Journal of Imaging

21 pages, 1453 KB

Open AccessReview

Current Trends and Future Opportunities of AI-Based Analysis in Mesenchymal Stem Cell Imaging: A Scoping Review

by Maksim Solopov, Elizaveta Chechekhina, Viktor Turchin, Andrey Popandopulo, Dmitry Filimonov, Anzhelika Burtseva and Roman Ishchenko

J. Imaging 2025, 11(10), 371; https://doi.org/10.3390/jimaging11100371 - 18 Oct 2025

Viewed by 537

Abstract

This scoping review explores the application of artificial intelligence (AI) methods for analyzing mesenchymal stem cells (MSCs) images. The aim of this study was to identify key areas where AI-based image processing techniques are utilized for MSCs analysis, assess their effectiveness, and highlight [...] Read more.

This scoping review explores the application of artificial intelligence (AI) methods for analyzing mesenchymal stem cells (MSCs) images. The aim of this study was to identify key areas where AI-based image processing techniques are utilized for MSCs analysis, assess their effectiveness, and highlight existing challenges. A total of 25 studies published between 2014 and 2024 were selected from six databases (PubMed, Dimensions, Scopus, Google Scholar, eLibrary, and Cochrane) for this review. The findings demonstrate that machine learning algorithms outperform traditional methods in terms of accuracy (up to 97.5%), processing speed and noninvasive capabilities. Among AI methods, convolutional neural networks (CNNs) are the most widely employed, accounting for 64% of the studies reviewed. The primary applications of AI in MSCs image analysis include cell classification (20%), segmentation and counting (20%), differentiation assessment (32%), senescence analysis (12%), and other tasks (16%). The advantages of AI methods include automation of image analysis, elimination of subjective biases, and dynamic monitoring of live cells without the need for fixation and staining. However, significant challenges persist, such as the high heterogeneity of the MSCs population, the absence of standardized protocols for AI implementation, and limited availability of annotated datasets. To advance this field, future efforts should focus on developing interpretable and multimodal AI models, creating standardized validation frameworks and open-access datasets, and establishing clear regulatory pathways for clinical translation. Addressing these challenges is crucial for accelerating the adoption of AI in MSCs biomanufacturing and enhancing the efficacy of cell therapies. Full article

(This article belongs to the Special Issue Deep Learning in Biomedical Image Segmentation and Classification: Advancements, Challenges and Applications, 2nd Edition)

► Show Figures

Figure 1

22 pages, 573 KB

Open AccessArticle

Federated Self-Supervised Few-Shot Face Recognition

by Nursultan Makhanov, Beibut Amirgaliyev, Talgat Islamgozhayev and Didar Yedilkhan

J. Imaging 2025, 11(10), 370; https://doi.org/10.3390/jimaging11100370 - 18 Oct 2025

Viewed by 404

Abstract

This paper presents a systematic framework that combines federated learning, self-supervised learning, and few-shot learning paradigms for privacy-preserving face recognition. We use the large-scale CASIA-WebFace dataset for self-supervised pre-training using SimCLR in a federated setting, followed by federated few-shot fine-tuning on the LFW [...] Read more.

This paper presents a systematic framework that combines federated learning, self-supervised learning, and few-shot learning paradigms for privacy-preserving face recognition. We use the large-scale CASIA-WebFace dataset for self-supervised pre-training using SimCLR in a federated setting, followed by federated few-shot fine-tuning on the LFW dataset using prototypical networks. Through comprehensive evaluation across six state-of-the-art architectures (ResNet, DenseNet, MobileViT, ViT-Small, CvT, and CoAtNet), we demonstrate that while our federated approach successfully preserves data privacy, it comes with significant performance trade-offs. Our results show 12–30% accuracy degradation compared to centralized methods, representing the substantial cost of privacy preservation. We find that traditional CNNs show superior robustness to federated constraints compared to transformer-based architectures, and that five-shot configurations provide an optimal balance between data efficiency and performance. This work provides important empirical insights and establishes benchmarks for federated few-shot face recognition, quantifying the privacy–utility trade-offs that practitioners must consider when deploying such systems in real-world applications. Full article

(This article belongs to the Section Computer Vision and Pattern Recognition)

► Show Figures

Figure 1

28 pages, 4277 KB

Open AccessArticle

Preclinical Application of Computer-Aided High-Frequency Ultrasound (HFUS) Imaging: A Preliminary Report on the In Vivo Characterization of Hepatic Steatosis Progression in Mouse Models

by Sara Gargiulo, Matteo Gramanzini, Denise Bonente, Tiziana Tamborrino, Giovanni Inzalaco, Lisa Gherardini, Lorenzo Franci, Eugenio Bertelli, Virginia Barone and Mario Chiariello

J. Imaging 2025, 11(10), 369; https://doi.org/10.3390/jimaging11100369 - 17 Oct 2025

Viewed by 337

Abstract

Metabolic dysfunction-associated steatotic liver disease (MASLD) is one of the most common chronic liver disorders worldwide and can lead to inflammation, fibrosis, and liver cancer. To better understand the impact of an unbalanced hypercaloric diet on liver phenotype in impaired autophagy, the study [...] Read more.

Metabolic dysfunction-associated steatotic liver disease (MASLD) is one of the most common chronic liver disorders worldwide and can lead to inflammation, fibrosis, and liver cancer. To better understand the impact of an unbalanced hypercaloric diet on liver phenotype in impaired autophagy, the study compared C57BL/6J wild type (WT) and MAPK15-ERK8 knockout (KO) male mice with C57BL/6J background fed for 17 weeks with “Western-type” (WD) or standard diet (SD). Liver features were monitored in vivo by high-frequency ultrasound (HFUS) using a semi-quantitative and parametric assessment of pathological changes in the parenchyma complemented by computer-aided diagnosis (CAD) methods. Liver histology was considered the reference standard. WD induced liver steatosis in both genotypes, although KO mice showed more pronounced dietary effects than WT mice. Overall, HFUS reliably detected steatosis-related parenchymal changes over time in the two mouse genotypes examined, consistent with histology. Furthermore, this study demonstrated the feasibility of extracting quantitative features from conventional B-mode ultrasound images of the liver in murine models at early clinical stages of MASLD using a computationally efficient and vendor-independent CAD method. This approach may contribute to the non-invasive characterization of genetically engineered mouse models of MASLD according to the principles of replacement, reduction, and refinement (3Rs), with interesting translational implications. Full article

(This article belongs to the Special Issue Translational Preclinical Imaging: Techniques, Applications and Perspectives)

► Show Figures

Graphical abstract

19 pages, 2488 KB

Open AccessArticle

Unsupervised Segmentation of Bolus and Residue in Videofluoroscopy Swallowing Studies

by Farnaz Khodami, Mehdy Dousty, James L. Coyle and Ervin Sejdić

J. Imaging 2025, 11(10), 368; https://doi.org/10.3390/jimaging11100368 - 17 Oct 2025

Viewed by 382

Abstract

Bolus tracking is a critical component of swallowing analysis, as the speed, course, and integrity of bolus movement from the mouth to the stomach, along with the presence of residue, serve as key indicators of potential abnormalities. Existing machine learning approaches for videofluoroscopic [...] Read more.

