SEARCH · PubMed Health

Results for “imaging”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources

PubMed2026

stDyer-image improves clustering analysis of spatially resolved transcriptomics and proteomics with morphological images.

MOTIVATION: Spatially resolved transcriptomics (SRT) and spatially resolved proteomics (SRP) data enable the study of gene expression and protein abundances within their precise spatial and cellular contexts in tissues. Certain SRT and SRP technologies also capture corresponding morphology images, adding another layer of valuable information. However, few existing methods developed for SRT data effectively leverage these supplementary images to enhance clustering performance. RESULTS: Here, we introduce stDyer-image, an end-to-end deep learning framework designed for clustering for SRT and SRP datasets with images. Unlike existing methods that utilize images to complement gene expression data, stDyer-image directly links image features to cluster labels. This approach draws inspiration from pathologists, who can visually identify specific cell types or tumor regions from morphological images without relying on gene expression or protein abundances. Benchmarks against state-of-the-art tools demonstrate that stDyer-image achieves superior performance in clustering. Moreover, it is capable of handling large-scale datasets across diverse technologies, making it a versatile and powerful tool for spatial omics analysis. AVAILABILITY AND IMPLEMENTATION: The source code of stDyer-image and detailed tutorials are available at https://github.com/ericcombiolab/stDyer-image.

Proteomics↗

PubMed2026

Genetic Correlation Between Brain Imaging Phenotypes and Externalizing Behavior: A Large-Scale LDSC Analysis of UK Biobank IDPs.

Externalizing has been associated with differences in brain structure and function; however, it remains unclear whether these associations reflect shared common-variant genetic influences. Cross-trait linkage disequilibrium score regression was used to estimate genome-wide genetic correlations between externalizing genome-wide association study (GWAS) results and 3,935 brain imaging-derived phenotypes from the UK Biobank BIG40 resource. The imaging phenotypes covered structural magnetic resonance imaging (MRI), diffusion MRI, susceptibility-weighted imaging, resting-state functional MRI, and task-based functional MRI. Results were included in the primary analysis when the imaging phenotype had positive single-nucleotide polymorphism (SNP) heritability, a heritability Z statistic of at least 1.96, a mean GWAS chi-square statistic of at least 1.02, at least 200,000 regression SNPs, and a complete LDSC result without a fatal error. Technical imaging quality-control phenotypes were excluded from biological inference. Individual results were corrected using the Benjamini-Hochberg false discovery rate procedure. Aggregated Cauchy association tests (ACATs) were used to evaluate evidence across all imaging phenotypes and within predefined imaging categories. Statistical power, simultaneous confidence bounds, and alternative quality-control definitions were examined in sensitivity analyses. Of the 3,935 imaging phenotypes, 3,716 produced estimable genetic correlations, 2,980 met the primary LDSC quality-control criteria, and 2,967 were classified as biological imaging phenotypes. No individual phenotype survived false discovery rate correction. The smallest unadjusted P value was 0.0005, and the minimum adjusted q value was 0.486. The distribution of genetic correlations was centered near zero, with a median genetic correlation of 0.0014 and a median absolute genetic correlation of 0.0338. ACAT provided no evidence of an aggregate association across all biological imaging phenotypes (P = 0.302), and no predefined imaging category survived multiple-testing correction. The median minimum detectable genetic correlation at 80% power was 0.216. Bonferroni-adjusted simultaneous confidence intervals were fully contained within the interval [-0.30, 0.30] for 80.0% of phenotypes in the primary analysis and 88.0% under the stringent heritability quality-control definition. Broad and stringent sensitivity analyses produced the same overall conclusions. In this study, no statistically robust evidence of genome-wide genetic correlations between externalizing and individual UK Biobank brain imaging phenotypes was found. Nevertheless, small, localized, mixed-direction, or developmentally specific genetic effects remain possible.

Journal Article↗

PubMed2026

How, When, and for Whom Does Daily Weight Bias Internalization Undermine Body Image Satisfaction? A Daily Diary Study of Perceived Weight, Exercise Duration, and Gender Differences Among Adolescents.

