[Skin lesion classification with multi-level fusion of Swin-T and ConvNeXt].

Skin cancer is a significant public health issue, and computer-aided diagnosis technology can effectively alleviate this burden. Accurate identification of skin lesion types is crucial when employing computer-aided diagnosis. This study proposes a multi-level attention cascaded fusion model based on Swin-T and ConvNeXt. It employed hierarchical Swin-T and ConvNeXt to extract global and local features, respectively, and introduced residual channel attention and spatial attention modules for further feature extraction. Multi-level attention mechanisms were utilized to process multi-scale global and local features. To address the problem of shallow features being lost due to their distance from the classifier, a hierarchical inverted residual fusion module was proposed to dynamically adjust the extracted feature information. Balanced sampling strategies and focal loss were employed to tackle the issue of imbalanced categories of skin lesions. Experimental testing on the ISIC2018 and ISIC2019 datasets yielded accuracy, precision, recall, and F1-Score of 96.01%, 93.67%, 92.65%, and 93.11%, respectively, and 92.79%, 91.52%, 88.90%, and 90.15%, respectively. Compared to Swin-T, the proposed method achieved an accuracy improvement of 3.60% and 1.66%, and compared to ConvNeXt, it achieved an accuracy improvement of 2.87% and 3.45%. The experiments demonstrate that the proposed method accurately classifies skin lesion images, providing a new solution for skin cancer diagnosis.

Grading differentiation in cutaneous squamous cell carcinoma: a review of the literature.

Poor differentiation is strongly associated with poor outcomes in cutaneous squamous cell carcinoma (CSCC). In addition, the National Comprehensive Cancer Network (NCCN) guidelines designate poorly differentiated tumors as "very high risk". Despite its clear prognostic implications, there is no standardized grading system for CSCC differentiation in common use today. CSCC differentiation is graded inconsistently by both dermatopathologists and Mohs surgeons, and reliability studies have demonstrated suboptimal inter- and intra-rater reliability in both of these groups. The absence of a standardized and reliable grading system has impeded the use of differentiation in CSCC staging, despite its apparent correlation with disease outcomes. We performed a comprehensive review of the literature summarizing historical CSCC differentiation grading systems, as well as grading systems in non-cutaneous head and neck SCC as a point of reference. Relevant articles were identified by searching Embase and PubMed, as well as by reviewing reference lists for additional articles and histology textbook excerpts. CSCC grading systems that were identified and summarized include the historical Broders system, the World Health Organization system, the College of American Pathologists' system, and a system described by a 2023 Delphi consensus panel of dermatopathologists.

Enhancing Skin Cancer Classification using Efficient Net B0-B7 through Convolutional Neural Networks and Transfer Learning with Patient-Specific Data.

BACKGROUND: Skin cancer diagnosis challenges dermatologists due to its complex visual variations across diagnostic categories. Convolutional neural networks (CNNs), specifically the Efficient Net B0-B7 series, have shown superiority in multiclass skin cancer classification. This study addresses the limitations of visual examination by presenting a tailored preprocessing pipeline designed for Efficient Net models. Leveraging transfer learning with pre-trained ImageNet weights, the research aims to enhance diagnostic accuracy in an imbalanced multiclass classification context. METHODS: The study develops a specialized image preprocessing pipeline involving image scaling, dataset augmentation, and artifact removal tailored to the nuances of Efficient Net models. Using the Efficient Net B0-B7 dataset, transfer learning fine-tunes CNNs with pre-trained ImageNet weights. Rigorous evaluation employs key metrics like Precision, Recall, Accuracy, F1 Score, and Confusion Matrices to assess the impact of transfer learning and fine-tuning on each Efficient Net variant's performance in classifying diverse skin cancer categories. RESULTS: The research showcases the effectiveness of the tailored preprocessing pipeline for Efficient Net models. Transfer learning and fine-tuning significantly enhance the models' ability to discern diverse skin cancer categories. The evaluation of eight Efficient Net models (B0-B7) for skin cancer classification reveals distinct performance patterns across various cancer classes. While the majority class, Benign Kertosis, achieves high accuracy (>87%), challenges arise in accurately classifying Eczema classes. Melanoma, despite its minority representation (2.42% of images), attains an average accuracy of 80.51% across all models. However, suboptimal performance is observed in predicting warts molluscum (90.7%) and psoriasis (84.2%) instances, highlighting the need for targeted improvements in accurately identifying specific skin cancer types. CONCLUSION: The study on skin cancer classification utilizes EfficientNets B0-B7 with transfer learning from ImageNet weights. The pinnacle performance is observed with EfficientNet-B7, achieving a groundbreaking top-1 accuracy of 84.4% and top-5 accuracy of 97.1%. Remarkably efficient, it is 8.4 times smaller than the leading CNN. Detailed per-class classification exactitudes through Confusion Matrices affirm its proficiency, signaling the potential of EfficientNets for precise dermatological image analysis.