Bolus tracking is a critical component of swallowing analysis, as the speed, course, and integrity of bolus movement from the mouth to the stomach, along with the presence of residue, serve as key indicators of potential abnormalities. Existing machine learning approaches for videofluoroscopic swallowing study (VFSS) analysis heavily rely on annotated data and often struggle to detect residue, which is visually subtle and underrepresented. This study proposes an unsupervised architecture to segment both bolus and residue, marking the first successful machine learning-based residue segmentation in swallowing analysis with quantitative evaluation. We introduce an unsupervised convolutional autoencoder that segments bolus and residue without requiring pixel-level annotations. To address the locality bias inherent in convolutional architectures, we incorporate positional encoding into the input representation, enabling the model to capture global spatial context. The proposed model was validated on a diverse set of VFSS images annotated by certified raters. Our method achieves an intersection over union (IoU) of 61% for bolus segmentation—comparable to state-of-the-art supervised methods—and 52% for residue detection. Despite not using pixel-wise labels for training, our model significantly outperforms top-performing supervised baselines in residue detection, as confirmed by statistical testing. These findings suggest that learning from negative space provides a robust and generalizable pathway for detecting clinically significant but sparsely represented features like residue. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

25 pages, 34240 KB

Open AccessArticle

ImbDef-GAN: Defect Image-Generation Method Based on Sample Imbalance

by Dengbiao Jiang, Nian Tao, Kelong Zhu, Yiming Wang and Haijian Shao

J. Imaging 2025, 11(10), 367; https://doi.org/10.3390/jimaging11100367 - 16 Oct 2025

Viewed by 346

Abstract

In industrial settings, defect detection using deep learning typically requires large numbers of defective samples. However, defective products are rare on production lines, creating a scarcity of defect samples and an overabundance of samples that contain only background. We introduce ImbDef-GAN, a sample [...] Read more.

In industrial settings, defect detection using deep learning typically requires large numbers of defective samples. However, defective products are rare on production lines, creating a scarcity of defect samples and an overabundance of samples that contain only background. We introduce ImbDef-GAN, a sample imbalance generative framework, to address three persistent limitations in defect image generation: unnatural transitions at defect background boundaries, misalignment between defects and their masks, and out-of-bounds defect placement. The framework operates in two stages: (i) background image generation and (ii) defect image generation conditioned on the generated background. In the background image-generation stage, a lightweight StyleGAN3 variant jointly generates the background image and its segmentation mask. A Progress-coupled Gated Detail Injection module uses global scheduling driven by training progress and per-pixel gating to inject high-frequency information in a controlled manner, thereby enhancing background detail while preserving training stability. In the defect image-generation stage, the design augments the background generator with a residual branch that extracts defect features. By blending defect features with a smoothing coefficient, the resulting defect boundaries transition more naturally and gradually. A mask-aware matching discriminator enforces consistency between each defect image and its mask. In addition, an Edge Structure Loss and a Region Consistency Loss strengthen morphological fidelity and spatial constraints within the valid mask region. Extensive experiments on the MVTec AD dataset demonstrate that ImbDef-GAN surpasses existing methods in both the realism and diversity of generated defects. When the generated data are used to train a downstream detector, YOLOv11 achieves a 5.4% improvement in mAP@0.5, indicating that the proposed approach effectively improves detection accuracy under sample imbalance. Full article

(This article belongs to the Section Image and Video Processing)

► Show Figures

Figure 1

4 pages, 170 KB

Open AccessEditorial

Editorial on the Special Issue: “Advances in Retinal Image Processing”

by P. Jidesh and Vasudevan Lakshminarayanan

J. Imaging 2025, 11(10), 366; https://doi.org/10.3390/jimaging11100366 - 16 Oct 2025

Viewed by 229

Abstract

Retinal disorders are one of the major causes of visual impairment [...] Full article

(This article belongs to the Special Issue Advances in Retinal Image Processing)

13 pages, 1736 KB

Open AccessArticle

Automatic Brain Tumor Segmentation in 2D Intra-Operative Ultrasound Images Using Magnetic Resonance Imaging Tumor Annotations

by Mathilde Gajda Faanes, Ragnhild Holden Helland, Ole Solheim, Sébastien Muller and Ingerid Reinertsen

J. Imaging 2025, 11(10), 365; https://doi.org/10.3390/jimaging11100365 - 16 Oct 2025

Viewed by 460

Abstract

Automatic segmentation of brain tumors in intra-operative ultrasound (iUS) images could facilitate localization of tumor tissue during the resection surgery. The lack of large annotated datasets limits the current models performances. In this paper, we investigated the use of tumor annotations in magnetic [...] Read more.

Automatic segmentation of brain tumors in intra-operative ultrasound (iUS) images could facilitate localization of tumor tissue during the resection surgery. The lack of large annotated datasets limits the current models performances. In this paper, we investigated the use of tumor annotations in magnetic resonance imaging (MRI) scans, which are more accessible than annotations in iUS images, for training of deep learning models for iUS brain tumor segmentation. We used 180 annotated MRI scans with corresponding unannotated iUS images, and 29 annotated iUS images. Image registration was performed to transfer the MRI annotations to the corresponding iUS images before training the nnU-Net model with different configurations of the data and label origins. The results showed similar performance for a model trained with only MRI annotated tumors compared to models trained with only iUS annotations and both, and to expert annotations, indicating that MRI tumor annotations can be used as a substitute for iUS tumor annotations to train a deep learning model for automatic brain tumor segmentation in the iUS images. The best model obtained an average Dice score of 0.62 ± 0.31, compared to 0.67 ± 0.25 for an expert neurosurgeon, where the performance on larger tumors was similar, but lower for the models on smaller tumors. In addition, the results showed that removing smaller tumors from the training sets improved the results. Full article

(This article belongs to the Special Issue Progress and Challenges in Biomedical Image Analysis—2nd Edition)

► Show Figures

Figure 1

11 pages, 1181 KB

Open AccessCommunication

Surgical Instrument Segmentation via Segment-Then-Classify Framework with Instance-Level Spatiotemporal Consistency Modeling

by Tiyao Zhang, Xue Yuan and Hongze Xu

J. Imaging 2025, 11(10), 364; https://doi.org/10.3390/jimaging11100364 - 15 Oct 2025

Viewed by 353

Abstract

Accurate segmentation of surgical instruments in endoscopic videos is crucial for robot-assisted surgery and intraoperative analysis. This paper presents a Segment-then-Classify framework that decouples mask generation from semantic classification to enhance spatial completeness and temporal stability. First, a Mask2Former-based segmentation backbone generates class-agnostic [...] Read more.

Accurate segmentation of surgical instruments in endoscopic videos is crucial for robot-assisted surgery and intraoperative analysis. This paper presents a Segment-then-Classify framework that decouples mask generation from semantic classification to enhance spatial completeness and temporal stability. First, a Mask2Former-based segmentation backbone generates class-agnostic instance masks and region features. Then, a bounding box-guided instance-level spatiotemporal modeling module fuses geometric priors and temporal consistency through a lightweight transformer encoder. This design improves interpretability and robustness under occlusion and motion blur. Experiments on the EndoVis 2017 and 2018 datasets demonstrate that our framework achieves mIoU improvements of 3.06%, 2.99%, and 1.67% and mcIoU gains of 2.36%, 2.85%, and 6.06%, respectively, over previously state-of-the-art methods, while maintaining computational efficiency. Full article

(This article belongs to the Section Image and Video Processing)

► Show Figures

Figure 1

11 pages, 1676 KB

Open AccessArticle

Radiographic Markers of Hip Dysplasia and Femoroacetabular Impingement Are Associated with Deterioration in Acetabular and Femoral Cartilage Quality: Insights from T2 MRI Mapping

by Adam Peszek, Kyle S. J. Jamar, Catherine C. Alder, Trevor J. Wait, Caleb J. Wipf, Carson L. Keeter, Stephanie W. Mayer, Charles P. Ho and James W. Genuario

J. Imaging 2025, 11(10), 363; https://doi.org/10.3390/jimaging11100363 - 14 Oct 2025

Viewed by 461

Abstract

Femoroacetabular impingement (FAI) and hip dysplasia have been shown to increase the risk of hip osteoarthritis in affected individuals. MRI with T2 mapping provides an objective measure of femoral and acetabular articular cartilage tissue quality. This study aims to evaluate the relationship between [...] Read more.