Although body image satisfaction is increasingly recognized as a dynamic experience that fluctuates across daily contexts, little is known about how weight bias internalization shapes these fluctuations or the cognitive mechanisms and contextual factors involved. The present study examined whether perceived weight mediates the association between weight bias internalization and body image satisfaction, whether exercise duration moderates this association, and whether these effects differ by gender. Daily diary data were collected over 10 consecutive days from 305 Chinese adolescents (Mage = 12.59 years, SD = 0.63 years; 49.2% girls). Multilevel analyses revealed that, at both within- and between-person levels, higher weight bias internalization was associated with lower body image satisfaction. Perceived weight significantly mediated the association between weight bias internalization and body image satisfaction at both levels. Additionally, exploratory analyses using dynamic structural equation modeling further indicated that perceived weight mediated the relationship between weight bias internalization and body image satisfaction at the between-person level, but not at the within-person level. Furthermore, exercise duration moderated the association between perceived weight and body image satisfaction among female adolescents but not among male adolescents. Specifically, the negative association between perceived weight and body image satisfaction was stronger on days when girls engaged in more exercise than usual and among girls with higher average exercise duration. In addition, exercise duration moderated the indirect association between weight bias internalization and body image satisfaction through perceived weight at both within- and between-person levels among girls. These findings highlight the dynamic cognitive processes linking weight bias internalization to body image satisfaction, and the importance of considering gender and daily exercise context in understanding adolescents' body image experiences.

Humans↗

PubMed2026

Diagnostic criteria and severity assessment for syndesmosis injury using magnetic resonance imaging: A systematic review.

High ankle sprains involving syndesmosis injury present challenges in both diagnosis and severity assessment. Magnetic resonance imaging is widely regarded as the preferred modality for evaluating syndesmosis injury and related structural damage. This systematic review primarily examined the diagnostic utility of magnetic resonance imaging. Secondarily, it explores grading and prognostics of syndesmosis injuries with magnetic resonance imaging and identified possible imaging parameters predictive of injury severity. A comprehensive search of MEDLINE, Embase, CINAHL Complete, and Scopus was performed through February 12, 2025, following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Peer-reviewed human studies in English that used magnetic resonance imaging to assess syndesmosis injury were included. Excluded were review articles, case reports, abstract-only studies, and biomechanical or cadaveric investigations. Twenty-seven studies comprising 1931 ankles met inclusion criteria. Magnetic resonance imaging demonstrated high diagnostic accuracy for complete tears of the anterior and posterior inferior tibiofibular ligaments. Ancillary signs such as the ring-of-fire edema pattern, distal tibiofibular joint effusion, and widening of the distal joint space exhibited high specificity with variable sensitivity and may assist in grading injury severity. Magnetic resonance imaging in chronic syndesmosis injury primarily detects fibrotic scarring and post-injury changes. Evidence gaps remain regarding the parameters that best determine injury severity and indicate early surgical intervention in competitive athletes. Consolidating multiple magnetic resonance imaging findings into standardized diagnostic criteria may improve reliability and clinical decision-making.

Humans↗

PubMed2025

Bioluminescence Imaging to Study Recombinant Orthopoxvirus Infection in Animal Models.

Bioluminescent images of viral replication in live animals (in vivo) reveal disease dynamics and effects of medical countermeasures over time. After selecting an appropriate orthopoxvirus animal model for the study, a recombinant virus with the firefly luciferase gene inserted in the genome is used to infect the animals. On the day of bioluminescent imaging, the substrate, D-luciferin, is prepared; animals are sedated and injected with the substrate and IVIS imager is utilized; various bioluminescent images are acquired; then animals recover and are able to continue in the study. Ex vivo imaging can also be completed after animals are euthanized at experimental endpoint. This approach allows real-time imaging of viral kinetics within an animal, and analysis of images can provide an additional quantitative measure throughout the study. Bioluminescent imaging not only provides scientific benefits but also benefits to animal welfare. For these reasons, bioluminescent imaging should be considered for any in vivo orthopoxvirus study.

Animals↗

PubMed2025

MedImg: An Integrated Database for Public Medical Images.

The advancements in deep learning algorithms for medical image analysis have garnered significant attention in recent years. While several studies have shown promising results, with models achieving or even surpassing human performance, translating these advancements into clinical practice is still accompanied by various challenges. A primary obstacle lies in the availability of large-scale, well-characterized datasets for validating the generalization of approaches. To address this challenge, we curated a diverse collection of medical image datasets from multiple public sources, containing 105 datasets and a total of 1,995,671 images. These images span 14 modalities, including X-ray, computed tomography, magnetic resonance imaging, optical coherence tomography, ultrasound, and endoscopy, and originate from 13 organs, such as the lung, brain, eye, and heart. Subsequently, we constructed an online database, MedImg, which incorporates and systematically organizes these medical images to facilitate data accessibility. MedImg serves as an intuitive and open-access platform for facilitating research in deep learning-based medical image analysis, accessible at https://www.cuilab.cn/medimg/.

