reduces background noise and corrects spatial distortions. The DLSE method follows three main steps: (1) depth estimation per pixel using a multi-resolution neural network to delineate the main scene range, filtering semantic noise; (2) conversion of perspective projection to orthographic projection per pixel to correct spatial distortion; and (3) improved land-use semantic extraction through MobileViT-enhanced feature extraction in SegNet. Experimental results demonstrate that DLSE achieves a 96.84% accuracy with more detailed outputs and operates at 23 fps, positioning it as an efficient tool for automated land surveys and land-use decision-making.