Femoroacetabular impingement (FAI) and hip dysplasia have been shown to increase the risk of hip osteoarthritis in affected individuals. MRI with T2 mapping provides an objective measure of femoral and acetabular articular cartilage tissue quality. This study aims to evaluate the relationship between hip morphology measurements collected from three-dimensional (3D) reconstructed computed tomography (CT) scans and the T2 mapping values of hip articular cartilage assessed by three independent, blinded reviewers on the optimal sagittal cut. Hip morphology measures including lateral center edge angle (LCEA), acetabular version, Tönnis angle, acetabular coverage, alpha angle, femoral torsion, neck-shaft angle (FNSA), and combined version were recorded from preoperative CT scans. The relationship between T2 values and hip morphology was assessed using univariate linear mixed models with random effects for individual patients. Significant associations were observed between femoral and acetabular articular cartilage T2 values and all hip morphology measures except femoral torsion. Hip morphology measurements consistent with dysplastic anatomy including decreased LCEA, increased Tönnis angle, and decreased acetabular coverage were associated with increased cartilage damage (p < 0.001 for all). Articular cartilage T2 values were strongly associated with the radiographic markers of hip dysplasia, suggesting hip microinstability significantly contributes to cartilage damage. The relationships between hip morphology measurements and T2 values were similar for the femoral and acetabular sides, indicating that damage to both surfaces is comparable rather than preferentially affecting one side. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

13 pages, 1871 KB

Open AccessArticle

CT Imaging Biomarkers in Rhinogenic Contact Point Headache: Quantitative Phenotyping and Diagnostic Correlations

by Salvatore Lavalle, Salvatore Ferlito, Jerome Rene Lechien, Mario Lentini, Placido Romeo, Alberto Maria Saibene, Gian Luca Fadda and Antonino Maniaci

J. Imaging 2025, 11(10), 362; https://doi.org/10.3390/jimaging11100362 - 14 Oct 2025

Viewed by 349

Abstract

Rhinogenic contact point headache (RCPH) represents a diagnostic challenge due to different anatomical presentations and unstandardized imaging markers. This prospective multicenter study involving 120 patients aimed to develop and validate a CT-based imaging framework for RCPH diagnosis. High-resolution CT scans were systematically assessed [...] Read more.

Rhinogenic contact point headache (RCPH) represents a diagnostic challenge due to different anatomical presentations and unstandardized imaging markers. This prospective multicenter study involving 120 patients aimed to develop and validate a CT-based imaging framework for RCPH diagnosis. High-resolution CT scans were systematically assessed for seven parameters: contact point (CP) type, contact intensity (CI), septal deviation, concha bullosa (CB) morphology, mucosal edema (ME), turbinate hypertrophy (TH), and associated anatomical variants. Results revealed CP-I (37.5%) and CP-II (22.5%) as predominant patterns, with moderate CI (45.8%) and septal deviation > 15° (71.7%) commonly observed. CB was found in 54.2% of patients, primarily bulbous type (26.7%). Interestingly, focal ME at CP was independently associated with greater pain severity in the multivariate model (p = 0.003). The framework demonstrated substantial to excellent interobserver reliability (κ = 0.76–0.91), with multivariate analysis identifying moderate–severe CI, focal ME, and specific septal deviation patterns as independent predictors of higher pain scores. Our imaging classification system highlights key radiological biomarkers associated with symptom severity and may facilitate future applications in quantitative imaging, automated phenotyping, and personalized treatment approaches. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

19 pages, 2435 KB

Open AccessArticle

A Lesion-Aware Patch Sampling Approach with EfficientNet3D-UNet for Robust Multiple Sclerosis Lesion Segmentation

by Hind Almaaz and Samia Dardouri

J. Imaging 2025, 11(10), 361; https://doi.org/10.3390/jimaging11100361 - 13 Oct 2025

Viewed by 431

Abstract

Accurate segmentation of multiple sclerosis (MS) lesions from 3D MRI scans is essential for diagnosis, disease monitoring, and treatment planning. However, this task remains challenging due to the sparsity, heterogeneity, and subtle appearance of lesions, as well as the difficulty in obtaining high-quality [...] Read more.

Accurate segmentation of multiple sclerosis (MS) lesions from 3D MRI scans is essential for diagnosis, disease monitoring, and treatment planning. However, this task remains challenging due to the sparsity, heterogeneity, and subtle appearance of lesions, as well as the difficulty in obtaining high-quality annotations. In this study, we propose Efficient-Net3D-UNet, a deep learning framework that combines compound-scaled MBConv3D blocks with a lesion-aware patch sampling strategy to improve volumetric segmentation performance across multi-modal MRI sequences (FLAIR, T1, and T2). The model was evaluated against a conventional 3D U-Net baseline using standard metrics including Dice similarity coefficient, precision, recall, accuracy, and specificity. On a held-out test set, EfficientNet3D-UNet achieved a Dice score of 48.39%, precision of 49.76%, and recall of 55.41%, outperforming the baseline 3D U-Net, which obtained a Dice score of 31.28%, precision of 32.48%, and recall of 43.04%. Both models reached an overall accuracy of 99.14%. Notably, EfficientNet3D-UNet also demonstrated faster convergence and reduced overfitting during training. These results highlight the potential of EfficientNet3D-UNet as a robust and computationally efficient solution for automated MS lesion segmentation, offering promising applicability in real-world clinical settings. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

23 pages, 4523 KB

Open AccessArticle

Lung Nodule Malignancy Classification Integrating Deep and Radiomic Features in a Three-Way Attention-Based Fusion Module

by Sadaf Khademi, Shahin Heidarian, Parnian Afshar, Arash Mohammadi, Abdul Sidiqi, Elsie T. Nguyen, Balaji Ganeshan and Anastasia Oikonomou

J. Imaging 2025, 11(10), 360; https://doi.org/10.3390/jimaging11100360 - 13 Oct 2025

Viewed by 431

Abstract

In this study, we propose a novel hybrid framework for assessing the invasiveness of an in-house dataset of 114 pathologically proven lung adenocarcinomas presenting as subsolid nodules on Computed Tomography (CT). Nodules were classified into group 1 (G1), which included atypical adenomatous hyperplasia, [...] Read more.

In this study, we propose a novel hybrid framework for assessing the invasiveness of an in-house dataset of 114 pathologically proven lung adenocarcinomas presenting as subsolid nodules on Computed Tomography (CT). Nodules were classified into group 1 (G1), which included atypical adenomatous hyperplasia, adenocarcinoma in situ, and minimally invasive adenocarcinomas, and group 2 (G2), which included invasive adenocarcinomas. Our approach includes a three-way Integration of Visual, Spatial, and Temporal features with Attention, referred to as I-VISTA, obtained from three processing algorithms designed based on Deep Learning (DL) and radiomic models, leading to a more comprehensive analysis of nodule variations. The aforementioned processing algorithms are arranged in the following three parallel paths: (i) The Shifted Window (SWin) Transformer path, which is a hierarchical vision Transformer that extracts nodules’ related spatial features; (ii) The Convolutional Auto-Encoder (CAE) Transformer path, which captures informative features related to inter-slice relations via a modified Transformer encoder architecture; and (iii) a 3D Radiomic-based path that collects quantitative features based on texture analysis of each nodule. Extracted feature sets are then passed through the Criss-Cross attention fusion module to discover the most informative feature patterns and classify nodules type. The experiments were evaluated based on a ten-fold cross-validation scheme. I-VISTA framework achieved the best performance of overall accuracy, sensitivity, and specificity (mean ± std) of 93.93 ± 6.80%, 92.66 ± 9.04%, and 94.99 ± 7.63% with an Area under the ROC Curve (AUC) of 0.93 ± 0.08 for lung nodule classification among ten folds. The hybrid framework integrating DL and hand-crafted 3D Radiomic model outperformed the standalone DL and hand-crafted 3D Radiomic model in differentiating G1 from G2 subsolid nodules identified on CT. Full article

