Search Results (466)

Background/Objectives: Melanoma is an aggressive type of skin cancer that poses serious health risks if not detected in its early stages. Although early diagnosis enables effective treatment, delays can result in life-threatening consequences. Traditional diagnostic processes predominantly rely on the subjective expertise of dermatologists, which can lead to variability and time inefficiencies. Consequently, there is an increasing demand for automated systems that can accurately classify melanoma lesions and retrieve visually similar cases to support clinical decision-making. Methods: This study proposes a transfer learning (TL)-based deep learning (DL) framework for the classification of melanoma images and the enhancement of content-based image retrieval (CBIR) systems. Pre-trained models including DenseNet121, InceptionV3, Vision Transformer (ViT), and Xception were employed to extract deep feature representations. These features were integrated using a weighted fusion strategy and classified through an Ensemble learning approach designed to capitalize on the complementary strengths of the individual models. The performance of the proposed system was evaluated using classification accuracy and mean Average Precision (mAP) metrics. Results: Experimental evaluations demonstrated that the proposed Ensemble model significantly outperformed each standalone model in both classification and retrieval tasks. The Ensemble approach achieved a classification accuracy of 95.25%. In the CBIR task, the system attained a mean Average Precision (mAP) score of 0.9538, indicating high retrieval effectiveness. The performance gains were attributed to the synergistic integration of features from diverse model architectures through the ensemble and fusion strategies. Conclusions: The findings underscore the effectiveness of TL-based DL models in automating melanoma image classification and enhancing CBIR systems. The integration of deep features from multiple pre-trained models using an Ensemble approach not only improved accuracy but also demonstrated robustness in feature generalization. This approach holds promise for integration into clinical workflows, offering improved diagnostic accuracy and efficiency in the early detection of melanoma. Full article

Background: The early detection of melanoma is crucial for improving patient outcomes, as survival rates decline dramatically with disease progression. Despite significant achievements in deep learning methods for skin lesion analysis, several challenges limit their effectiveness in clinical practice. One of the key issues is the lack of knowledge about the melanoma stage distribution in the training data, raising concerns about the ability of these models to detect early-stage melanoma accurately. Additionally, publicly available datasets that include detailed information on melanoma stage and tumor thickness remain scarce, restricting researchers from developing and benchmarking methods specifically tailored for early diagnosis. Another major limitation is the lack of cross-dataset evaluations. Most deep learning models are tested on the same dataset they were trained on, so they fail to assess their generalization ability when applied to unseen data. This reduces their reliability in real-world clinical settings. Methods: We introduce an early-stage melanoma benchmark dataset to address these issues, featuring images labeled according to T-category based on Breslow thickness. Results: We evaluated several state-of-the-art deep learning models on this dataset and observed a significant drop in performance compared to their results on the ISIC Challenge datasets. Conclusions: This finding highlights the models’ limited capability in detecting early-stage melanoma. This work seeks to advance the development and clinical applicability of automated melanoma diagnostic systems by providing a resource for T-category-specific analysis and supporting cross-dataset evaluation. Full article

We performed a diagnostic disease investigation on a wild smallmouth bass (Micropterus dolomieu) with skin ulcers that was collected from Lake Oahe, South Dakota, following reports from anglers of multiple fish with similar lesions. Gross and histologic lesions of ulcerative dermatitis, myositis, and lymphocytolysis within the spleen and kidneys were consistent with largemouth bass virus (LMBV) infection. LMBV was detected by conventional PCR in samples of a skin ulcer, and the complete genome sequence of the LMBV (99,184 bp) was determined from a virus isolate obtained from a homogenized skin sample. A maximum likelihood (ML) phylogenetic analysis based on the major capsid protein (MCP) gene alignment supported the LMBV isolate (LMBV-SD-2023) as a member of the species Ranavirus micropterus1, branching within the subclade of LMBV isolates recovered from North American largemouth (Micropterus salmoides) and smallmouth bass. This is the first detection of LMBV in wild smallmouth bass from South Dakota. The ultrastructure of the LMBV isolate exhibited the expected icosahedral shape of virions budding from cellular membranes. Viral nucleic acid in infected cells was visualized via in situ hybridization (ISH) within dermal granulomas, localized predominantly at the margin of epithelioid macrophages and central necrosis. Further sampling is needed to determine the geographic distribution, affected populations, and evolutionary relationship between isolates of LMBV. Full article

