Journal of Imaging最新文献_第2页

Facial Recognition Algorithms: A Systematic Literature Review.

IF 2.7

Journal of Imaging Pub Date : 2025-02-13 DOI: 10.3390/jimaging11020058

Nazar El Fadel

{"title":"Facial Recognition Algorithms: A Systematic Literature Review.","authors":"Nazar El Fadel","doi":"10.3390/jimaging11020058","DOIUrl":"10.3390/jimaging11020058","url":null,"abstract":"This systematic literature review aims to understand new developments and challenges in facial recognition technology. This will provide an understanding of the system principles, performance metrics, and applications of facial recognition technology in various fields such as health, society, and security from various academic publications, conferences, and industry news. A comprehensive approach was adopted in the literature review of various facial recognition technologies. It emphasizes the most important techniques in algorithm development, examines performance metrics, and explores their applications in various fields. The review mainly emphasizes the recent development in deep learning techniques, especially CNNs, which greatly improved the accuracy and efficiency of facial recognition systems. The findings reveal that there has been a noticeable evolution in facial recognition technology, especially with the current use of deep learning techniques. Nevertheless, it highlights important challenges, including privacy concerns, ethical dilemmas, and biases in the systems. These factors highlight the necessity of using facial recognition technology in an ethical and regulated manner. In conclusion, the paper proposes several future research directions to establish the reliability of facial recognition systems and reduce biases while building user confidence. These considerations are key to responsibly advancing facial recognition technology by ensuring ethical practices and safeguarding privacy.","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"11 2","pages":""},"PeriodicalIF":2.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11856072/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Adapting Classification Neural Network Architectures for Medical Image Segmentation Using Explainable AI. 利用可解释的人工智能，为医学图像分割调整分类神经网络架构。

IF 2.7

Journal of Imaging Pub Date : 2025-02-13 DOI: 10.3390/jimaging11020055

Arturs Nikulins, Edgars Edelmers, Kaspars Sudars, Inese Polaka

{"title":"Adapting Classification Neural Network Architectures for Medical Image Segmentation Using Explainable AI.","authors":"Arturs Nikulins, Edgars Edelmers, Kaspars Sudars, Inese Polaka","doi":"10.3390/jimaging11020055","DOIUrl":"10.3390/jimaging11020055","url":null,"abstract":"Segmentation neural networks are widely used in medical imaging to identify anomalies that may impact patient health. Despite their effectiveness, these networks face significant challenges, including the need for extensive annotated patient data, time-consuming manual segmentation processes and restricted data access due to privacy concerns. In contrast, classification neural networks, similar to segmentation neural networks, capture essential parameters for identifying objects during training. This paper leverages this characteristic, combined with explainable artificial intelligence (XAI) techniques, to address the challenges of segmentation. By adapting classification neural networks for segmentation tasks, the proposed approach reduces dependency on manual segmentation. To demonstrate this concept, the Medical Segmentation Decathlon 'Brain Tumours' dataset was utilised. A ResNet classification neural network was trained, and XAI tools were applied to generate segmentation-like outputs. Our findings reveal that GuidedBackprop is among the most efficient and effective methods, producing heatmaps that closely resemble segmentation masks by accurately highlighting the entirety of the target object.","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"11 2","pages":""},"PeriodicalIF":2.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11856553/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multiphoton Microscopy to Visualize Live Renal Nerves in Reanimated Kidney Blocks.