(This article belongs to the Special Issue Progress and Challenges in Biomedical Image Analysis—2nd Edition)

► Show Figures

Figure 1

29 pages, 2757 KB

Open AccessArticle

Non-Contrast Brain CT Images Segmentation Enhancement: Lightweight Pre-Processing Model for Ultra-Early Ischemic Lesion Recognition and Segmentation

by Aleksei Samarin, Alexander Savelev, Aleksei Toropov, Aleksandra Dozortseva, Egor Kotenko, Artem Nazarenko, Alexander Motyko, Galiya Narova, Elena Mikhailova and Valentin Malykh

J. Imaging 2025, 11(10), 359; https://doi.org/10.3390/jimaging11100359 - 13 Oct 2025

Viewed by 510

Abstract

Timely identification and accurate delineation of ultra-early ischemic stroke lesions in non-contrast computed tomography (CT) scans of the human brain are of paramount importance for prompt medical intervention and improved patient outcomes. In this study, we propose a deep learning-driven methodology specifically designed [...] Read more.

Timely identification and accurate delineation of ultra-early ischemic stroke lesions in non-contrast computed tomography (CT) scans of the human brain are of paramount importance for prompt medical intervention and improved patient outcomes. In this study, we propose a deep learning-driven methodology specifically designed for segmenting ultra-early ischemic regions, with a particular emphasis on both the ischemic core and the surrounding penumbra during the initial stages of stroke progression. We introduce a lightweight preprocessing model based on convolutional filtering techniques, which enhances image clarity while preserving the structural integrity of medical scans, a critical factor when detecting subtle signs of ultra-early ischemic strokes. Unlike conventional preprocessing methods that directly modify the image and may introduce artifacts or distortions, our approach ensures the absence of neural network-induced artifacts, which is especially crucial for accurate diagnosis and segmentation of ultra-early ischemic lesions. The model employs predefined differentiable filters with trainable parameters, allowing for artifact-free and precision-enhanced image refinement tailored to the challenges of ultra-early stroke detection. In addition, we incorporated into the combined preprocessing pipeline a newly proposed trainable linear combination of pretrained image filters, a concept first introduced in this study. For model training and evaluation, we utilize a publicly available dataset of acute ischemic stroke cases, focusing on the subset relevant to ultra-early stroke manifestations, which contains annotated non-contrast CT brain scans from 112 patients. The proposed model demonstrates high segmentation accuracy for ultra-early ischemic regions, surpassing existing methodologies across key performance metrics. The results have been rigorously validated on test subsets from the dataset, confirming the effectiveness of our approach in supporting the early-stage diagnosis and treatment planning for ultra-early ischemic strokes. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

16 pages, 571 KB

Open AccessArticle

Lightweight Statistical and Texture Feature Approach for Breast Thermogram Analysis

by Ana P. Romero-Carmona, Jose J. Rangel-Magdaleno, Francisco J. Renero-Carrillo, Juan M. Ramirez-Cortes and Hayde Peregrina-Barreto

J. Imaging 2025, 11(10), 358; https://doi.org/10.3390/jimaging11100358 - 13 Oct 2025

Viewed by 365

Abstract

Breast cancer is the most commonly diagnosed cancer in women globally and represents the leading cause of mortality related to malignant tumors. Currently, healthcare professionals are focused on developing and implementing innovative techniques to improve the early detection of this disease. Thermography, studied [...] Read more.

Breast cancer is the most commonly diagnosed cancer in women globally and represents the leading cause of mortality related to malignant tumors. Currently, healthcare professionals are focused on developing and implementing innovative techniques to improve the early detection of this disease. Thermography, studied as a complementary method to traditional approaches, captures infrared radiation emitted by tissues and converts it into data about skin surface temperature. During tumor development, angiogenesis occurs, increasing blood flow to support tumor growth, which raises the surface temperature in the affected area. Automatic classification techniques have been explored to analyze thermographic images and develop an optimal classification tool to identify thermal anomalies. This study aims to design a concise description using statistical and texture features to accurately classify thermograms as control or highly probable to be cancer (with thermal anomalies). The importance of employing a short description lies in facilitating interpretation by medical professionals. In contrast, a characterization based on a large number of variables could make it more challenging to identify which values differentiate the thermograms between groups, thereby complicating the explanation of results to patients. A maximum accuracy of 91.97% was achieved by applying only seven features and using a Coarse Decision Tree (DT) classifier and robust Machine Learning (ML) model, which demonstrated competitive performance compared with previously reported studies. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

18 pages, 3321 KB

Open AccessArticle

New Solution for Segmental Assessment of Left Ventricular Wall Thickness, Using Anatomically Accurate and Highly Reproducible Automated Cardiac MRI Software

by Balázs Mester, Kristóf Attila Farkas-Sütő, Júlia Magdolna Tardy, Kinga Grebur, Márton Horváth, Flóra Klára Gyulánczi, Hajnalka Vágó, Béla Merkely and Andrea Szűcs

J. Imaging 2025, 11(10), 357; https://doi.org/10.3390/jimaging11100357 - 11 Oct 2025

Viewed by 405

Abstract

Introduction: Changes in left ventricular (LV) wall thickness serve as important diagnostic and prognostic indicators in various cardiovascular diseases. To date, no automated software exists for the measurement of myocardial segmental wall thickness in cardiac MRI (CMR), which leads to reliance on manual [...] Read more.

Introduction: Changes in left ventricular (LV) wall thickness serve as important diagnostic and prognostic indicators in various cardiovascular diseases. To date, no automated software exists for the measurement of myocardial segmental wall thickness in cardiac MRI (CMR), which leads to reliance on manual caliper measurements that carry risks of inaccuracy. Aims: This paper aims to present a new automated segmental wall thickness measurement software, OptiLayer, developed to address this issue and to compare it with the conventional manual measurement method. Methods: In our pilot study, the algorithm of the OptiLayer software was tested on 50 HEALTHY individuals, and 50 excessively trabeculated noncompaction (LVET) subjects with preserved LV function, whose morphology makes it more challenging to measure left ventricular wall thickness, although often occurring with myocardial thinning. Measurements were performed by two independent investigators who assessed LV wall thicknesses in 16 segments, both manually using the Medis Suite QMass program and automatically with the new OptiLayer method, which enables high-density sampling across the distance between the epicardial and endocardial contours. Results: The results showed that the segmental wall thickness measurement values of the OptiLayer algorithm were significantly higher than those of the manual caliper. In comparisons of the HEALTHY and LVET subgroups, OptiLayer measurements demonstrated differences at several points than manual measurements. Between the investigators, manual measurements showed low intraclass correlations (ICC below 0.6 on average), while measurements with OptiLayer gave excellent agreement (ICC above 0.9 in 75% of segments). Conclusions: Our study suggests that OptiLayer, a new automated wall thickness measurement software based on high-precision anatomical segmentation, offers a faster, more accurate, and more reproducible alternative to manual measurements. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

20 pages, 5063 KB

Open AccessArticle

AI Diffusion Models Generate Realistic Synthetic Dental Radiographs Using a Limited Dataset

by Brian Kirkwood, Byeong Yeob Choi, James Bynum and Jose Salinas

J. Imaging 2025, 11(10), 356; https://doi.org/10.3390/jimaging11100356 - 11 Oct 2025

Viewed by 626

Abstract

Generative Artificial Intelligence (AI) has the potential to address the limited availability of dental radiographs for the development of Dental AI systems by creating clinically realistic synthetic dental radiographs (SDRs). Evaluation of artificially generated images requires both expert review and objective measures of [...] Read more.