Humans↗

PubMed2026

Optimizing prostate biopsy decision making - (MUSIC-screen): A 1:1 randomized controlled trial comparing micro-ultrasound versus multiparametric magnetic resonance imaging for prostate cancer diagnosis.

BACKGROUND: Micro-ultrasound (microUS) represents a potential alternative to multiparametric magnetic resonance (mpMRI) in guiding prostate biopsy, with level 1 evidence demonstrating non-inferiority to detect Grade Group ≥2 (GG ≥ 2) prostate cancer in biopsy-naïve men. However, a critical gap remains in the screening pathway, in which imaging is needed to identify men at risk and determine whether biopsy is warranted. METHODS: MUSIC-Screen is a phase 3, open-label, noninferiority 1:1 randomized controlled trial evaluating microUS as an alternative imaging compared to mpMRI for determining the need for prostate biopsy in biopsy- and imaging- naïve men at risk for GG ≥ 2. A total of 1284 men will be randomized to undergo either microUS or mpMRI. Men with PRI-MUS 3-5 or PI-RADS 3-5 lesion will undergo targeted and systematic biopsy. Men with negative imaging and PSA density ≥ 0.15 will undergo systematic biopsy, while those with PSA density < 0.15 will defer biopsy. RESULTS: The primary outcome is GG ≥ 2 detection in each study arm. The primary hypothesis is that microUS is non-inferior to mpMRI for screening and detection of GG ≥ 2. Secondary objectives include comparison of GG ≥ 2 detection rates in targeted cores among patients with PRI-MUS or PI-RADS scores of 3-5, proportion of men who defer biopsy but are diagnosed with GG ≥ 2 within 8 years, assessment of the negative predictive value of each imaging modality, and health economic analyses. CONCLUSION: MUSIC-Screen will determine whether microUS can be used as an imaging modality to inform biopsy decision making that is non-inferior to mpMRI for GG ≥ 2 prostate cancer detection. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT06626022.

Humans↗

PubMed2025

Knowledge-enhanced protein subcellular localization prediction from 3D fluorescence microscope images.

MOTIVATION: Pinpointing the subcellular location of proteins is essential for studying protein function and related diseases. Advances in spatial proteomics have shown that automatic recognition of protein subcellular localization from images could highly facilitate protein translocation analysis and biomarker discovery, but existing machine-learning works have been mostly limited to processing 2D images. By contrast, 3D images have higher spatial resolution and allow researchers to observe cellular structures in their natural context, but currently, there are only a few studies of 3D image processing for protein distribution analysis due to the lack of data and complexity of modeling. RESULTS: We developed a knowledge-enhanced protein subcellular localization model, KE3DLoc, which could recognize distribution patterns in 3D fluorescence microscope images using deep learning methods. The model designs an image feature extraction module that incorporates information from 3D and 2D projected cells and implements asymmetric loss and confidence weights to address data imbalance and weak cell annotation issues. Besides, considering that the biological knowledge in the Gene Ontology (GO) database can provide valuable support for protein location understanding, the KE3DLoc model incorporates a novel knowledge enhancement module that optimizes the protein representation by related knowledge graphs derived from the GO. Since the image module and the knowledge module calculate features from different levels, KE3DLoc designs protein ID aggregation to enhance the consistency of protein features across different cells. Experimental results on three public datasets have demonstrated that the KE3DLoc significantly outperforms existing methods and provides valuable insights for spatial proteomics research. AVAILABILITY AND IMPLEMENTATION: All datasets and codes used in this study are available at GitHub: https://github.com/PRBioimages/KE3DLoc.

Microscopy, Fluorescence↗

PubMed2026

Tumor-Intrinsic Blood and Imaging Correlatives in Advanced Prostate Cancer Treated with Combination Radiopharmaceutical Therapy and Immunotherapy.