IF 2.7

Journal of Imaging Pub Date : 2025-02-13 DOI: 10.3390/jimaging11020056

Joerg Reifart, Patrick T Willey, Paul A Iaizzo

{"title":"Multiphoton Microscopy to Visualize Live Renal Nerves in Reanimated Kidney Blocks.","authors":"Joerg Reifart, Patrick T Willey, Paul A Iaizzo","doi":"10.3390/jimaging11020056","DOIUrl":"10.3390/jimaging11020056","url":null,"abstract":"Renal denervation to treat arterial hypertension is growing in adoption but still shows inconsistent results. Device improvement is difficult, as there is currently no way to study the immediate success of renal denervation devices in living tissue. In an effort to visualize live renal nerves surrounding their arteries using multiphoton microscopy, kidney pairs were explanted from Yorkshire pigs. They were maintained viable with a pulsatile perfusion apparatus using Visible Kidney™ methodologies, in which blood is replaced by a modified, oxygenated, and warmed (37 °C) Krebs-Henseleit buffer. The block resection allows catheter placement for nerve ablation treatment. Subsequently, the kidney block was disconnected from the perfusion system and underwent multiphoton microscopy (Nikon A1R 1024 MP). A total of three renal blocks were imaged using this model. Using 780 nm excitation for autofluorescence, we were able to selectively image peri-arterial nerves (2.5-23 μm diameter) alongside arteriolar elastin fibers (1.96 ± 0.87 μm; range: 0.3-4.27) at 25× magnification at a pixel size of 1.02 µm). Autofluoresecence was not strong enough to identify nerves at 4× magnification. There was a high but variable signal-to-noise ratio of 52.3 (median, IQR 159). This model may be useful for improving future physician training and innovations in renal denervation technologies.","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"11 2","pages":""},"PeriodicalIF":2.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11856253/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Novel 3D Magnetic Resonance Imaging Registration Framework Based on the Swin-Transformer UNet+ Model with 3D Dynamic Snake Convolution Scheme.

IF 2.7

Journal of Imaging Pub Date : 2025-02-11 DOI: 10.3390/jimaging11020054

Yaolong Han, Lei Wang, Zizhen Huang, Yukun Zhang, Xiao Zheng

{"title":"A Novel 3D Magnetic Resonance Imaging Registration Framework Based on the Swin-Transformer UNet+ Model with 3D Dynamic Snake Convolution Scheme.","authors":"Yaolong Han, Lei Wang, Zizhen Huang, Yukun Zhang, Xiao Zheng","doi":"10.3390/jimaging11020054","DOIUrl":"10.3390/jimaging11020054","url":null,"abstract":"Transformer-based image registration methods have achieved notable success, but they still face challenges, such as difficulties in representing both global and local features, the inability of standard convolution operations to focus on key regions, and inefficiencies in restoring global context using the decoder. To address these issues, we extended the Swin-UNet architecture and incorporated dynamic snake convolution (DSConv) into the model, expanding it into three dimensions. This improvement enables the model to better capture spatial information at different scales, enhancing its adaptability to complex anatomical structures and their intricate components. Additionally, multi-scale dense skip connections were introduced to mitigate the spatial information loss caused by downsampling, enhancing the model's ability to capture both global and local features. We also introduced a novel optimization-based weakly supervised strategy, which iteratively refines the deformation field generated during registration, enabling the model to produce more accurate registered images. Building on these innovations, we proposed OSS DSC-STUNet+ (Swin-UNet+ with 3D dynamic snake convolution). Experimental results on the IXI, OASIS, and LPBA40 brain MRI datasets demonstrated up to a 16.3% improvement in Dice coefficient compared to five classical methods. The model exhibits outstanding performance in terms of registration accuracy, efficiency, and feature preservation.","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"11 2","pages":""},"PeriodicalIF":2.7,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11856140/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Comparative Analysis of Deep Learning Architectures for Macular Hole Segmentation in OCT Images: A Performance Evaluation of U-Net Variants.

IF 2.7

Journal of Imaging Pub Date : 2025-02-11 DOI: 10.3390/jimaging11020053

H M S S Herath, S L P Yasakethu, Nuwan Madusanka, Myunggi Yi, Byeong-Il Lee

{"title":"Comparative Analysis of Deep Learning Architectures for Macular Hole Segmentation in OCT Images: A Performance Evaluation of U-Net Variants.","authors":"H M S S Herath, S L P Yasakethu, Nuwan Madusanka, Myunggi Yi, Byeong-Il Lee","doi":"10.3390/jimaging11020053","DOIUrl":"10.3390/jimaging11020053","url":null,"abstract":"This study presents a comprehensive comparison of U-Net variants with different backbone architectures for Macular Hole (MH) segmentation in optical coherence tomography (OCT) images. We evaluated eleven architectures, including U-Net combined with InceptionNetV4, VGG16, VGG19, ResNet152, DenseNet121, EfficientNet-B7, MobileNetV2, Xception, and Transformer. Models were assessed using the Dice coefficient and HD95 metrics on the OIMHS dataset. While HD95 proved unreliable for small regions like MH, often returning 'nan' values, the Dice coefficient provided consistent performance evaluation. InceptionNetV4 + U-Net achieved the highest Dice coefficient (0.9672), demonstrating superior segmentation accuracy. Although considered state-of-the-art, Transformer + U-Net showed poor performance in MH and intraretinal cyst (IRC) segmentation. Analysis of computational resources revealed that MobileNetV2 + U-Net offered the most efficient performance with minimal parameters, while InceptionNetV4 + U-Net balanced accuracy with moderate computational demands. Our findings suggest that CNN-based backbones, particularly InceptionNetV4, are more effective than Transformer architectures for OCT image segmentation, with InceptionNetV4 + U-Net emerging as the most promising model for clinical applications.","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"11 2","pages":""},"PeriodicalIF":2.7,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11856554/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Highly Robust Encoder-Decoder Network with Multi-Scale Feature Enhancement and Attention Gate for the Reduction of Mixed Gaussian and Salt-and-Pepper Noise in Digital Images. 具有多尺度特征增强和注意门的高鲁棒性编码器-解码器网络，用于降低数字图像中的混合高斯和椒盐噪声

IF 2.7

Journal of Imaging Pub Date : 2025-02-10 DOI: 10.3390/jimaging11020051

Milan Tripathi, Waree Kongprawechnon, Toshiaki Kondo

{"title":"A Highly Robust Encoder-Decoder Network with Multi-Scale Feature Enhancement and Attention Gate for the Reduction of Mixed Gaussian and Salt-and-Pepper Noise in Digital Images.","authors":"Milan Tripathi, Waree Kongprawechnon, Toshiaki Kondo","doi":"10.3390/jimaging11020051","DOIUrl":"10.3390/jimaging11020051","url":null,"abstract":"Image denoising is crucial for correcting distortions caused by environmental factors and technical limitations. We propose a novel and highly robust encoder-decoder network (HREDN) for effectively removing mixed salt-and-pepper and Gaussian noise from digital images. HREDN integrates a multi-scale feature enhancement block in the encoder, allowing the network to capture features at various scales and handle complex noise patterns more effectively. To mitigate information loss during encoding, skip connections transfer essential feature maps from the encoder to the decoder, preserving structural details. However, skip connections can also propagate redundant information. To address this, we incorporate attention gates within the skip connections, ensuring that only relevant features are passed to the decoding layers. We evaluate the robustness of the proposed method across facial, medical, and remote sensing domains. The experimental results demonstrate that HREDN excels in preserving edge details and structural features in denoised images, outperforming state-of-the-art techniques in both qualitative and quantitative measures. Statistical analysis further highlights the model's ability to effectively remove noise in diverse, complex scenarios with images of varying resolutions across multiple domains.","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"11 2","pages":""},"PeriodicalIF":2.7,"publicationDate":"2025-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11856137/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SAVE: Self-Attention on Visual Embedding for Zero-Shot Generic Object Counting.

IF 2.7

Journal of Imaging Pub Date : 2025-02-10 DOI: 10.3390/jimaging11020052

Ahmed Zgaren, Wassim Bouachir, Nizar Bouguila

{"title":"SAVE: Self-Attention on Visual Embedding for Zero-Shot Generic Object Counting.","authors":"Ahmed Zgaren, Wassim Bouachir, Nizar Bouguila","doi":"10.3390/jimaging11020052","DOIUrl":"10.3390/jimaging11020052","url":null,"abstract":"Zero-shot counting is a subcategory of Generic Visual Object Counting, which aims to count objects from an arbitrary class in a given image. While few-shot counting relies on delivering exemplars to the model to count similar class objects, zero-shot counting automates the operation for faster processing. This paper proposes a fully automated zero-shot method outperforming both zero-shot and few-shot methods. By exploiting feature maps from a pre-trained detection-based backbone, we introduce a new Visual Embedding Module designed to generate semantic embeddings within object contextual information. These embeddings are then fed to a Self-Attention Matching Module to generate an encoded representation for the head counter. Our proposed method has outperformed recent zero-shot approaches, achieving the best Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) results of 8.89 and 35.83, respectively, on the FSC147 dataset. Additionally, our method demonstrates competitive performance compared to few-shot methods, advancing the capabilities of visual object counting in various industrial applications such as tree counting, wildlife animal counting, and medical applications like blood cell counting.","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"11 2","pages":""},"PeriodicalIF":2.7,"publicationDate":"2025-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11856060/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143494010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Differentiation of Atypical Lipomatous Tumors from Lipomas: Our Experience with Visual Analysis of Conventional Magnetic Resonance Imaging. 非典型脂肪瘤与脂肪瘤的鉴别：我们对传统磁共振成像进行视觉分析的经验。

IF 2.7

Journal of Imaging Pub Date : 2025-02-08 DOI: 10.3390/jimaging11020047

Luz Maria Moran, Chao Yuan Li Cai, Alberto Ramirez, Ana Royuela

{"title":"Differentiation of Atypical Lipomatous Tumors from Lipomas: Our Experience with Visual Analysis of Conventional Magnetic Resonance Imaging.","authors":"Luz Maria Moran, Chao Yuan Li Cai, Alberto Ramirez, Ana Royuela","doi":"10.3390/jimaging11020047","DOIUrl":"10.3390/jimaging11020047","url":null,"abstract":"Differentiating atypical lipomatous tumors (ALTs) from lipomas using imaging techniques is a challenge, and the biopsy with immunohistochemical determination of murine double minute 2 (MDM2) oncogene is the gold standard. We are looking for a management algorithm with the visual analysis of magnetic resonance images in these two fatty soft tissue tumors that allow us to avoid some biopsies. Two radiologists, blinded to the final diagnosis, independently assessed various features on conventional magnetic resonance imaging (MRI), in 79 patients with pathologically confirmed fatty tumors as either lipoma (MDM2 negative) or ALT (MDM2 positive). Results: The interobserver agreement for the most MRI features was moderate and the musculoskeletal radiologist accuracy for final diagnosis was 90% sensitivity and 66% specificity. Tumors with homogeneous fat signals and a maximum size < 8 cm were always lipomas (p < 0.001), and the tumors with septa thickness ≥ 2 mm, or more than one non-fat nodule, and a maximum size ≥ 12.8 cm were typically ALTs. While those tumors with septa < 2 mm or one non-fat nodule, independently of maximum size, the diagnosis of lipoma versus ALT is uncertain and a biopsy is required.","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"11 2","pages":""},"PeriodicalIF":2.7,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11856569/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing U-Net Segmentation Accuracy Through Comprehensive Data Preprocessing.

IF 2.7

Journal of Imaging Pub Date : 2025-02-08 DOI: 10.3390/jimaging11020050

Talshyn Sarsembayeva, Madina Mansurova, Assel Abdildayeva, Stepan Serebryakov

引用次数: 0

Resampling Point Clouds Using Series of Local Triangulations.

IF 2.7

Journal of Imaging Pub Date : 2025-02-08 DOI: 10.3390/jimaging11020049

Vijai Kumar Suriyababu, Cornelis Vuik, Matthias Möller

引用次数: 0