Generative Artificial Intelligence (AI) has the potential to address the limited availability of dental radiographs for the development of Dental AI systems by creating clinically realistic synthetic dental radiographs (SDRs). Evaluation of artificially generated images requires both expert review and objective measures of fidelity. A stepwise approach was used to processing 10,000 dental radiographs. First, a single dentist screened images to determine if specific image selection criterion was met; this identified 225 images. From these, 200 images were randomly selected for training an AI image generation model. Second, 100 images were randomly selected from the previous training dataset and evaluated by four dentists; the expert review identified 57 images that met image selection criteria to refine training for two additional AI models. The three models were used to generate 500 SDRs each and the clinical realism of the SDRs was assessed through expert review. In addition, the SDRs generated by each model were objectively evaluated using quantitative metrics: Fréchet Inception Distance (FID) and Kernel Inception Distance (KID). Evaluation of the SDR by a dentist determined that expert-informed curation improved SDR realism, and refinement of model architecture produced further gains. FID and KID analysis confirmed that expert input and technical refinement improve image fidelity. The convergence of subjective and objective assessments strengthens confidence that the refined model architecture can serve as a foundation for SDR image generation, while highlighting the importance of expert-informed data curation and domain-specific evaluation metrics. Full article

(This article belongs to the Topic Machine Learning and Deep Learning in Medical Imaging)

► Show Figures

Figure 1

28 pages, 2961 KB

Open AccessArticle

An Improved Capsule Network for Image Classification Using Multi-Scale Feature Extraction

by Wenjie Huang, Ruiqing Kang, Lingyan Li and Wenhui Feng

J. Imaging 2025, 11(10), 355; https://doi.org/10.3390/jimaging11100355 - 10 Oct 2025

Viewed by 434

Abstract

In the realm of image classification, the capsule network is a network topology that packs the extracted features into many capsules, performs sophisticated capsule screening using a dynamic routing mechanism, and finally recognizes that each capsule corresponds to a category feature. Compared with [...] Read more.

In the realm of image classification, the capsule network is a network topology that packs the extracted features into many capsules, performs sophisticated capsule screening using a dynamic routing mechanism, and finally recognizes that each capsule corresponds to a category feature. Compared with previous network topologies, the capsule network has more sophisticated operations, uses a large number of parameter matrices and vectors to express picture attributes, and has more powerful image classification capabilities. However, in the practical application field, the capsule network has always been constrained by the quantity of calculation produced by the complicated structure. In the face of basic datasets, it is prone to over-fitting and poor generalization and often cannot satisfy the high computational overhead when facing complex datasets. Based on the aforesaid problems, this research proposes a novel enhanced capsule network topology. The upgraded network boosts the feature extraction ability of the network by incorporating a multi-scale feature extraction module based on proprietary star structure convolution into the standard capsule network. At the same time, additional structural portions of the capsule network are changed, and a variety of optimization approaches such as dense connection, attention mechanism, and low-rank matrix operation are combined. Image classification studies are carried out on different datasets, and the novel structure suggested in this paper has good classification performance on CIFAR-10, CIFAR-100, and CUB datasets. At the same time, we also achieved 98.21% and 95.38% classification accuracy on two complicated datasets of skin cancer ISIC derived and Forged Face EXP. Full article

(This article belongs to the Section Image and Video Processing)

► Show Figures

Figure 1

19 pages, 3418 KB

Open AccessArticle

WSVAD-CLIP: Temporally Aware and Prompt Learning with CLIP for Weakly Supervised Video Anomaly Detection

by Min Li, Jing Sang, Yuanyao Lu and Lina Du

J. Imaging 2025, 11(10), 354; https://doi.org/10.3390/jimaging11100354 - 10 Oct 2025

Viewed by 828

Abstract

Weakly Supervised Video Anomaly Detection (WSVAD) is a critical task in computer vision. It aims to localize and recognize abnormal behaviors using only video-level labels. Without frame-level annotations, it becomes significantly challenging to model temporal dependencies. Given the diversity of abnormal events, it [...] Read more.

Weakly Supervised Video Anomaly Detection (WSVAD) is a critical task in computer vision. It aims to localize and recognize abnormal behaviors using only video-level labels. Without frame-level annotations, it becomes significantly challenging to model temporal dependencies. Given the diversity of abnormal events, it is also difficult to model semantic representations. Recently, the cross-modal pre-trained model Contrastive Language-Image Pretraining (CLIP) has shown a strong ability to align visual and textual information. This provides new opportunities for video anomaly detection. Inspired by CLIP, WSVAD-CLIP is proposed as a framework that uses its cross-modal knowledge to bridge the semantic gap between text and vision. First, the Axial-Graph (AG) Module is introduced. It combines an Axial Transformer and Lite Graph Attention Networks (LiteGAT) to capture global temporal structures and local abnormal correlations. Second, a Text Prompt mechanism is designed. It fuses a learnable prompt with a knowledge-enhanced prompt to improve the semantic expressiveness of category embeddings. Third, the Abnormal Visual-Guided Text Prompt (AVGTP) mechanism is proposed to aggregate anomalous visual context for adaptively refining textual representations. Extensive experiments on UCF-Crime and XD-Violence datasets show that WSVAD-CLIP notably outperforms existing methods in coarse-grained anomaly detection. It also achieves superior performance in fine-grained anomaly recognition tasks, validating its effectiveness and generalizability. Full article

(This article belongs to the Section Computer Vision and Pattern Recognition)

► Show Figures

Figure 1

20 pages, 34236 KB

Open AccessArticle

ILD-Slider: A Parameter-Efficient Model for Identifying Progressive Fibrosing Interstitial Lung Disease from Chest CT Slices

by Jiahao Zhang, Shoya Wada, Kento Sugimoto, Takayuki Niitsu, Kiyoharu Fukushima, Hiroshi Kida, Bowen Wang, Shozo Konishi, Katsuki Okada, Yuta Nakashima and Toshihiro Takeda

J. Imaging 2025, 11(10), 353; https://doi.org/10.3390/jimaging11100353 - 9 Oct 2025

Viewed by 441

Abstract

Progressive Fibrosing Interstitial Lung Disease (PF-ILD) is a severe phenotype of Interstitial Lung Disease (ILD) with a poor prognosis, typically requiring prolonged clinical observation and multiple CT examinations for diagnosis. Such requirements delay early detection and treatment initiation. To enable earlier identification of [...] Read more.