Multi-scale feature fusion and class weight loss for skin lesion classification.

Skin cancer is one of the common types of cancer. It spreads quickly and is not easy to detect in the early stages, posing a major threat to human health. In recent years, deep learning methods have attracted widespread attention for skin cancer detection in dermoscopic images. However, training a practical classifier becomes highly challenging due to inter-class similarity and intra-class variation in skin lesion images. To address these problems, we propose a multi-scale fusion structure that combines shallow and deep features for more accurate classification. Simultaneously, we implement three approaches to the problem of class imbalance: class weighting, label smoothing, and resampling. In addition, the HAM10000_RE dataset strips out hair features to demonstrate the role of hair features in the classification process. We demonstrate that the region of interest is the most critical classification feature for the HAM10000_SE dataset, which segments lesion regions. We evaluated the effectiveness of our model using the HAM10000 and ISIC2019 dataset. The results showed that this method performed well in dermoscopic classification tasks, with ACC and AUC of 94.0% and 99.3%, on the HAM10000 dataset and ACC of 89.8% for the ISIC2019 dataset. The overall performance of our model is excellent in comparison to state-of-the-art models.

An Ensemble Architecture for Melanoma Classification.

Melanoma represents an extremely aggressive type of skin lesion. Despite its high mortality rate, when detected in its initial stage, the projected five-year survival rate is notably high. The advancement of Artificial Intelligence in recent years has facilitated the creation of diverse solutions aimed at assisting medical diagnosis. This proposal presents an architecture for melanoma classification.

A novel skin cancer detection model using modified finch deep CNN classifier model.

Skin cancer is one of the most life-threatening diseases caused by the abnormal growth of the skin cells, when exposed to ultraviolet radiation. Early detection seems to be more crucial for reducing aberrant cell proliferation because the mortality rate is rapidly rising. Although multiple researches are available based on the skin cancer detection, there still exists challenges in improving the accuracy, reducing the computational time and so on. In this research, a novel skin cancer detection is performed using a modified falcon finch deep Convolutional neural network classifier (Modified Falcon finch deep CNN) that efficiently detects the disease with higher efficiency. The usage of modified falcon finch deep CNN classifier effectively analyzed the information relevant to the skin cancer and the errors are also minimized. The inclusion of the falcon finch optimization in the deep CNN classifier is necessary for efficient parameter tuning. This tuning enhanced the robustness and boosted the convergence of the classifier that detects the skin cancer in less stipulated time. The modified falcon finch deep CNN classifier achieved accuracy, sensitivity, and specificity values of 93.59%, 92.14%, and 95.22% regarding k-fold and 96.52%, 96.69%, and 96.54% regarding training percentage, proving more effective than literary works.

Evaluation of High-Dimensional Data Classification for Skin Malignancy Detection Using DL-Based Techniques.

Skin cancer can be detected through visual screening and skin analysis based on the biopsy and pathological state of the human body. The survival rate of cancer patients is low, and millions of people are diagnosed annually. By determining the different comparative analyses, the skin malignancy classification is evaluated. Using the Isomap with the vision transformer, we analyze the high-dimensional images with dimensionality reduction. Skin cancer can present with severe cases and life-threatening symptoms. Overall performance evaluation and classification tend to improve the accuracy of the high-dimensional skin lesion dataset when completed. In deep learning methodologies, the distinct phases of skin malignancy classification are determined by its accuracy, specificity, F1 recall, and sensitivity while implementing the classification methodology. A nonlinear dimensionality reduction technique called Isomap preserves the data's underlying nonlinear relationships intact. This is essential for the categorization of skin malignancies, as the features that separate malignant from benign skin lesions may not be linearly separable. Isomap decreases the data's complexity while maintaining its essential characteristics, making it simpler to analyze and explain the findings. High-dimensional datasets for skin lesions have been evaluated and classified more effectively when evaluated and classified using Isomap with the vision transformer.

Mesenchymal skin tumors - Novel entities in the 5th edition of WHO classification of skin tumors.

The section on mesenchymal tumors in 5th edition of WHO classification of skin tumors has undergone several changes, the most important of which, as usual, is the inclusion of newly identified tumor entities, which will be the main focus of this review article. These specifically include three novel cutaneous mesenchymal tumors with melanocytic differentiation, and rearrangements of the CRTC1::TRIM11, ACTIN::MITF, and MITF::CREM genes. In addition, EWSR1::SMAD3-rearranged fibroblastic tumors, superficial CD34-positive fibroblastic tumors, and NTRK-rearranged spindle cell neoplasms were newly included. Of the other changes, only the most important ones will be briefly mentioned.

Comprehensive analysis of clinical images contributions for melanoma classification using convolutional neural networks.

BACKGROUND: Timely diagnosis plays a critical role in determining melanoma prognosis, prompting the development of deep learning models to aid clinicians. Questions persist regarding the efficacy of clinical images alone or in conjunction with dermoscopy images for model training. This study aims to compare the classification performance for melanoma of three types of CNN models: those trained on clinical images, dermoscopy images, and a combination of paired clinical and dermoscopy images from the same lesion. MATERIALS AND METHODS: We divided 914 image pairs into training, validation, and test sets. Models were built using pre-trained Inception-ResNetV2 convolutional layers for feature extraction, followed by binary classification. Training comprised 20 models per CNN type using sets of random hyperparameters. Best models were chosen based on validation AUC-ROC. RESULTS: Significant AUC-ROC differences were found between clinical versus dermoscopy models (0.661 vs. 0.869, p < 0.001) and clinical versus clinical + dermoscopy models (0.661 vs. 0.822, p = 0.001). Significant sensitivity differences were found between clinical and dermoscopy models (0.513 vs. 0.799, p = 0.01), dermoscopy versus clinical + dermoscopy models (0.799 vs. 1.000, p = 0.02), and clinical versus clinical + dermoscopy models (0.513 vs. 1.000, p < 0.001). Significant specificity differences were found between dermoscopy versus clinical + dermoscopy models (0.800 vs. 0.288, p < 0.001) and clinical versus clinical + dermoscopy models (0.650 vs. 0.288, p < 0.001). CONCLUSION: CNN models trained on dermoscopy images outperformed those relying solely on clinical images under our study conditions. The potential advantages of incorporating paired clinical and dermoscopy images for CNN-based melanoma classification appear less clear based on our findings.

A novel SpaSA based hyper-parameter optimized FCEDN with adaptive CNN classification for skin cancer detection.

Skin cancer is the most prevalent kind of cancer in people. It is estimated that more than 1 million people get skin cancer every year in the world. The effectiveness of the disease's therapy is significantly impacted by early identification of this illness. Preprocessing is the initial detecting stage in enhancing the quality of skin images by removing undesired background noise and objects. This study aims is to compile preprocessing techniques for skin cancer imaging that are currently accessible. Researchers looking into automated skin cancer diagnosis might use this article as an excellent place to start. The fully convolutional encoder-decoder network and Sparrow search algorithm (FCEDN-SpaSA) are proposed in this study for the segmentation of dermoscopic images. The individual wolf method and the ensemble ghosting technique are integrated to generate a neighbour-based search strategy in SpaSA for stressing the correct balance between navigation and exploitation. The classification procedure is accomplished by using an adaptive CNN technique to discriminate between normal skin and malignant skin lesions suggestive of disease. Our method provides classification accuracies comparable to commonly used incremental learning techniques while using less energy, storage space, memory access, and training time (only network updates with new training samples, no network sharing). In a simulation, the segmentation performance of the proposed technique on the ISBI 2017, ISIC 2018, and PH2 datasets reached accuracies of 95.28%, 95.89%, 92.70%, and 98.78%, respectively, on the same dataset and assessed the classification performance. It is accurate 91.67% of the time. The efficiency of the suggested strategy is demonstrated through comparisons with cutting-edge methodologies.

Cutaneous lymphoproliferative disorders: Back to the future.

In the 1980s, immunohistochemistry and clonality analyses became instrumental in the recognition and definition of new types of cutaneous T-cell lymphoma (CTCL) and cutaneous B-cell lymphoma (CBCL) and the development of new classifications. By accepting loss of pan-T-cell antigens and clonal T-cell receptor gene rearrangements as important criteria to differentiate between benign and malignant T-cell proliferations, and monotypic immunoglobulin light-chain expression and clonal immunoglobulin gene rearrangements as crucial criteria to distinguish between benign and malignant B-cell proliferations, many cases, until then diagnosed as cutaneous lymphoid hyperplasia or pseudolymphoma, were reclassified as primary cutaneous CD4+ small/medium T-cell lymphoma (PCSM-TCL) or primary cutaneous marginal zone lymphoma (PCMZL), respectively. However, in recent years there is growing awareness that neither these immunohistochemical criteria nor demonstration of T-cell or B-cell clonality is specific for malignant lymphomas. In addition, many studies have reported that these low-grade malignant CTCL and CBCL have an indolent clinical behavior and an excellent prognosis with disease-specific survival rates of or close to 100%. As a result, recent classifications have downgraded several low-grade malignant cutaneous lymphomas to lymphoproliferative disorder (LPD). Both the 5th edition of the WHO classification (2022) and the 2022 International Consensus Classification (ICC) of mature lymphoid neoplasms reclassified PCSM-TCL as primary cutaneous CD4+ small/medium T-cell LPD and primary cutaneous acral CD8+ T-cell lymphoma as primary cutaneous acral CD8+ T cell LPD. While the 2022 ICC introduced the term "primary cutaneous marginal zone LPD," in the 5th edition of the WHO classification PCMZL is maintained. In this review we describe the background and rationale of the continually changing terminology of these conditions and discuss the clinical consequences of downgrading malignant lymphomas to LPDs.

Tumores de colisión en la piel: un hallazgo incidental en la mayoría de los casos. Estudio retrospectivo y revisión de la literatura / Collision tumours in skin: usually an incidental finding. A retrospective study with literature revision

Se denomina tumor de colisión (TC) a la coexistencia de dos o más neoplasias independientes en la misma resección. Suelen ser hallazgos incidentales en la piel, de patogénesis y prevalencia desconocidas, con pocas referencias en la literatura. Aquí mostramos un estudio retrospectivo de TC diagnosticados por un dermatopatólogo entre los años 2019-2022 en nuestro centro. Se han definido las lesiones de manera independiente y organizado cada colisión en categorías: benigno-benigno (BB), benigno-maligno (BM) y maligno-maligno (MM). Del total de 108 TC (1,4% de las biopsias totales del dermatopatólogo en ese periodo), se detecta que la colisión más frecuente es la formada entre BM (48,5%), con un carcinoma basocelular (CBC) como lesión maligna más frecuente de forma global y con un nevus melanocítico (NM) como lesión benigna principal. Se ha realizado el análisis estadístico de los resultados con el software Stata 14.2, detectando una diferencia estadísticamente significativa entre edad y tipo de colisión. (AU)