Introduction: Acute Myeloid Leukemia (AML) is a hematologic malignancy characterized by the clonal expansion of myeloid progenitors. While it primarily affects the bone marrow, extramedullary relapse occurs in 3–5% of cases, and it is linked to poor prognosis. Central nervous system (CNS) involvement presents diagnostic challenges due to nonspecific symptoms. CNS manifestations include leptomeningeal dissemination, nerve infiltration, parenchymal lesions, and myeloid sarcoma, occurring at any disease stage and frequently asymptomatic. Methods: A 62-year-old man with a recent history of AML in remission presented with diplopia and aching paresthesias in the left periorbital region spreading to the left frontal area. The diagnostic workup included neurological and hematological evaluation, lumbar puncture, brain CT, brain magnetic resonance imaging (MRI) with contrast, and dermatological evaluation with skin biopsy due to the appearance of nodular skin lesions on the abdomen and thorax. Results: Neurological evaluation showed hypoesthesia in the left mandibular region, consistent with left trigeminal nerve involvement, extending to the periorbital and frontal areas, and impaired adduction of the left eye with divergent strabismus in the primary position due to left oculomotor nerve palsy. Brain MRI showed an equivocal thickening of the left oculomotor nerve without enhancement. Cerebrospinal fluid (CSF) analysis initially showed elevated protein (47 mg/dL) with negative cytology; a repeat lumbar puncture one week later detected leukemic cells. Skin biopsy revealed cutaneous AML localization. A diagnosis of AML relapse with CNS and cutaneous localization was made. Salvage therapy with FLAG-IDA-VEN (fludarabine, cytarabine, idarubicin, venetoclax) and intrathecal methotrexate, cytarabine, and dexamethasone was started. Subsequent lumbar punctures were negative for leukemic cells. Due to high-risk status and extramedullary disease, the patient underwent allogeneic hematopoietic stem cell transplantation. Post-transplant aplasia was complicated by septic shock; the patient succumbed to an invasive fungal infection. Conclusions: This case illustrates the diagnostic complexity and poor prognosis of extramedullary AML relapse involving the CNS. Early recognition of neurological signs, including cranial nerve dysfunction, is crucial for timely diagnosis and management. Although initial investigations were negative, further analyses—including repeated CSF examinations and skin biopsy—led to the identification of leukemic involvement. Although neuroleukemiosis cannot be confirmed without nerve biopsy, the combination of clinical presentation, neuroimaging, and CSF data strongly supports the diagnosis of extramedullary relapse of AML. Multidisciplinary evaluation remains essential for detecting extramedullary relapse. Despite treatment achieving CSF clearance, the prognosis remains unfavorable, underscoring the need for vigilant clinical suspicion in hematologic patients presenting with neurological symptoms. Full article

A 10-year-old intact female Bichon Frise presented with multiple firm skin nodules on all four limbs. The nodules progressively increased in number and size over seven months. Diagnostic tests included cytology of fine-needle aspirates, histopathology of skin biopsies, radiography, and abdominal ultrasonography. Cytology revealed spindle-shaped mesenchymal cells and extracellular matrix components, and histopathology confirmed ND characterized by mature collagen deposition without evidence of malignancy. Ultrasonography detected multiple kidney cysts bilaterally, although their exact nature (benign or malignant) could not be confirmed histologically. Genetic analysis was performed, revealing no mutation in the traditionally implicated FLCN gene but multiple nonsynonymous mutations in the BRCA2 gene. This case suggests a potential association between BRCA2 gene mutations and the development of ND with renal cystic lesions, broadening the known genetic causes beyond the commonly reported FLCN mutation. Regular genetic screening and close monitoring of dermatological and renal conditions in atypical breeds are recommended. To the best of current knowledge, this is the first case report demonstrating ND and renal cysts associated with BRCA2 mutations in a Bichon Frise. Full article

The absence of skin color information in skin cancer datasets poses a significant challenge for accurate diagnosis using artificial intelligence models, particularly for non-white populations. In this paper, based on the Monk Skin Tone (MST) scale, which is less biased than the Fitzpatrick scale, we propose MST-AI, a novel method for detecting skin color in images of large datasets, such as the International Skin Imaging Collaboration (ISIC) archive. The approach includes automatic frame, lesion removal, and lesion segmentation using convolutional neural networks, and modeling normal skin tones with a Variational Bayesian Gaussian Mixture Model (VB-GMM). The distribution of skin color predictions was compared with MST scale probability distribution functions (PDFs) using the Kullback-Leibler Divergence (KLD) metric. Validation against manual annotations and comparison with K-means clustering of image and skin mean RGBs demonstrated the superior performance of the MST-AI, with Kendall’s Tau, Spearman’s Rho, and Normalized Discounted Cumulative Gain (NDGC) of 0.68, 0.69, and 1.00, respectively. This research lays the groundwork for developing unbiased AI models for early skin cancer diagnosis by addressing skin color imbalances in large datasets. Full article