Progressive Fibrosing Interstitial Lung Disease (PF-ILD) is a severe phenotype of Interstitial Lung Disease (ILD) with a poor prognosis, typically requiring prolonged clinical observation and multiple CT examinations for diagnosis. Such requirements delay early detection and treatment initiation. To enable earlier identification of PF-ILD, we propose ILD-Slider, a parameter-efficient and lightweight deep learning framework that enables accurate PF-ILD identification from a limited number of CT slices. ILD-Slider introduces anatomy-based position markers (PMs) to guide the selection of representative slices (RSs). A PM extractor, trained via a multi-class classification model, achieves high PM detection accuracy despite severe class imbalance by leveraging a peak slice mining (PSM)-based strategy. Using the PM extractor, we automatically select three, five, or nine RSs per case, substantially reducing computational cost while maintaining diagnostic accuracy. The selected RSs are then processed by a slice-level 3D Adapter (Slider) for PF-ILD identification. Experiments on 613 cases from The University of Osaka Hospital (UOH) and the National Hospital Organization Osaka Toneyama Medical Center (OTMC) demonstrate the effectiveness of ILD-Slider, achieving an AUPRC of 0.790 (AUROC 0.847) using only five automatically extracted RSs. ILD-Slider further validates the feasibility of diagnosing PF-ILD from non-contiguous slices, which is particularly valuable for real-world and public datasets where contiguous volumes are often unavailable. These results highlight ILD-Slider as a practical and efficient solution for early PF-ILD identification. Full article

(This article belongs to the Special Issue Advances in Medical Imaging and Machine Learning)

► Show Figures

Figure 1

17 pages, 4166 KB

Open AccessArticle

Non-Destructive Volume Estimation of Oranges for Factory Quality Control Using Computer Vision and Ensemble Machine Learning

by Wattanapong Kurdthongmee and Arsanchai Sukkuea

J. Imaging 2025, 11(10), 352; https://doi.org/10.3390/jimaging11100352 - 9 Oct 2025

Viewed by 282

Abstract

A crucial task in industrial quality control, especially in the food and agriculture sectors, is the quick and precise estimation of an object’s volume. This study combines cutting-edge machine learning and computer vision techniques to provide a comprehensive, non-destructive method for predicting orange [...] Read more.

A crucial task in industrial quality control, especially in the food and agriculture sectors, is the quick and precise estimation of an object’s volume. This study combines cutting-edge machine learning and computer vision techniques to provide a comprehensive, non-destructive method for predicting orange volume. We created a reliable pipeline that employs top and side views of every orange to estimate four important dimensions using a calibrated marker. These dimensions are then fed into a machine learning model that has been fine-tuned. Our method uses a range of engineered features, such as complex surface-area-to-volume ratios and new shape-based descriptors, to go beyond basic geometric formulas. Based on a dataset of 150 unique oranges, we show that the Stacking Regressor performs significantly better than other single-model benchmarks, including the highly tuned LightGBM model, achieving an

R^{2}

score of 0.971. Because of its reliance on basic physical characteristics, the method is extremely resilient to the inherent variability in fruit and may be used with a variety of produce types. Because it allows for the real-time calculation of density (mass over volume) for automated defect detection and quality grading, this solution is directly applicable to a factory sorting environment. Full article

(This article belongs to the Topic Nondestructive Testing and Evaluation)

► Show Figures

Figure 1

19 pages, 3520 KB

Open AccessArticle

Multifactorial Imaging Analysis as a Platform for Studying Cellular Senescence Phenotypes

by Shatalova Rimma, Larin Ilya and Shevyrev Daniil

J. Imaging 2025, 11(10), 351; https://doi.org/10.3390/jimaging11100351 - 8 Oct 2025

Viewed by 585

Abstract

Cellular senescence is a heterogeneous and dynamic state characterised by stable proliferation arrest, macromolecular damage and metabolic remodelling. Although markers such as SA-β-galactosidase staining, yH2AX foci and p53 activation are widely used as de facto standards, they are imperfect and differ in terms [...] Read more.

Cellular senescence is a heterogeneous and dynamic state characterised by stable proliferation arrest, macromolecular damage and metabolic remodelling. Although markers such as SA-β-galactosidase staining, yH2AX foci and p53 activation are widely used as de facto standards, they are imperfect and differ in terms of sensitivity, specificity and dependence on context. We present a multifactorial imaging platform integrating scanning electron, flow cytometry and high-resolution confocal microscopy. This allows us to identify senescence phenotypes in three in vitro models: replicative ageing via serial passaging; dose-graded genotoxic stress under serum deprivation; and primary fibroblasts from young and elderly donors. We present a multimodal imaging framework to characterise senescence-associated phenotypes by integrating LysoTracker and MitoTracker microscopy and SA-β-gal/FACS, p16INK4a immunostaining provides independent confirmation of proliferative arrest. Combined nutrient deprivation and genotoxic challenge elicited the most pronounced and concordant organelle alterations relative to single stressors, aligning with age-donor differences. Our approach integrates structural and functional readouts across modalities, reducing the impact of phenotypic heterogeneity and providing reproducible multiparametric endpoints. Although the framework focuses on a robustly validated panel of phenotypes, it is extensible by nature and sensitive to distributional shifts. This allows both drug-specific redistribution of established markers and the emergence of atypical or transient phenotypes to be detected. This flexibility renders the platform suitable for comparative studies and the screening of senolytics and geroprotectors, as well as for refining the evolving landscape of senescence-associated states. Full article

(This article belongs to the Section Image and Video Processing)

► Show Figures

Figure 1

30 pages, 10084 KB

Open AccessArticle

Automatic Visual Inspection for Industrial Application

by António Gouveia Ribeiro, Luís Vilaça, Carlos Costa, Tiago Soares da Costa and Pedro Miguel Carvalho

J. Imaging 2025, 11(10), 350; https://doi.org/10.3390/jimaging11100350 - 8 Oct 2025

Viewed by 808

Abstract

Quality control represents a critical function in industrial environments, ensuring that manufactured products meet strict standards and remain free from defects. In highly regulated sectors such as the pharmaceutical industry, traditional manual inspection methods remain widely used. However, these are time-consuming and prone [...] Read more.

Quality control represents a critical function in industrial environments, ensuring that manufactured products meet strict standards and remain free from defects. In highly regulated sectors such as the pharmaceutical industry, traditional manual inspection methods remain widely used. However, these are time-consuming and prone to human error, and they lack the reliability required for large-scale operations, highlighting the urgent need for automated solutions. This is crucial for industrial applications, where environments evolve and new defect types can arise unpredictably. This work proposes an automated visual defect detection system specifically designed for pharmaceutical bottles, with potential applicability in other manufacturing domains. Various methods were integrated to create robust tools capable of real-world deployment. A key strategy is the use of incremental learning, which enables machine learning models to incorporate new, unseen data without full retraining, thus enabling adaptation to new defects as they appear, allowing models to handle rare cases while maintaining stability and performance. The proposed solution incorporates a multi-view inspection setup to capture images from multiple angles, enhancing accuracy and robustness. Evaluations in real-world industrial conditions demonstrated high defect detection rates, confirming the effectiveness of the proposed approach. Full article

(This article belongs to the Section Computer Vision and Pattern Recognition)

► Show Figures

Figure 1

16 pages, 5738 KB

Open AccessArticle

Image-Processing-Driven Modeling and Reconstruction of Traditional Patterns via Dual-Channel Detection and B-Spline Analysis

by Xuemei He, Siyi Chen, Yin Kuang and Xinyue Yang

J. Imaging 2025, 11(10), 349; https://doi.org/10.3390/jimaging11100349 - 7 Oct 2025

Viewed by 409

Abstract

This study aims to address the research gap in the digital analysis of traditional patterns by proposing an image-processing-driven parametric modeling method that combines graphic primitive function modeling with topological reconstruction. The image is processed using a dual-channel image processing algorithm (Canny edge [...] Read more.