A collision tumour (CT) is a neoplastic lesion comprised of two or more distinct cell populations that maintain distinct borders. Mostly, these are incidental findings in skin biopsies, whose pathologic mechanism and prevalence remain unknown, with few references among literature. Here, we present a retrospective study of CT, diagnosed by a dermatopathologist in our hospital between 2019-2022. Lesions have been defined individually and organized into three categories: benign-benign (BB), benign-malignant (BM) and malignant-malignant (MM). A total of 108 CT were diagnosed (1,4% of the biopsies from the dermatopathologist during this period), from which BM was the most frequent collision (48,5%). Globally, basal cell carcinoma (BCC) was the main malignant lesion and melanocytic nevus (MN) the main benign lesion. We have used the software Stata 14.2 in order to analyse results, and we have detected a statistically significant difference between age and collision type. (AU)

Tumores de colisión en la piel: un hallazgo incidental en la mayoría de los casos. Estudio retrospectivo y revisión de la literatura / Collision tumours in skin: usually an incidental finding. A retrospective study with literature revision

Se denomina tumor de colisión (TC) a la coexistencia de dos o más neoplasias independientes en la misma resección. Suelen ser hallazgos incidentales en la piel, de patogénesis y prevalencia desconocidas, con pocas referencias en la literatura. Aquí mostramos un estudio retrospectivo de TC diagnosticados por un dermatopatólogo entre los años 2019-2022 en nuestro centro. Se han definido las lesiones de manera independiente y organizado cada colisión en categorías: benigno-benigno (BB), benigno-maligno (BM) y maligno-maligno (MM). Del total de 108 TC (1,4% de las biopsias totales del dermatopatólogo en ese periodo), se detecta que la colisión más frecuente es la formada entre BM (48,5%), con un carcinoma basocelular (CBC) como lesión maligna más frecuente de forma global y con un nevus melanocítico (NM) como lesión benigna principal. Se ha realizado el análisis estadístico de los resultados con el software Stata 14.2, detectando una diferencia estadísticamente significativa entre edad y tipo de colisión. (AU)

A collision tumour (CT) is a neoplastic lesion comprised of two or more distinct cell populations that maintain distinct borders. Mostly, these are incidental findings in skin biopsies, whose pathologic mechanism and prevalence remain unknown, with few references among literature. Here, we present a retrospective study of CT, diagnosed by a dermatopathologist in our hospital between 2019-2022. Lesions have been defined individually and organized into three categories: benign-benign (BB), benign-malignant (BM) and malignant-malignant (MM). A total of 108 CT were diagnosed (1,4% of the biopsies from the dermatopathologist during this period), from which BM was the most frequent collision (48,5%). Globally, basal cell carcinoma (BCC) was the main malignant lesion and melanocytic nevus (MN) the main benign lesion. We have used the software Stata 14.2 in order to analyse results, and we have detected a statistically significant difference between age and collision type. (AU)

Restrospective reappraisal of the prognostic classification of spitzoid melanocytic neoplasms after BRAF and NRAS mutation characterisation: a single institution experience.

AIMS: The current WHO classification of melanocytic tumours excludes neoplasms showing BRAF or NRAS mutations from the Spitz category. This study aimed to review and reclassify atypical melanocytic tumours with spitzoid morphological features diagnosed between 2009 and 2021 in our hospital after expanding the molecular profile, including BRAF and NRAS mutations in all cases. METHODS AND RESULTS: A total of 71 neoplasms showing spitzoid features (Spitz-like) and atypia were included. The risk of progression of tumours was first studied by integrating the morphology, immunohistochemistry (p16, Ki67, HMB45 and PRAME) and fluorescence in-situ hybridisation (FISH) results (melanoma multiprobe and 9p21). In a second step, after expanding the molecular study, including BRAF and NRAS mutational status, the neoplasms were finally classified into four subgroups: atypical Spitz tumour (AST, n = 45); BRAF-mutated naevus/low-grade melanocytoma with spitzoid morphology (BAMS, n = 2); Spitz melanoma (SM, n = 14); and BRAF or NRAS mutated melanoma with spitzoid features (MSF, n = 10). Follow-up of patients revealed uneventful results for AST and BAMS. Only one SM presented lymph node metastasis after 134 months. Conversely, patients with MSF showed an unfavourable outcome: three developed lymph node metastases after a mean time of 22 months, with one patient presenting distant metastasis and dying of the disease 64 months from diagnosis. The progression-free survival showed significant differences between the four groups of spitzoid tumours (P < 0.001) and between both melanoma subtypes (P = 0.012). CONCLUSIONS: The classification and prognostication of atypical neoplasms with spitzoid features requires the integration of histomorphology with the molecular investigation of tumours, which should include BRAF and NRAS mutational status.

Registro de linfomas cutáneos primarios (RELCP) de la AEDV: datos tras 5 años de funcionamiento / Primary Cutaneous Lymphoma Registry of the Spanish Academy of Dermatology and Venereology (AEDV): Data for the First 5 Years

Antecedentes y objetivos Los linfomas cutáneos primarios (LCP) son un conjunto de entidades poco frecuentes. En febrero del 2018 se describieron los resultados del primer año de funcionamiento del Registro de linfomas cutáneos primarios de la AEDV. En el presente trabajo actualizamos los resultados tras 5 años de funcionamiento. Pacientes y métodos Registro de enfermedad de pacientes con LCP. Se recogieron datos prospectivamente de los pacientes, incluyendo diagnóstico, tratamientos, pruebas realizadas y estado actual del paciente. Se realizó un análisis descriptivo. Resultados En diciembre del 2021 se había incluido a un total de 2020 pacientes en el Registro, pertenecientes a 33 hospitales españoles. El 59% fueron hombres, la edad media fue de 62,2 años. Se agruparon en 4grandes grupos diagnósticos: micosis fungoide/síndrome de Sézary (1.112, 55%), LCP de células B (547, 27,1%), trastornos linfoproliferativos de células T CD30+(222, 11%) y otros linfomas T (116, 5,8%). La mayoría presentó estadio T1, encontrándose actualmente casi el 75% en remisión completa (43,5%) o enfermedad estable (EE; 27%). Los tratamientos más usados fueron corticoides tópicos (1.369, 67,8%), fototerapia (890, 44,1%), cirugía (412, 20,4%) y radioterapia (384, 19%). Conclusión Las características del paciente con LCP en España no difieren de otras series. El mayor tamaño del registro permite precisar mejor los datos con respecto a los resultados del primer año. Este registro facilita al grupo de linfomas de la AEDV realizar investigación clínica, surgiendo ya trabajos publicados de dicho registro (AU)

Background and objective Primary cutaneous lymphomas (PCL) are uncommon. Observations based on the first year of data from the Spanish Registry of Primary Cutaneous Lymphomas (RELCP, in its Spanish abbreviation) of the Spanish Academy of Dermatology and Venereology (AEDV) were published in February 2018. This report covers RELCP data for the first 5 years. Patients and methods RELCP data were collected prospectively and included diagnosis, treatments, tests, and the current status of patients. We compiled descriptive statistics of the data registered during the first 5 years. Results Information on 2020 patients treated at 33 Spanish hospitals had been included in the RELCP by December 2021. Fifty-nine percent of the patients were men; the mean age was 62.2 years. The lymphomas were grouped into 4 large diagnostic categories: mycosis fungoides/Sézary syndrome, 1112 patients (55%); primary B-cell cutaneous lymphoma, 547 patients (27.1%); primary CD30+lymphoproliferative disorders, 222 patients (11%), and other T-cell lymphomas, 116 patients (5.8%). Nearly 75% of the tumors were registered in stage I. After treatment, 43.5% achieved complete remission and 27% were stable at the time of writing. Treatments prescribed were topical corticosteroids (1369 [67.8%]), phototherapy (890 patients [44.1%]), surgery (412 patients [20.4%]), and radiotherapy (384 patients [19%]). Conclusion The characteristics of cutaneous lymphomas in Spain are similar to those reported for other series. The large size of the RELCP registry at 5 years has allowed us to give more precise descriptive statistics than in the first year. This registry facilitates the clinical research of the AEDV's lymphoma interest group, which has already published articles based on the RELCP data (AU)

[Translated article] Primary Cutaneous Lymphoma Registry of the Spanish Academy of Dermatology and Venereology (AEDV): Data for the First 5 Years / Registro de linfomas cutáneos primarios (RELCP) de la AEDV: datos tras 5 años de funcionamiento

Background and objective Primary cutaneous lymphomas (PCL) are uncommon. Observations based on the first year of data from the Spanish Registry of Primary Cutaneous Lymphomas (RELCP, in its Spanish abbreviation) of the Spanish Academy of Dermatology and Venereology (AEDV) were published in February 2018. This report covers RELCP data for the first 5 years. Patients and methods RELCP data were collected prospectively and included diagnosis, treatments, tests, and the current status of patients. We compiled descriptive statistics of the data registered during the first 5 years. Results Information on 2020 patients treated at 33 Spanish hospitals had been included in the RELCP by December 2021. Fifty-nine percent of the patients were men; the mean age was 62.2 years. The lymphomas were grouped into 4 large diagnostic categories: mycosis fungoides/Sézary syndrome, 1112 patients (55%); primary B-cell cutaneous lymphoma, 547 patients (27.1%); primary CD30+lymphoproliferative disorders, 222 patients (11%), and other T-cell lymphomas, 116 patients (5.8%). Nearly 75% of the tumors were registered in stage I. After treatment, 43.5% achieved complete remission and 27% were stable at the time of writing. Treatments prescribed were topical corticosteroids (1369 [67.8%]), phototherapy (890 patients [44.1%]), surgery (412 patients [20.4%]), and radiotherapy (384 patients [19%]). Conclusion The characteristics of cutaneous lymphomas in Spain are similar to those reported for other series. The large size of the RELCP registry at 5 years has allowed us to give more precise descriptive statistics than in the first year. This registry facilitates the clinical research of the AEDV's lymphoma interest group, which has already published articles based on the RELCP data (AU)

Antecedentes y objetivos Los linfomas cutáneos primarios (LCP) son un conjunto de entidades poco frecuentes. En febrero del 2018 se describieron los resultados del primer año de funcionamiento del Registro de linfomas cutáneos primarios de la AEDV. En el presente trabajo actualizamos los resultados tras 5 años de funcionamiento. Pacientes y métodos Registro de enfermedad de pacientes con LCP. Se recogieron datos prospectivamente de los pacientes, incluyendo diagnóstico, tratamientos, pruebas realizadas y estado actual del paciente. Se realizó un análisis descriptivo. Resultados En diciembre del 2021 se había incluido a un total de 2020 pacientes en el Registro, pertenecientes a 33 hospitales españoles. El 59% fueron hombres, la edad media fue de 62,2 años. Se agruparon en 4grandes grupos diagnósticos: micosis fungoide/síndrome de Sézary (1.112, 55%), LCP de células B (547, 27,1%), trastornos linfoproliferativos de células T CD30+(222, 11%) y otros linfomas T (116, 5,8%). La mayoría presentó estadio T1, encontrándose actualmente casi el 75% en remisión completa (43,5%) o enfermedad estable (EE; 27%). Los tratamientos más usados fueron corticoides tópicos (1.369, 67,8%), fototerapia (890, 44,1%), cirugía (412, 20,4%) y radioterapia (384, 19%). Conclusión Las características del paciente con LCP en España no difieren de otras series. El mayor tamaño del registro permite precisar mejor los datos con respecto a los resultados del primer año. Este registro facilita al grupo de linfomas de la AEDV realizar investigación clínica, surgiendo ya trabajos publicados de dicho registro (AU)

DUNEScan: a web server for uncertainty estimation in skin cancer detection with deep neural networks.

Recent years have seen a steep rise in the number of skin cancer detection applications. While modern advances in deep learning made possible reaching new heights in terms of classification accuracy, no publicly available skin cancer detection software provide confidence estimates for these predictions. We present DUNEScan (Deep Uncertainty Estimation for Skin Cancer), a web server that performs an intuitive in-depth analysis of uncertainty in commonly used skin cancer classification models based on convolutional neural networks (CNNs). DUNEScan allows users to upload a skin lesion image, and quickly compares the mean and the variance estimates provided by a number of new and traditional CNN models. Moreover, our web server uses the Grad-CAM and UMAP algorithms to visualize the classification manifold for the user's input, hence providing crucial information about its closeness to skin lesion images  from the popular ISIC database. DUNEScan is freely available at: https://www.dunescan.org .

Multi-features extraction based on deep learning for skin lesion classification.

For various forms of skin lesion, many different feature extraction methods have been investigated so far. Indeed, feature extraction is a crucial step in machine learning processes. In general, we can distinct handcrafted and deep learning features. In this paper, we investigate the efficiency of using 17 commonly pre-trained convolutional neural networks (CNN) architectures as feature extractors and of 24 machine learning classifiers to evaluate the classification of skin lesions from two different datasets: ISIC 2019 and PH2. In this research, we find out that a DenseNet201 combined with Fine KNN or Cubic SVM achieved the best results in accuracy (92.34% and 91.71%) for the ISIC 2019 dataset. The results also show that the suggested method outperforms others approaches with an accuracy of 99% on the PH2 dataset.

Characterization of multiple diagnostic terms in melanocytic skin lesion pathology reports.

BACKGROUND: Histopathologically ambiguous melanocytic lesions lead some pathologists to list multiple diagnostic considerations in the pathology report. The frequency and circumstance of multiple diagnostic considerations remain poorly characterized. METHODS: Two hundred and forty skin biopsy samples were interpreted by 187 pathologists (8976 independent diagnoses) and classified according to a diagnostic/treatment stratification (MPATH-Dx). RESULTS: Multiple diagnoses in different MPATH-Dx classes were used in n = 1320 (14.7%) interpretations, with 97% of pathologists and 91% of cases having at least one such interpretation. Multiple diagnoses were more common for intermediate risk lesions and are associated with greater subjective difficulty and lower confidence. We estimate that 6% of pathology reports for melanocytic lesions in the United States contain two diagnoses of different MPATH-Dx prognostic classes, and 2% of cases are given two diagnoses with significant treatment implications. CONCLUSIONS: Difficult melanocytic diagnoses in skin may necessitate multiple diagnostic considerations; however, as patients increasingly access their health records and retrieve pathology reports (as mandated by US law), uncertainty should be expressed unambiguously.

Deep learning data augmentation for Raman spectroscopy cancer tissue classification.

Recently, Raman Spectroscopy (RS) was demonstrated to be a non-destructive way of cancer diagnosis, due to the uniqueness of RS measurements in revealing molecular biochemical changes between cancerous vs. normal tissues and cells. In order to design computational approaches for cancer detection, the quality and quantity of tissue samples for RS are important for accurate prediction. In reality, however, obtaining skin cancer samples is difficult and expensive due to privacy and other constraints. With a small number of samples, the training of the classifier is difficult, and often results in overfitting. Therefore, it is important to have more samples to better train classifiers for accurate cancer tissue classification. To overcome these limitations, this paper presents a novel generative adversarial network based skin cancer tissue classification framework. Specifically, we design a data augmentation module that employs a Generative Adversarial Network (GAN) to generate synthetic RS data resembling the training data classes. The original tissue samples and the generated data are concatenated to train classification modules. Experiments on real-world RS data demonstrate that (1) data augmentation can help improve skin cancer tissue classification accuracy, and (2) generative adversarial network can be used to generate reliable synthetic Raman spectroscopic data.