The PRINCE trial showed the clinical activity for 177Lu-PSMA-617 in combination with pembrolizumab for metastatic castration-resistant prostate cancer. To refine patient selection and improve response monitoring strategies to this combination, we investigated candidate tumor-intrinsic biomarkers of treatment response and resistance. Methods: We performed circulating tumor DNA (ctDNA), circulating tumor cell (CTC), and PET imaging analyses at baseline, 12 wk on-treatment, and disease progression in participants enrolled in PRINCE (n = 37). We performed targeted sequencing for ctDNA quantification and genomic analysis of more than 70 prostate cancer genes. CTC enumeration was performed on the EpicSciences platform and was combined with selective single-cell whole-genome sequencing. PET imaging included serial PSMA PET as well as 18F-FDG PET imaging at baseline. Results: A low baseline ctDNA fraction and high PSMA avidity in metastatic lesions were linked to superior treatment responses and may have composite biomarker value. Genomic alterations in tumor suppressor genes TP53, RB1, or PTEN were associated with higher 18F-FDG avidity and metabolic tumor volume on 18F-FDG PET imaging and worse prognosis. At 12-wk on-treatment, both ctDNA detection and PSMA PET imaging were strong indicators of response depth and durability. At disease progression, PSMA expression on PET imaging was lower compared with baseline and supported by subclonal remodeling of ctDNA and CTC copy number profiles and by clonal expansions of tumor suppressor gene mutations. Conclusion: We provide the first integrated molecular and imaging insights into determinants of response and resistance to combined radiopharmaceutical therapy and immunotherapy in prostate cancer and propose biomarker strategies to inform future clinical development.

177Lu-PSMA-617↗

PubMed2026

A STORM-based protocol for nanoscale imaging and quantitative analysis of protein-associated and phospholipid-associated structures in natural rubber.

Stochastic Optical Reconstruction Microscopy (STORM) enables nanoscale mapping of molecular components beyond the diffraction limit; however, its reproducible implementation in hydrophobic polymer matrices remains challenging because fluorescence-labeling specificity, fluorophore photoswitching, three-dimensional localization, chromatic registration, and quantitative image analysis must be carefully controlled. This protocol presents a standardized experimental workflow for dual-color labeling, astigmatism-based three-dimensional STORM acquisition, and quantitative analysis of protein-associated and phospholipid-associated structures in natural rubber (NR). The workflow covers sample pretreatment, Cy5 NHS ester labeling of protein-associated primary amines, DiI labeling of phospholipid-rich domains, STORM imaging-buffer preparation, three-dimensional single-molecule localization, dual-channel registration, generation of standardized xy projections, aggregate-size analysis, and projected lateral spatial correlation assessment. Reproducibility is supported by defined acquisition and localization criteria, three independent sample preparations with at least five fields of view analyzed per condition, and unlabeled, single-color, dye-only matrix, and processing-associated Cy5 controls. Mean lateral localization precisions of 11.8 ± 2.3 nm for Cy5 and 13.5 ± 2.9 nm for DiI were obtained, while two-dimensional Fourier ring correlation analysis of the xy projections yielded effective lateral image resolutions of approximately 25 and 28 nm, respectively. Image-based particle segmentation and localization-coordinate-based density-based spatial clustering of applications with noise (DBSCAN) were applied to standardized xy projections as complementary quantitative approaches. Application of the protocol to untreated, centrifuged, and protease-treated NR samples demonstrated treatment-associated changes in the detected abundance and projected size distributions of protein- and phospholipid-associated aggregates, together with a non-monotonic change in their projected lateral spatial correlation. These observations describe alterations in nanoscale organization but do not, by themselves, establish stable protein-phospholipid complex formation. Unlike previous studies that primarily demonstrated the feasibility of STORM imaging in rubber materials, the principal contribution of this work is an end-to-end, step-by-step protocol incorporating defined controls, three-dimensional localization, image-quality metrics, chromatic-registration procedures, and complementary quantitative-analysis pipelines for non-expert users. The workflow may be adaptable to other hydrophobic polymers and soft-material systems after appropriate optimization and validation.

Rubber↗

PubMed2026

Imaging techniques for assessing the hand in systemic sclerosis: a systematic review.

BACKGROUND: Systemic sclerosis (SSc) is a rare autoimmune connective tissue disease frequently associated with hand involvement, leading to significant functional impairment. Imaging techniques provide unique opportunities to visualize and quantify structural and functional abnormalities of the hand, supporting diagnosis, monitoring, and treatment evaluation. This systematic review summarizes the imaging techniques used in SSc. METHODS: A systematic search of PubMed and Embase was conducted. Eligible studies included original research articles in English that applied or evaluated imaging techniques of the hands in SSc, published after 2000. Ultrasound and nailfold capillaroscopy were excluded, given their established use. Screening was performed independently by two authors. Findings were synthesized by clinical manifestations, study quality was assessed using the QUADAS-2 tool. RESULTS: Sixty-one studies met the inclusion criteria. In total, 25 distinct imaging techniques were identified, enabling assessment of various hand structures, including vascular involvement, inflammation, fibrosis, calcifications, erosions, and bone marrow edema. Vascular imaging was most extensively studied, particularly in the context of Raynaud's phenomenon and digital ischemia, with multiple techniques demonstrating impaired perfusion and altered thermoregulatory responses. MRI consistently detected subclinical inflammatory and erosive changes of joints and soft tissues,. CT-based techniques provided detailed assessment of calcinosis cutis, while optical and photoacoustic methods showed promise for quantifying skin fibrosis. CONCLUSION: Imaging techniques provide valuable, complementary insights into hand involvement in SSc, often revealing subclinical disease. Despite promising results, limited standardization and longitudinal validation currently restrict clinical implementation. Future studies should focus on harmonizing protocols and validating against clinically meaningful outcomes.

Humans↗

PubMed2025

The Consortium for Clarity in ADRD Research Through Imaging (CLARiTI): Overview of consortium sites and anticipated enrollment.

INTRODUCTION: The Consortium for Clarity in Alzheimer's disease related dementias (ADRD) Research Through Imaging (CLARiTI) is a study that aims to collect standardized imaging and plasma biomarkers on 2000 Clinical Core participants enrolled across all Alzheimer's Disease Research Centers (ADRC) sites. We sought to summarize the known heterogeneity across centers regarding scientific focus and initial enrollment plans for CLARiTI. METHODS: We developed and distributed a survey capturing information on the 36 CLARiTI site's theme/expertise, recruitment plans, and the intersection of CLARiTI with other ADRC imaging efforts. RESULTS: Anticipated CLARiTI enrollees spanned 11 different categories of suspected etiologies underlying impairment. A wide range of risk factors were endorsed across sites regarding the enrollment of unimpaired individuals. Variability also existed regarding site-level strategies in enrollment into CLARiTI versus other imaging efforts. DISCUSSION: We anticipate that the 2000 individuals that will enroll into CLARiTI will reflect the clinical heterogeneity already in place across the ADRC network. HIGHLIGHTS: The ADRC Consortium for Clarity in ADRD Research Through Imaging (CLARiTI) will leverage and contribute to the existing Alzheimer's Disease Research Centers (ADRC) program by supporting standardized imaging and plasma collection across all centers. We summarize the variation in scientific focus and enrollment plans across ADRC sites participating in CLARiTI. The anticipated CLARiTI cohort will reflect the clinical heterogeneity that already exists across the ADRC network. CLARiTI will contribute to scientific goals related to the detection of multi-etiological signatures relevant for Alzheimer's disease and related disorders (ADRDs).

Humans↗

PubMed2026

Strategy for Simultaneous Multiomic Survey of N-Glycomic and Extracellular Matrix Proteome by Mass Spectrometry Imaging.

Recent advances in spatially resolved molecular profiling have positioned matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) as a powerful platform for multiomic tissue analyses. However, conventional workflows that sequentially target distinct molecular classes are time- and resource-intensive, requiring repeated sequential sample preparation, imaging, and data integration. Here, we evaluate streamlined strategies for simultaneous or combined acquisition of N-glycan and collagen-derived peptide information using PNGase F and collagenase. In-solution studies demonstrate that simultaneous enzymatic digestion yields comparable peptide identifications and glycan profiles relative to traditional sequential workflows, with minimal impact on enzymatic specificity. On the basis of these findings, we developed and optimized MALDI-MSI protocols enabling either simultaneous enzyme application or sequential enzyme treatment with unified matrix deposition and single-pass imaging. While direct coapplication reduced image uniformity, a hybrid approach that used sequential enzyme deposition with combined imaging preserved spatial fidelity and spectral quality while significantly reducing processing and computational demands. Application to human tissues, including vertebral bone and ocular samples, highlights the utility of this workflow for fragile specimens and exploratory multiomic surveys. Collectively, these results establish a framework for integrated glycomic and proteomic imaging targeting the extracellular microenvironment, expanding multiomic MALDI-MSI analyses.

Spectrometry, Mass, Matrix-Assisted Laser Desorpti↗

PubMed2026

Imaging and genomics in stroke.

Imaging after ischemic and hemorrhagic stroke may allow measurement of key phenotypes of injury and recovery for which targeted therapies are still lacking. Such imaging endophenotypes provide quantifiable and heritable biomarkers that can represent mechanistic aspects of disease processes better than clinical measures. Artificial intelligence is allowing extraction of these imaging biomarkers in large cohorts, which can be paired with genomic and other omics data. This will allow the evaluation of what genetic and other biologic variations impact stroke injury and recovery. Integration of these analyses with bioinformatics tools (such as Mendelian randomization and multi-trait analysis) could further dissect how stroke complications overlap with other biologic processes and how they may be causally linked to risk factors. Further work is required to confirm the translational impact of these methods in elucidating mechanisms and drug targets for stroke. However, global collaborations are accelerating analyses on large multi-ethnic stroke cohorts, with availability of imaging data facilitated by federally-funded repositories such as the Imaging Repository for the Cerebrovascular Disease Knowledge Portal (iCDKP).

Humans↗

PubMed2026

Effectiveness of psychologically informed physical therapy, tendon-specific exercise program and routine physical therapy in prolonged unilateral shoulder pain and symptom correlations with imaging: a single-center, randomized, parallel-group, three-arm study (RESPECT).

BACKGROUND: Shoulder complaints are one of the most common musculoskeletal ailments. Patient-specific characteristics such as obesity, depression and physical labor are established risk factors, whereas imaging findings are common and associations between specific imaging findings and symptomatology is limited. General exercises are considered useful in treatment whereas evidence for specific tendon exercises is lacking. Biopsychosocial model is also recommended, but has not been extensively studied concerning shoulder symptoms. This article describes the study protocol designed to evaluate the effectiveness and the cost-effectiveness of routine and specific physical therapy (PT) interventions. Imaging is performed for descriptive, longitudinal and imaging-symptom correlation studies. METHODS: The Rehabilitation of Shoulder Pain: Evaluation and Clinical Trial (RESPECT) is a randomized three-arm parallel-group study involving 300 participants aged 20 to 60 years with prolonged unilateral shoulder pain. Participants will receive either routine PT, physiotherapist-guided tendon-specific exercise program or psychologically informed PT. Bilateral shoulder radiographs, ultrasound and magnetic resonance imaging will be done at the baseline and at 12 and 36 months. Electronic surveys will be completed at the baseline and at 3, 6, 12 and 36 months. The primary outcome will be patient-specific functional scale (PSFS) at 12 months, analyzed using analysis of covariance (ANCOVA), adjusted for baseline PSFS. DISCUSSION: RESPECT will provide systematic and controlled data regarding different PT interventions in prolonged shoulder symptoms, which is currently limited. Being one of the most common sources of musculoskeletal pain, improved management could reduce symptom-related burden and prolonged functional impairment at individual and population level. CLINICALTRIALS: gov; Registration number NCT07235969; Registered November 18th, 2025; Version: 1.0.

Humans↗

PubMed2026

Automated Classification of Lymphoma Subtypes From Histopathological Images Using a U-Net Deep Learning Model: Comparative Evaluation Study.

BACKGROUND: Accurate classification and grading of lymphoma subtypes are essential for treatment planning. Traditional diagnostic methods face challenges of subjectivity and inefficiency, highlighting the need for automated solutions based on deep learning techniques. OBJECTIVE: This study aimed to investigate the application of deep learning technology, specifically the U-Net model, in classifying and grading lymphoma subtypes to enhance diagnostic precision and efficiency. METHODS: In this study, the U-Net model was used as the primary tool for image segmentation integrated with attention mechanisms and residual networks for feature extraction and classification. A total of 620 high-quality histopathological images representing 3 major lymphoma subtypes were collected from The Cancer Genome Atlas and the Cancer Imaging Archive. All images underwent standardized preprocessing, including Gaussian filtering for noise reduction, histogram equalization, and normalization. Data augmentation techniques such as rotation, flipping, and scaling were applied to improve the model's generalization capability. The dataset was divided into training (70%), validation (15%), and test (15%) subsets. Five-fold cross-validation was used to assess model robustness. Performance was benchmarked against mainstream convolutional neural network architectures, including fully convolutional network, SegNet, and DeepLabv3+. RESULTS: The U-Net model achieved high segmentation accuracy, effectively delineating lesion regions and improving the quality of input for classification and grading. The incorporation of attention mechanisms further improved the model's ability to extract key features, whereas the residual structure of the residual network enhanced classification accuracy for complex images. In the test set (N=1250), the proposed fusion model achieved an accuracy of 92% (1150/1250), a sensitivity of 91.04% (1138/1250), a specificity of 89.04% (1113/1250), and an F1-score of 90% (1125/1250) for the classification of the 3 lymphoma subtypes, with an area under the receiver operating characteristic curve of 0.95 (95% CI 0.93-0.97). The high sensitivity and specificity of the model indicate strong clinical applicability, particularly as an assistive diagnostic tool. CONCLUSIONS: Deep learning techniques based on the U-Net architecture offer considerable advantages in the automated classification and grading of lymphoma subtypes. The proposed model significantly improved diagnostic accuracy and accelerated pathological evaluation, providing efficient and precise support for clinical decision-making. Future work may focus on enhancing model robustness through integration with advanced algorithms and validating performance across multicenter clinical datasets. The model also holds promise for deployment in digital pathology platforms and artificial intelligence-assisted diagnostic workflows, improving screening efficiency and promoting consistency in pathological classification.

Humans↗

PubMed2026

Imaging‑based models for predicting cerebrovascular complications of carotid stenosis.

This is a protocol for a Cochrane review (prognosis). The objectives are as follows: Primary objective To systematically review and critically appraise multivariable prognostic models developed for adults (≥ 18 years) with carotid stenosis in which imaging biomarkers (e.g. plaque characteristics derived from magnetic resonance imaging (MRI), computed tomography (CT), or ultrasound) constitute the core predictors. The primary focus is to evaluate the predictive performance of these models for cerebrovascular complications - specifically ipsilateral ischaemic stroke and transient ischaemic attack (TIA) - which are the clinical outcomes to be predicted. Where feasible, we will summarise and compare the models' discrimination (C‑statistic/area under the curve (AUC)) and calibration (calibration‑in‑the‑large, calibration slope, observed‑to‑expected ratio) across studies, and assess their potential for clinical application and external validation. For the purpose of defining symptomatic carotid stenosis as an eligibility criterion and subgroup variable, we will include studies that also considered retinal ischaemia (e.g. retinal embolism, amaurosis fugax) as a qualifying event. Secondary objectives To describe the combinations of imaging markers, modelling techniques, sample sizes, and variable‑selection strategies used in the development of the included models To evaluate the performance of these models for additional secondary clinical outcomes: plaque progression or regression, incident high‑risk imaging features, and the transition from asymptomatic to symptomatic disease To explore whether predictive performance differs according to imaging modality (MRI versus CT versus contrast‑enhanced ultrasound (CEUS)) or technical protocol (e.g. 3 T versus 1.5 T, spectral CT versus conventional CT) For studies that report both cerebrovascular and broader cardiovascular outcomes (major adverse cardiovascular events, myocardial infarction, etc.), we will only extract the performance metrics relating to cerebrovascular events for the primary analysis. Performance metrics for cardiovascular outcomes will be considered exploratory and will not form part of the main synthesis.

Humans↗

PubMed2026

Imaging-Guided Omics Technologies for Resolving Rare Cancer States and Advancing Nanomedicine.

The ability to resolve rare and transient cellular states is critical for understanding metastasis, immune evasion, and therapy resistance in cancer, yet these dynamic processes often escape detection by conventional sequencing and imaging approaches. Recent advances at the interface of nanotechnology, high-resolution live-cell imaging, and single-cell/spatial multiomics methods have enabled functional profiling of cells with unprecedented precision within their native microenvironment. In this Mini-Review, we highlight emerging nanoscale platforms that couple real-time phenotypic imaging with molecular readouts, such as FUNseq and CIN-seq, to directly link functional heterogeneity to transcriptomic, proteomic, and epigenomic information. By integrating nanoscale optical imaging, microengineered perturbation tools, and AI-driven computational analysis, these technologies open up new avenues for dissecting rare metastatic, therapy-resistant, or immune-evasive subpopulations. We further discuss how these next-generation imaging-guided single-cell and spatial omics platforms not only advance fundamental cancer biology but also create opportunities to accelerate the development of nanomedicine applications.

Humans↗