This study aims to address the research gap in the digital analysis of traditional patterns by proposing an image-processing-driven parametric modeling method that combines graphic primitive function modeling with topological reconstruction. The image is processed using a dual-channel image processing algorithm (Canny edge detection and grayscale mapping) to extract and vectorize graphic primitives. These primitives are uniformly represented using B-spline curves, with variations generated through parametric control. A topological reconstruction approach is introduced, incorporating mapped geometric parameters, topological combination rules, and geometric adjustments to output topological configurations. The generated patterns are evaluated using fractal dimension analysis for complexity quantification and applied in cultural heritage imaging practice. The proposed image processing pipeline enables flexible parametric control and continuous structural integration of the graphic primitives and demonstrates high reproducibility and expandability. This study establishes a novel computational framework for traditional patterns, offering a replicable technical pathway that integrates image processing, parametric modeling, and topological reconstruction for digital expression, stylistic innovation, and heritage conservation. Full article

(This article belongs to the Section Computational Imaging and Computational Photography)

► Show Figures

Figure 1

23 pages, 5437 KB

Open AccessArticle

Hierarchical Deep Learning for Abnormality Classification in Mouse Skeleton Using Multiview X-Ray Images: Convolutional Autoencoders Versus ConvNeXt

by Muhammad M. Jawaid, Rasneer S. Bains, Sara Wells and James M. Brown

J. Imaging 2025, 11(10), 348; https://doi.org/10.3390/jimaging11100348 - 7 Oct 2025

Viewed by 394

Abstract

Single-view-based anomaly detection approaches present challenges due to the lack of context, particularly for multi-label problems. In this work, we demonstrate the efficacy of using multiview image data for improved classification using a hierarchical learning approach. Using 170,958 images from the International Mouse [...] Read more.

Single-view-based anomaly detection approaches present challenges due to the lack of context, particularly for multi-label problems. In this work, we demonstrate the efficacy of using multiview image data for improved classification using a hierarchical learning approach. Using 170,958 images from the International Mouse Phenotyping Consortium (IMPC) repository, a specimen-wise multiview dataset comprising 54,046 specimens was curated. Next, two hierarchical classification frameworks were developed by customizing ConvNeXT and a convolutional autoencoder (CAE) as CNN backbones, respectively. The customized architectures were trained at three hierarchy levels with increasing anatomical granularity, enabling specialized layers to learn progressively more detailed features. At the top level (L1), multiview (MV) classification performed about the same as single views, with a high mean AUC of 0.95. However, using MV images in the hierarchical model greatly improved classification at levels 2 and 3. The model showed consistently higher average AUC scores with MV compared to single views such as dorsoventral or lateral. For example, at Level 2 (L2), the model divided abnormal cases into three subclasses, achieving AUCs of 0.65 for DV, 0.76 for LV, and 0.87 for MV. Then, at Level 3 (L3), it further divided these into ten specific abnormalities, with AUCs of 0.54 for DV, 0.59 for LV, and 0.82 for MV. A similar performance was achieved by the CAE-driven architecture, with mean AUCs of 0.87, 0.88, and 0.89 at Level 2 (L2) and 0.74, 0.78, and 0.81 at Level 3 (L3), respectively, for DV, LV, and MV views. The overall results demonstrate the advantage of multiview image data coupled with hierarchical learning for skeletal abnormality detection in a multi-label context. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

12 pages, 9239 KB

Open AccessArticle

Effects of Motion in Ultrashort Echo Time Quantitative Susceptibility Mapping for Musculoskeletal Imaging

by Sam Sedaghat, Jinil Park, Eddie Fu, Fang Liu, Youngkyoo Jung and Hyungseok Jang

J. Imaging 2025, 11(10), 347; https://doi.org/10.3390/jimaging11100347 - 6 Oct 2025

Viewed by 507

Abstract

Quantitative susceptibility mapping (QSM) is a powerful magnetic resonance imaging (MRI) technique for assessing tissue composition in the human body. For imaging short-T2 tissues in the musculoskeletal (MSK) system, ultrashort echo time (UTE) imaging plays a key role. However, UTE-based QSM (UTE-QSM) often [...] Read more.

Quantitative susceptibility mapping (QSM) is a powerful magnetic resonance imaging (MRI) technique for assessing tissue composition in the human body. For imaging short-T2 tissues in the musculoskeletal (MSK) system, ultrashort echo time (UTE) imaging plays a key role. However, UTE-based QSM (UTE-QSM) often involves repeated acquisitions, making it vulnerable to inter-scan motion. In this study, we investigate the effects of motion on UTE-QSM and introduce strategies to reduce motion-induced artifacts. Eight healthy male volunteers underwent UTE-QSM imaging of the knee joint, while an additional seven participated in imaging of the ankle joint. UTE-QSM was conducted using multiple echo acquisitions, including both UTE and gradient-recalled echoes, and processed using the iterative decomposition of water and fat with echo asymmetry and least-squares estimation (IDEAL) and morphology-enabled dipole inversion (MEDI) algorithms. To assess the impact of motion, datasets were reconstructed both with and without motion correction. Furthermore, we evaluated a two-step UTE-QSM approach that incorporates tissue boundary information. This method applies edge detection, excludes pixels near detected edges, and performs a two-step QSM reconstruction to reduce motion-induced streaking artifacts. In participants exhibiting substantial inter-scan motion, prominent streaking artifacts were evident. Applying motion registration markedly reduced these artifacts in both knee and ankle UTE-QSM. Additionally, the two-step UTE-QSM approach, which integrates tissue boundary information, further enhanced image quality by mitigating residual streaking artifacts. These results indicate that motion-induced errors near tissue boundaries play a key role in generating streaking artifacts in UTE-QSM. Inter-scan motion poses a fundamental challenge in UTE-QSM due to the need for multiple acquisitions. However, applying motion registration along with a two-step QSM approach that excludes tissue boundaries can effectively suppress motion-induced streaking artifacts, thereby improving the accuracy of musculoskeletal tissue characterization. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

15 pages, 2364 KB

Open AccessArticle

Optimized Lung Nodule Classification Using CLAHE-Enhanced CT Imaging and Swin Transformer-Based Deep Feature Extraction

by Dorsaf Hrizi, Khaoula Tbarki and Sadok Elasmi

J. Imaging 2025, 11(10), 346; https://doi.org/10.3390/jimaging11100346 - 4 Oct 2025

Viewed by 377

Abstract

Lung cancer remains one of the most lethal cancers globally. Its early detection is vital to improving survival rates. In this work, we propose a hybrid computer-aided diagnosis (CAD) pipeline for lung cancer classification using Computed Tomography (CT) scan images. The proposed CAD [...] Read more.

Lung cancer remains one of the most lethal cancers globally. Its early detection is vital to improving survival rates. In this work, we propose a hybrid computer-aided diagnosis (CAD) pipeline for lung cancer classification using Computed Tomography (CT) scan images. The proposed CAD pipeline integrates ten image preprocessing techniques and ten pretrained deep learning models for feature extraction including convolutional neural networks and transformer-based architectures, and four classical machine learning classifiers. Unlike traditional end-to-end deep learning systems, our approach decouples feature extraction from classification, enhancing interpretability and reducing the risk of overfitting. A total of 400 model configurations were evaluated to identify the optimal combination. The proposed approach was evaluated on the publicly available Lung Image Database Consortium and Image Database Resource Initiative dataset, which comprises 1018 thoracic CT scans annotated by four thoracic radiologists. For the classification task, the dataset included a total of 6568 images labeled as malignant and 4849 images labeled as benign. Experimental results show that the best performing pipeline, combining Contrast Limited Adaptive Histogram Equalization, Swin Transformer feature extraction, and eXtreme Gradient Boosting, achieved an accuracy of 95.8%. Full article

(This article belongs to the Special Issue Advancements in Imaging Techniques for Detection of Cancer)

► Show Figures

Figure 1

24 pages, 14242 KB

Open AccessArticle

DBA-YOLO: A Dense Target Detection Model Based on Lightweight Neural Networks

by Zhiyong He, Jiahong Yang, Hongtian Ning, Chengxuan Li and Qiang Tang

J. Imaging 2025, 11(10), 345; https://doi.org/10.3390/jimaging11100345 - 4 Oct 2025

Viewed by 683

Abstract

Current deep learning-based dense target detection models face dual challenges in industrial scenarios: high computational complexity leading to insufficient inference efficiency on mobile devices, and missed/false detections caused by dense small targets, high inter-class similarity, and complex background interference. To address these issues, [...] Read more.