Objective: The classification of skin cancer is very helpful in its early diagnosis and treatment, considering the complexity involved in differentiating AK from BCC and SK. These conditions are generally not easily detectable due to their comparable clinical presentations. Method: This paper presents a new approach to hyperspectral imaging for enhancing the visualization of skin lesions called the Spectrum-Aided Vision Enhancer (SAVE), which has the ability to convert any RGB image into a narrow-band image (NBI) by combining hyperspectral imaging (HSI) to increase the contrast of the area of the cancerous lesions when compared with the normal tissue, thereby increasing the accuracy of classification. The current study investigates the use of ten different machine learning algorithms for the purpose of classification of AK, BCC, and SK, including convolutional neural network (CNN), random forest (RF), you only look once (YOLO) version 8, support vector machine (SVM), ResNet50, MobileNetV2, Logistic Regression, SVM with stochastic gradient descent (SGD) Classifier, SVM with logarithmic (LOG) Classifier and SVM- Polynomial Classifier, in assessing the capability of the system to differentiate AK from BCC and SK with heightened accuracy. Results: The results demonstrated that SAVE enhanced classification performance and increased its accuracy, sensitivity, and specificity compared to a traditional RGB imaging approach. Conclusions: This advanced method offers dermatologists a tool for early and accurate diagnosis, reducing the likelihood of misclassification and improving patient outcomes. Full article

Recently, there has been a rise in skin cancer cases, for which early detection is highly relevant, as it increases the likelihood of a cure. In this context, this work presents a benchmarking study of standard Convolutional Neural Network (CNN) architectures for automated skin lesion classification. A total of 38 CNN architectures from ten families (ConvNeXt, DenseNet, EfficientNet, Inception, InceptionResNet, MobileNet, NASNet, ResNet, VGG, and Xception) were evaluated using transfer learning on the HAM10000 dataset for seven-class skin lesion classification, namely, actinic keratoses, basal cell carcinoma, benign keratosis-like lesions, dermatofibroma, melanoma, melanocytic nevi, and vascular lesions. The comparative analysis used standardized training conditions, with all models utilizing frozen pre-trained weights. Cross-database validation was then conducted using the ISIC 2019 dataset to assess generalizability across different data distributions. The ConvNeXtXLarge architecture achieved the best performance, despite having one of the lowest performance-to-number-of-parameters ratios, with 87.62% overall accuracy and 76.15% F1 score on the test set, demonstrating competitive results within the established performance range of existing HAM10000-based studies. A proof-of-concept multiplatform mobile application was also implemented using a client–server architecture with encrypted image transmission, demonstrating the viability of integrating high-performing models into healthcare screening tools. Full article

Background: Integrating artificial intelligence into clinical photography offers great potential for monitoring skin conditions such as actinic keratosis (AK) and skin field cancerization. Identifying the extent of AK lesions often requires more than analyzing lesion morphology—it also depends on contextual cues, such as surrounding photodamage. This highlights the need for models that can combine fine-grained local features with a comprehensive global view. Methods: To address this challenge, we propose AKTransU-net, a hybrid U-net-based architecture. The model incorporates Transformer blocks to enrich feature representations, which are passed through ConvLSTM modules within the skip connections. This configuration allows the network to maintain semantic coherence and spatial continuity in AK detection. This global awareness is critical when applying the model to whole-image detection via tile-based processing, where continuity across tile boundaries is essential for accurate and reliable lesion segmentation. Results: The effectiveness of AKTransU-net was demonstrated through comparative evaluations with state-of-the-art segmentation models. A proprietary annotated dataset of 569 clinical photographs from 115 patients with actinic keratosis was used to train and evaluate the models. From each photograph, crops of 512 × 512 pixels were extracted using translation lesion boxes that encompassed lesions in different positions and captured different contexts. AKtransU-net exhibited a more robust context awareness and achieved a median Dice score of 65.13%, demonstrating significant progress in whole-image assessments. Conclusions: Transformer-driven context modeling offers a promising approach for robust AK lesion monitoring, supporting its application in real-world clinical settings where accurate, context-aware analysis is crucial for managing skin field cancerization. Full article

Cutaneous neoplasia is among the most common illnesses in dogs and can pose significant risks. Accurate morphological diagnosis of these conditions is vital for effective treatment and management. In this retrospective study, a total of 3746 canine skin biopsies were submitted to a veterinary reference diagnostic laboratory and evaluated using histopathology. The variables assessed included age, sex, breed, lesion, location, and histopathological diagnosis. Non-neoplastic lesions accounted for 61% of all analyzed samples, while neoplastic tumors accounted for 39%. When looking at age, dogs ranging 3–6 years and 7–9 years had at least six times higher risk of developing malignant neoplasia compared to those aged 0–2 years. Among the malignant neoplasms, mast cell tumors, hemangiosarcoma, and squamous cell carcinoma were the most observed, representing 30%, 18%, and 12% of cases, respectively. The breeds most frequently affected by malignant neoplasms included Pit Bull Terriers, Boxers, and mixed breeds, all of which comprised the majority of mast cell tumor cases at 50.54%. These findings are novel in this field and may assist small animal veterinarians in making preliminary diagnoses, while also helping pet owners understand the importance of skin cancer and its early detection. Full article

Background/Objective: The global spread of Monkeypox (Mpox) has highlighted the urgent need for rapid, accurate diagnostic tools. Traditional methods like polymerase chain reaction (PCR) are resource-intensive, while skin image-based detection offers a promising alternative. This study evaluates the effectiveness of vision transformers (ViTs) for automated Mpox detection. Methods: By fine-tuning a pre-trained ViT model on an Mpox lesion image dataset, a robust ViT-based transfer learning (TL) model was created. Performance was assessed relative to convolutional neural network (CNN)-based TL models and ViT models trained from scratch across key metrics: accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC). Furthermore, a transferability measure was utilized to assess the effectiveness of feature transfer to Mpox images. Results: The results show that the ViT model outperformed a CNN, achieving an AUC of 0.948 and an accuracy of 0.942 with a p-value of less than 0.05 across all metrics, highlighting its potential for accurate and scalable Mpox detection. Moreover, the ViT models yielded a better hypothesis margin-based transferability measure, highlighting its effectiveness in transferring useful learning weights to Mpox images. Gradient-weighted Class Activation Mapping (Grad-CAM) visualizations also confirmed that the ViT model attends to clinically relevant features, supporting its interpretability and reliability for diagnostic use. Conclusions: The results from this study suggest that ViT offers superior accuracy, making it a valuable tool for Mpox early detection in field settings, especially where conventional diagnostics are limited. This approach could support faster outbreak response and improved resource allocation in public health systems. Full article

Objective: Skin lesions, including dermatofibroma, lichenoid lesions, and acrochordons, are increasingly prevalent worldwide and often require timely identification for effective clinical management. However, conventional RGB-based imaging can overlook subtle vascular characteristics, potentially delaying diagnosis. Methods: A novel spectrum-aided vision enhancer (SAVE) that transforms standard RGB images into simulated narrowband imaging representations in a single step was proposed. The performances of five cutting-edge object detectors, based on You Look Only Once (YOLOv11, YOLOv10, YOLOv9, YOLOv8, and YOLOv5) models, were assessed across three lesion categories using white-light imaging (WLI) and SAVE modalities. Each YOLO model was trained separately on SAVE and WLI images, and performance was measured using precision, recall, and F1 score. Results: Among all tested configurations, YOLOv10 attained the highest overall performance, particularly under the SAVE modality, demonstrating superior precision and recall across the majority of lesion types. YOLOv9 exhibited robust performance, especially for dermatofibroma detection under SAVE, albeit slightly lagging behind YOLOv10. Conversely, YOLOv11 underperformed on acrochordon detection (cumulative F1  =  65.73%), and YOLOv8 and YOLOv5 displayed lower accuracy and higher false-positive rates, especially in WLI mode. Although SAVE improved the performance of YOLOv8 and YOLOv5, their results remained below those of YOLOv10 and YOLOv9. Conclusions: Combining the SAVE modality with advanced YOLO-based object detectors, specifically YOLOv10 and YOLOv9, markedly enhances the accuracy of lesion detection compared to conventional WLI, facilitating expedited real-time dermatological screening. These findings indicate that integrating snapshot-based narrowband imaging with deep learning object detection models can improve early diagnosis and has potential applications in broader clinical contexts. Full article

SARS-CoV-2, a β-coronavirus, primarily affects the lungs, with non-specific lesions and no cytopathic viral effect in the skin. Cutaneous antiviral mechanisms include activation of TLR/IRF pathways and production of type I IFN. We evaluated the antiviral mechanisms involved in the skin of COVID-19 patients, including skin samples from 35 deceased patients who had contracted COVID-19 before the launch of the vaccine. Detection of SARS-CoV-2 in the skin was performed using transmission electron microscopy and RT-qPCR. Microscopic and molecular effects of the virus in skin were evaluated by histopathology, RT-qPCR, and immunohistochemistry (IHC). The results revealed the presence of SARS-CoV-2 and microscopic changes, including microvascular hyaline thrombi, perivascular dermatitis, and eccrine gland necrosis. There was increased transcription of TBK1 and a reduction in transcription of TNFα by RT-qPCR in the COVID-19 group. IHC revealed reduced expression of ACE2, TLR7, and IL-6, and elevated expression of IFN-β by epidermal cells. In the dermis, there was decreased expression of STING, IFN-β, and TNF-α and increased expression of IL-6 in sweat glands. Our results highlight the role of type I IFN in the skin of COVID-19 patients, which may modulate the cutaneous response to SARS-CoV-2. Full article

Although porcine sapelovirus (PSV) is generally subclinical, it can cause a wide range of clinical signs in some individuals, including respiratory distress, acute diarrhea, pneumonia, skin lesions, reproductive failure, and neurological diseases. In this study, we investigated the prevalence and genotype of PSV isolated from domestic pigs and wild boars in Korea. We also analyzed potential recombination events, and assessed the pathogenicity of the virus through animal experiments. In wild boars, the prevalence of PSV antibodies decreased slightly (by 1.8%) over 5 years (from 2019 to 2024); however, prevalence increased significantly (by 17.8%) in breeding sows. In samples from animals with diarrhea and respiratory clinical signs, the prevalence of PSV alone was 21.1%, whereas the prevalence of PSV mixed with other pathogens was also 21.1%. The whole genome of the PSV/Goryeong/KR-2019 strain isolated from a piglet with diarrhea was closely related to the Jpsv447 strain isolated in Japan in 2009, and recombination analysis predicted that the PSV/Goryeong/KR-2019 strain was generated by genetic recombination between the KS05151 strain and the Jpsv447 strain. However, when the PSV/Goryeong/KR-2019 strain was orally administered to 5-day-old suckling pigs, diarrhea clinical signs were mild, and no significant changes were observed in villus height and ridge depth in the duodenum, jejunum, or ileum. In addition, no neurological clinical signs were observed when the isolated virus was administered to 130-day-old pigs, and no specific lesions were found upon histopathological examination of brain tissue. In conclusion, PSV/Goryeong/KR-2019 appears to be a weakly pathogenic virus that does not cause severe diarrhea in suckling pigs, and does not cause neurological clinical signs in fattening pigs. Therefore, it is presumed that most PSVs detected in Korean pig farms are weakly pathogenic strains. Full article

In 2019, an unusual mortality event (UME) affected bottlenose dolphins (Tursiops truncatus) in the Mississippi Sound (MSS) following an extended dual opening of the Bonnet Carré Spillway (BCS), which prevents flooding in New Orleans. This resulted in low salinity, skin lesions, and electrolyte imbalances in dolphins. Additionally, the influx likely altered the MSS’s environmental chemical composition, including levels of heavy metals and metalloids; thus, we quantified heavy metals, metalloids, and essential elements in the tissues of dolphins that stranded in the MSS before and after the 2019 UME. We hypothesized that levels of heavy metals and metalloids (such as mercury (Hg), arsenic (As), lead (Pb), and cadmium (Cd)) would not show significant changes post-UME. Indeed, we found no major changes associated with the UME in most metals; sodium (Na) and magnesium (Mg) levels were lower in several tissues after 2019, which correlated with the average yearly salinity measured from the MSS. Toxic metals and metalloids were detectable with some changes over time; however, the selenium (Se):Hg molar ratio increased in some tissues post-2019. Additionally, we confirmed that Hg can bioaccumulate, with positive correlations between Hg levels and dolphin size as assessed by straight length. Overall, our findings indicate that freshwater incursions into the MSS can alter dolphin exposure to essential and toxic elements. Full article