Current deep learning-based dense target detection models face dual challenges in industrial scenarios: high computational complexity leading to insufficient inference efficiency on mobile devices, and missed/false detections caused by dense small targets, high inter-class similarity, and complex background interference. To address these issues, this paper proposes DBA-YOLO, a lightweight model based on YOLOv10, which significantly reduces computational complexity through model compression and algorithm optimization while maintaining high accuracy. Key improvements include the following: (1) a C2f PA module for enhanced feature extraction, (2) a parameter-refined BIMAFPN neck structure to improve small target detection, and (3) a DyDHead module integrating scale, space, and task awareness for spatial feature weighting. To validate DBA-YOLO, we constructed a real-world dataset from cigarette package images. Experiments on SKU-110K and our dataset show that DBA-YOLO achieves 91.3% detection accuracy (1.4% higher than baseline), with mAP and mAP75 improvements of 2–3%. Additionally, the model reduces parameters by 3.6%, balancing efficiency and performance for resource-constrained devices. Full article

(This article belongs to the Section Computer Vision and Pattern Recognition)

► Show Figures

Figure 1

36 pages, 462 KB

Open AccessArticle

No Reproducibility, No Progress: Rethinking CT Benchmarking

by Dmitry Polevoy, Danil Kazimirov, Marat Gilmanov and Dmitry Nikolaev

J. Imaging 2025, 11(10), 344; https://doi.org/10.3390/jimaging11100344 - 2 Oct 2025

Viewed by 550

Abstract

Reproducibility is a cornerstone of scientific progress, yet in X-ray computed tomography (CT) reconstruction, it remains a critical and unresolved challenge. Current benchmarking practices in CT are hampered by the scarcity of openly available datasets, the incomplete or task-specific nature of existing resources, [...] Read more.

Reproducibility is a cornerstone of scientific progress, yet in X-ray computed tomography (CT) reconstruction, it remains a critical and unresolved challenge. Current benchmarking practices in CT are hampered by the scarcity of openly available datasets, the incomplete or task-specific nature of existing resources, and the lack of transparent implementations of widely used methods and evaluation metrics. As a result, even the fundamental property of reproducibility is frequently violated, undermining objective comparison and slowing methodological progress. In this work, we analyze the systemic limitations of current CT benchmarking, drawing parallels with broader reproducibility issues across scientific domains. We propose an extended data model and formalized schemes for data preparation and quality assessment, designed to improve reproducibility and broaden the applicability of CT datasets across multiple tasks. Building on these schemes, we introduce checklists for dataset construction and quality assessment, offering a foundation for reliable and reproducible benchmarking pipelines. A key aspect of our recommendations is the integration of virtual CT (vCT), which provides highly realistic data and analytically computable phantoms, yet remains underutilized despite its potential to overcome many current barriers. Our work represents a first step toward a methodological framework for reproducible benchmarking in CT. This framework aims to enable transparent, rigorous, and comparable evaluation of reconstruction methods, ultimately supporting their reliable adoption in clinical and industrial applications. Full article

(This article belongs to the Special Issue Tools and Techniques for Improving Radiological Imaging Applications)

► Show Figures

Figure 1

22 pages, 4682 KB

Open AccessArticle

Development of a Fully Optimized Convolutional Neural Network for Astrocytoma Classification in MRI Using Explainable Artificial Intelligence

by Christos Ch. Andrianos, Spiros A. Kostopoulos, Ioannis K. Kalatzis, Dimitris Th. Glotsos, Pantelis A. Asvestas, Dionisis A. Cavouras and Emmanouil I. Athanasiadis

J. Imaging 2025, 11(10), 343; https://doi.org/10.3390/jimaging11100343 - 2 Oct 2025

Viewed by 427

Abstract

Astrocytoma is the most common type of brain glioma and is classified by the World Health Organization into four grades, providing prognostic insights and guiding treatment decisions. The accurate determination of astrocytoma grade is critical for patient management, especially in high-malignancy-grade cases. This [...] Read more.

Astrocytoma is the most common type of brain glioma and is classified by the World Health Organization into four grades, providing prognostic insights and guiding treatment decisions. The accurate determination of astrocytoma grade is critical for patient management, especially in high-malignancy-grade cases. This study proposes a fully optimized Convolutional Neural Network (CNN) for the classification of astrocytoma MRI slices across the three malignant grades (G2–4). The training dataset consisted of 1284 pre-operative axial 2D MRI slices from T1-weighted contrast-enhanced and FLAIR sequences derived from 69 patients. To provide the best possible model performance, an extensive hyperparameter tuning was carried out through the Hyperband method, a variant of Successive Halving. Training was conducted using Repeated Hold-Out Validation across four randomized data splits, achieving a mean classification accuracy of 98.05%, low loss values, and an AUC of 0.997. Comparative evaluation against state-of-the-art pre-trained models using transfer learning demonstrated superior performance. For validation purposes, the proposed CNN trained on an altered version of the training set yielded 93.34% accuracy on unmodified slices, which confirms the model’s robustness and potential use for clinical deployment. Model interpretability was ensured through the application of two Explainable AI (XAI) techniques, SHAP and LIME, which highlighted the regions of the slices contributing to the decision-making process. Full article

(This article belongs to the Section Medical Imaging)

► Show Figures

Figure 1

15 pages, 2112 KB

Open AccessArticle

Radiomics-Based Preoperative Assessment of Muscle-Invasive Bladder Cancer Using Combined T2 and ADC MRI: A Multicohort Validation Study

by Dmitry Kabanov, Natalia Rubtsova, Aleksandra Golbits, Andrey Kaprin, Valentin Sinitsyn and Mikhail Potievskiy

J. Imaging 2025, 11(10), 342; https://doi.org/10.3390/jimaging11100342 - 1 Oct 2025

Viewed by 496

Abstract

Accurate preoperative staging of bladder cancer on MRI remains challenging because visual reads vary across observers. We investigated a multiparametric MRI (mpMRI) radiomics approach to predict muscle invasion (≥T2) and prospectively tested it on a validation cohort. Eighty-four patients with urothelial carcinoma underwent [...] Read more.

Accurate preoperative staging of bladder cancer on MRI remains challenging because visual reads vary across observers. We investigated a multiparametric MRI (mpMRI) radiomics approach to predict muscle invasion (≥T2) and prospectively tested it on a validation cohort. Eighty-four patients with urothelial carcinoma underwent 1.5-T mpMRI per VI-RADS (T2-weighted imaging and DWI-derived ADC maps). Two blinded radiologists performed 3D tumor segmentation; 37 features per sequence were extracted (LifeX) using absolute resampling. In the training cohort (n = 40), features that differed between non-muscle-invasive and muscle-invasive tumors (Mann–Whitney p < 0.05) underwent ROC analysis with cut-offs defined by the Youden index. A compact descriptor combining GLRLM-LRLGE from T2 and GLRLM-SRLGE from ADC was then fixed and applied without re-selection to a prospective validation cohort (n = 44). Histopathology within 6 weeks—TURBT or cystectomy—served as the reference. Eleven T2-based and fifteen ADC-based features pointed to invasion; DWI texture features were not informative. The descriptor yielded AUCs of 0.934 (training) and 0.871 (validation) with 85.7% sensitivity and 96.2% specificity in validation. Collectively, these findings indicate that combined T2/ADC radiomics can provide high diagnostic accuracy and may serve as a useful decision support tool, after multicenter, multi-vendor validation. Full article

(This article belongs to the Topic Machine Learning and Deep Learning in Medical Imaging)

► Show Figures

Figure 1

Journal Menu

Journal Browser

J. Imaging, Volume 11, Issue 10 (October 2025) – 47 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI