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.

Expression Profiles of ASIC1/2 and TRPV1/4 in Common Skin Tumors.

The acid-sensing ion channels ASIC1 and ASIC2, as well as the transient receptor potential vanilloid channels TRPV1 and TRPV4, are proton-gated cation channels that can be activated by low extracellular pH (pHe), which is a hallmark of the tumor microenvironment in solid tumors. However, the role of these channels in the development of skin tumors is still unclear. In this study, we investigated the expression profiles of ASIC1, ASIC2, TRPV1 and TRPV4 in malignant melanoma (MM), squamous cell carcinoma (SCC), basal cell carcinoma (BCC) and in nevus cell nevi (NCN). We conducted immunohistochemistry using paraffin-embedded tissue samples from patients and found that most skin tumors express ASIC1/2 and TRPV1/4. Striking results were that BCCs are often negative for ASIC2, while nearly all SCCs express this marker. Epidermal MM sometimes seem to lack ASIC1 in contrast to NCN. Dermal portions of MM show strong expression of TRPV1 more frequently than dermal NCN portions. Some NCN show a decreasing ASIC1/2 expression in deeper dermal tumor tissue, while MM seem to not lose ASIC1/2 in deeper dermal portions. ASIC1, ASIC2, TRPV1 and TRPV4 in skin tumors might be involved in tumor progression, thus being potential diagnostic and therapeutic targets.

TERTp mutations and p53 expression in head and neck cutaneous basal cell carcinomas with different aggressive features.

Cutaneous basal cell carcinoma (cBCC) is an economic burden to health services, due to its great morbidity and increasing incidence in old people. Infiltrative cBCCs and cBCCs with micronodular pattern are considered as more aggressive. The role of p53 expression and TERTp mutation on cBCC behavior remains to be clarified. We aimed to assess TERTp mutations and p53 expression in relation to the cBCC histological subtype in a cohort of patients referred to an ENT Department of a tertiary Hospital of Northern Portugal. We performed a retrospective clinicopathological and histological review of the head and neck cBCCs followed-up at the otorhinolaryngology department of Trás-os-Montes e Alto Douro hospital (January 2007-June 2018). We assessed TERTp mutations in 142 cBCCs and p53 protein expression, through immunohistochemistry, in 157 cBCCs. We detected TERTp mutations in 43.7% of cBCCs and p53 overexpression in 60.5% of cBCCs. We spotted association of p53 overexpression and TERTp mutation with necrosis. In the infitrative-growth pattern cBCCs, there was no significant association with the clinical and histological features evaluated, except for necrosis. In the indolent-growth cBCCs, we identified a significant association of TERTp mutation status with female sex, necrosis, multiple cBCCs, and p53 positive expression. Our results suggest that TERTp mutation may be useful to identify more aggressive features in the indolent-growth pattern cBCCs (nodular and superficial subtypes). Further studies with larger cohorts are warranted to clarify the relevance of TERTp mutation in cBCCs.

Line-field optical coherence tomography: in vivo diagnosis of basal cell carcinoma subtypes compared with histopathology.

BACKGROUND: Basal cell carcinoma (BCC) is the most common skin cancer in the general population. Treatments vary from Mohs surgery to topical therapy, depending on the subtype. Dermoscopy, reflectance confocal microscopy (RCM) and optical coherence tomography (OCT) have gained a foothold in daily clinical practice to optimize diagnosis and subtype-oriented treatment. The new technique of line-field confocal OCT (LC-OCT) allows imaging at high resolution and depth, but its use has not yet been investigated in larger studies. AIM: To evaluate the main LC-OCT criteria for the diagnosis and subtyping of BCC compared with histopathology, OCT and RCM. METHODS: In total, 52 histopathologically confirmed BCCs were evaluated for imaging criteria. Their frequency, predictive values and ROC curves were calculated. A multinominal regression with stepwise variables selection to distinguish BCC subtypes was performed. RESULTS: Nodular BCCs were mainly characterized by atypical keratinocytes, altered dermoepidermal junction (DEJ), tumour nests in the dermis, dark clefting, prominent vascularization and white hyper-reflective stroma. Superficial BCCs showed a thickening of the epidermis due to a series of tumour lobules with clear connection to the DEJ (string of pearls pattern). Infiltrative BCCs were characterized by elongated hyporeflective tumour strands, surrounded by bright collagen (shoal of fish pattern). The overall BCC subtype agreement between LC-OCT and conventional histology was 90.4% (95% CI 79.0-96.8). CONCLUSION: LC-OCT allows noninvasive, real-time identification of BCCs and their subtypes in vertical, horizontal and three-dimension mode compared with histology, RCM and OCT. Further larger studies are needed to better explore the clinical applications of this promising device.

Subtype-Specific Analyses Reveal Infiltrative Basal Cell Carcinomas Are Highly Interactive with their Environment.

Little is known regarding the molecular differences between basal cell carcinoma (BCC) subtypes, despite clearly distinct phenotypes and clinical outcomes. In particular, infiltrative BCCs have poorer clinical outcomes in terms of response to therapy and propensity for dissemination. In this project, we aimed to use exome sequencing and RNA sequencing to identify somatic mutations and molecular pathways leading to infiltrative BCCs. Using whole-exome sequencing of 36 BCC samples (eight infiltrative) combined with previously reported exome data (58 samples), we determine that infiltrative BCCs do not contain a distinct somatic variant profile and carry classical UV-induced mutational signatures. RNA sequencing on both datasets revealed key differentially expressed genes, such as POSTN and WISP1, suggesting increased integrin and Wnt signaling. Immunostaining for periostin and WISP1 clearly distinguished infiltrative BCCs, and nuclear ß-catenin staining patterns further validated the resulting increase in Wnt signaling in infiltrative BCCs. Of significant interest, in BCCs with mixed morphology, infiltrative areas expressed WISP1, whereas nodular areas did not, supporting a continuum between subtypes. In conclusion, infiltrative BCCs do not differ in their genomic alteration in terms of initiating mutations. They display a specific type of interaction with the extracellular matrix environment regulating Wnt signaling.

Integrative analysis of breast cancer profiles in TCGA by TNBC subgrouping reveals novel microRNA-specific clusters, including miR-17-92a, distinguishing basal-like 1 and basal-like 2 TNBC subtypes.

BACKGROUND: The term triple-negative breast cancer (TNBC) is used to describe breast cancers without expression of estrogen receptor, progesterone receptor or HER2 amplification. To advance targeted treatment options for TNBC, it is critical that the subtypes within this classification be described in regard to their characteristic biology and gene expression. The Cancer Genome Atlas (TCGA) dataset provides not only clinical and mRNA expression data but also expression data for microRNAs. RESULTS: In this study, we applied the Lehmann classifier to TCGA-derived TNBC cases which also contained microRNA expression data and derived subtype-specific microRNA expression patterns. Subsequent analyses integrated known and predicted microRNA-mRNA regulatory nodes as well as patient survival data to identify key networks. Notably, basal-like 1 (BL1) TNBCs were distinguished from basal-like 2 TNBCs through up-regulation of members of the miR-17-92 cluster of microRNAs and suppression of several known miR-17-92 targets including inositol polyphosphate 4-phosphatase type II, INPP4B. CONCLUSIONS: These data demonstrate TNBC subtype-specific microRNA and target mRNA expression which may be applied to future biomarker and therapeutic development studies.

Single versus Multiple Level Sectioning for the Subtyping of Basal-Cell Carcinoma: A Retrospective Study.

BACKGROUND: The histological subtype of basal-cell carcinoma (BCC) is often based on a punch biopsy; only a small part is evaluated, possibly leading to misclassification. Consensus on the optimal approach to process punch biopsies is lacking, though accurate subtyping is important for appropriate treatment. OBJECTIVE: The aim is to investigate whether evaluating 4 levels of a punch biopsy instead of 1 or 2 levels leads to more accurate subtyping of BCC. METHODS: In a retrospective study we evaluated 87 punch biopsies of histologically confirmed BCCs. The primary outcome was the proportion of "more aggressive" BCCs (nonsuperficial vs. superficial, infiltrative vs. nodular subtype) that was missed by evaluation on 1 or 2 levels, using 4-level diagnosis as reference standard. RESULTS: Eighty-five cases were available for analysis. Subtyping based on 1 level resulted in discrepancies with 4-level diagnosis in 16.5% of all cases. Underdiagnosis occurred in 14 of 58 nonsuperficial BCCs (24.1%, 95% CI: 13.9-37.2). Seven of 38 nodular BCCs (18.4%, 95% CI: 7.74-34.3) were diagnosed as superficial in 1 level, and 7 of 20 infiltrative BCCs (35%, 95% CI: 15.4-59.2) were diagnosed as superficial (n = 2) or nodular (n = 5) in 1 level. CONCLUSION: In order to maximize correct subtyping and plan appropriate treatment, we advise to evaluate at least 2, but preferably more, levels of a punch biopsy to determine the BCC subtype.

Evaluation of Patched-1 Protein Expression Level in Low Risk and High Risk Basal Cell Carcinoma Subtypes.

Objective: Basal cell carcinoma (BCC) is the most common malignancy in humans and represents a growing public health care problem. The major etiological factors contributing to BCC development are exposure to ultraviolet radiation and genetic alterations. BCC is primarily caused by dysregulation of sonic Hedgehog (HH) signaling pathway in basal cells of the skin. BCC can be classified into low risk non-aggressive and high risk aggressive subtypes. BCC subtypes differentiation is essential for prognosis and for better disease management and treatment strategies. The aim of this study was to assess the correlation between PTCH1 protein expression level and the aggressiveness of BCC histopathology. Methods: Archival paraffin embedded blocks containing BCC were retrieved from a cohort of 101 patients. Immunohistochemistry staining was performed to assess the expression level of PTCH1 which is a key component of Hedgehog pathway. Results: 101 paraffin embedded samples were evaluated and classified as high risk and low risk BCC subtypes by histopathological finding. High risk BCC subtypes were found in 40 samples (39.6%) and low risk subtypes were identified in 61 samples (60.4%). Nodular was the most frequent subtype which was found in (56/ 101), followed by infiltrative (22/101) and micronodular (14/ 101) subtypes. Positive PTCH1 expression was found highest in nodular subtypes (46.5%). Conclusion: In this study, the correlation between low risk or high risk BCC subtypes and PTCH1 expression level was not statistically significant (p>0.05), but the frequency of positive PTCH1 expression was found to be higher in low risk subtypes than high risk BCC subtypes.

Convolutional Neural Network Approach to Classify Skin Lesions Using Reflectance Confocal Microscopy.

We propose an approach based on a convolutional neural network to classify skin lesions using the reflectance confocal microscopy (RCM) mosaics. Skin cancers are the most common type of cancers and a correct, early diagnosis significantly lowers both morbidity and mortality. RCM is an in-vivo non-invasive screening tool that produces virtual biopsies of skin lesions but its proficient and safe use requires hard to obtain expertise. Therefore, it may be useful to have an additional tool to aid diagnosis. The proposed network is based on the ResNet architecture. The dataset consists of 429 RCM mosaics and is divided into 3 classes: melanoma, basal cell carcinoma, and benign naevi with the ground-truth confirmed by a histopathological examination. The test set classification accuracy was 87%, higher than the accuracy achieved by medical, confocal users. The results show that the proposed classification system can be a useful tool to aid in early, noninvasive melanoma detection.

Lesions Mimicking Melanoma at Dermoscopy Confirmed Basal Cell Carcinoma: Evaluation with Reflectance Confocal Microscopy.

BACKGROUND: Atypical basal cell carcinoma (BCC), characterized by equivocal dermoscopic features typical of malignant melanoma (MM), can be difficult to diagnose. Reflectance confocal microscopy (RCM) enables in vivo imaging at nearly histological resolution. OBJECTIVES: To evaluate with RCM atypical melanocytic lesions identified in dermoscopy, according to common RCM criteria for the differential diagnosis of BCC, and to identify representative RCM parameters for superficial (sBCCs) and nonsuperficial (nsBCCs) basal cell carcinomas (BCCs). METHODS: A retrospective analysis of consecutive patients evaluated with RCM, selecting excised lesions classified at dermoscopy with ≥1 score from the re visited 7-point checklist, mimicking melanoma, registered between 2010 and 2016. Cluster analysis identified BCC subclassifications. RESULTS: Of 178 atypical lesions, 34 lesions were diagnosed as BCCs with RCM. Lesions were confirmed BCCs with histopathology. Dermoscopic features included atypical network (55.9%) and regression structures (35.5%) associated with sBCCs, and an atypical vascular pattern (58.8%) and irregular blotches (58.8%) with nsBCCs. Hierarchical cluster analysis identified 2 clusters: cluster 1 (100% sBCCs) was characterized by the presence of cords connected to the epidermis (90%, p < 0.001), tumor islands located in the epidermis (100%, p < 0.001), smaller vascular diameter (100%, p < 0.001) and solar elastosis (90%, p = 0.017), and cluster 2 (nsBCCs 85%) was defined by the dermic location of tumor islands (87.5%, p < 0.001) with branch-like structures (70.8%, p = 0.007) and surrounding collagen (83.3%, p = 0.012), peripheral palisading (83.3%, p = 0.012) and coiled vascular morphology (79.2%, p < 0.001) with a larger vascular diameter (50%, p < 0.001). CONCLUSIONS: RCM is able to diagnose BCCs mimicking melanoma at dermoscopy and seems able to identify sBCCs and nsBCCs.

A novel cumulative level difference mean based GLDM and modified ABCD features ranked using eigenvector centrality approach for four skin lesion types classification.

BACKGROUND AND OBJECTIVE: Melanoma is one of the major death causes while basal cell carcinoma (BCC) is the utmost incident skin lesion type. At their early stages, medical experts may be confused between both types with benign nevus and pigmented benign keratoses (BKL). This inspired the current study to develop an accurate automated, user-friendly skin lesion identification system. METHODS: The current work targets a novel discrimination technique of four pre-mentioned skin lesion classes. A novel proposed texture feature, named cumulative level-difference mean (CLDM) based on the gray-level difference method (GLDM) is extracted. The asymmetry, border irregularity, color variation and diameter are summed up as the ABCD rule feature vector is originally used to classify the melanoma from benign lesions. The proposed method improved the ABCD rule to also classify BCC and BKL by using the proposed modified-ABCD feature vector. In the modified set of ABCD features, each border feature, such as compact index, fractal dimension, and edge abruptness is considered a separate feature. Then, the composite feature vector having the pre-mentioned features is ranked using the Eigenvector Centrality (ECFS) feature ranking method. The ranked features are then classified by a cubic support vector machine for different numbers of selected features. RESULTS: The proposed CLDM texture features combined with the ranked ABCD features achieved outstanding performance to classify the four targeted classes (melanoma, BCC, nevi and BKL). The results report 100% outstanding performance of the sensitivity, accuracy and specificity per each class compared to other features when using the highest seven ranked features. CONCLUSIONS: The proposed system established that Melanoma, BCC, nevus and BKL are efficiently classified using cubic SVM with the new feature set. In addition, the comparative studies proved the superiority of the cubic SVM to classify the four classes.

Discovering lncRNA mediated sponge interactions in breast cancer molecular subtypes.

BACKGROUND: Long non-coding RNAs (lncRNAs) can indirectly regulate mRNAs expression levels by sequestering microRNAs (miRNAs), and act as competing endogenous RNAs (ceRNAs) or as sponges. Previous studies identified lncRNA-mediated sponge interactions in various cancers including the breast cancer. However, breast cancer subtypes are quite distinct in terms of their molecular profiles; therefore, ceRNAs are expected to be subtype-specific as well. RESULTS: To find lncRNA-mediated ceRNA interactions in breast cancer subtypes, we develop an integrative approach. We conduct partial correlation analysis and kernel independence tests on patient gene expression profiles and further refine the candidate interactions with miRNA target information. We find that although there are sponges common to multiple subtypes, there are also distinct subtype-specific interactions. Functional enrichment of mRNAs that participate in these interactions highlights distinct biological processes for different subtypes. Interestingly, some of the ceRNAs also reside in close proximity in the genome; for example, those involving HOX genes, HOTAIR, miR-196a-1 and miR-196a-2. We also discover subtype-specific sponge interactions with high prognostic potential. We found that patients differ significantly in their survival distributions if they are group based on the expression patterns of specific ceRNA interactions. However, it is not the case if the expression of individual RNAs participating in ceRNA is used. CONCLUSION: These results can help shed light on subtype-specific mechanisms of breast cancer, and the methodology developed herein can help uncover sponges in other diseases.

Sclerodermiform basal cell carcinoma: how much can we rely on dermatoscopy to differentiate from non-aggressive basal cell carcinomas? Analysis of 1256 cases.

BACKGROUND: The behaviour of each basal cell carcinoma is known to be different according to the histological growth pattern. Among these aggressive lesions, sclerodermiform basal cell carcinomas are the most common type. This is a challenging-to-treat lesion due to its deep tissue invasion, rapid growth, risk of metastasis and overall poor prognosis if not diagnosed in early stages. OBJECTIVE: To investigate if sclerodermiform basal cell carcinomas are diagnosed later compared to non-sclerodermiform basal cell carcinoma Method: All lesions excised from 2000 to 2010 were included. A pathologist classified the lesions in two cohorts: one with specimens of non-aggressive basal cell carcinoma (superficial, nodular and pigmented), and other with sclerodermiform basal cell carcinoma. For each lesion, we collected patient's information from digital medical records regarding: gender, age when first attending the clinic and the tumor location. RESULTS: 1256 lesions were included, out of which 296 (23.6%) corresponded to sclerodermiform basal cell carcinoma, whereas 960 (76.4%) were non-aggressive subtypes of basal cell carcinoma. The age of diagnosis was: 72.78±12.31 years for sclerodermiform basal cell and 69.26±13.87 years for non-aggressive basal cell carcinoma (P<.0001). Sclerodermiform basal cell carcinomas are diagnosed on average 3.52 years later than non-aggressive basal cell carcinomas. Sclerodermiform basal cell carcinomas were diagnosed 3.40 years and 2.34 years later than non-aggressive basal cell carcinomas in younger and older patients respectively (P=.002 and P=.03, respectively). STUDY LIMITATIONS: retrospective design. CONCLUSION: The diagnostic accuracy and primary clinic conjecture of sclerodermiform basal cell carcinomas is quite low compared to other forms of basal cell carcinoma such as nodular, superficial and pigmented. The dermoscopic vascular patterns, which is the basis for the diagnosis of non-melanocytic nonpigmented skin tumors, may not be particularly useful in identifying sclerodermiform basal cell carcinomas in early stages. As a distinct entity, sclerodermiform basal cell carcinomas show a lack of early diagnosis compared to less-aggressive subtypes of BCC, and thus, more accurate diagnostic tools apart from dermatoscopy are required to reach the goal of early-stage diagnosis of sclerodermiform basal cell carcinomas.

Sclerodermiform basal cell carcinoma: how much can we rely on dermatoscopy to differentiate from non-aggressive basal cell carcinomas? Analysis of 1256 cases

Abstract: Background: The behaviour of each basal cell carcinoma is known to be different according to the histological growth pattern. Among these aggressive lesions, sclerodermiform basal cell carcinomas are the most common type. This is a challenging-to-treat lesion due to its deep tissue invasion, rapid growth, risk of metastasis and overall poor prognosis if not diagnosed in early stages. Objective: To investigate if sclerodermiform basal cell carcinomas are diagnosed later compared to non-sclerodermiform basal cell carcinoma Method: All lesions excised from 2000 to 2010 were included. A pathologist classified the lesions in two cohorts: one with specimens of non-aggressive basal cell carcinoma (superficial, nodular and pigmented), and other with sclerodermiform basal cell carcinoma. For each lesion, we collected patient's information from digital medical records regarding: gender, age when first attending the clinic and the tumor location. Results: 1256 lesions were included, out of which 296 (23.6%) corresponded to sclerodermiform basal cell carcinoma, whereas 960 (76.4%) were non-aggressive subtypes of basal cell carcinoma. The age of diagnosis was: 72.78±12.31 years for sclerodermiform basal cell and 69.26±13.87 years for non-aggressive basal cell carcinoma (P<.0001). Sclerodermiform basal cell carcinomas are diagnosed on average 3.52 years later than non-aggressive basal cell carcinomas. Sclerodermiform basal cell carcinomas were diagnosed 3.40 years and 2.34 years later than non-aggressive basal cell carcinomas in younger and older patients respectively (P=.002 and P=.03, respectively). Study limitations: retrospective design. Conclusion: The diagnostic accuracy and primary clinic conjecture of sclerodermiform basal cell carcinomas is quite low compared to other forms of basal cell carcinoma such as nodular, superficial and pigmented. The dermoscopic vascular patterns, which is the basis for the diagnosis of non-melanocytic nonpigmented skin tumors, may not be particularly useful in identifying sclerodermiform basal cell carcinomas in early stages. As a distinct entity, sclerodermiform basal cell carcinomas show a lack of early diagnosis compared to less-aggressive subtypes of BCC, and thus, more accurate diagnostic tools apart from dermatoscopy are required to reach the goal of early-stage diagnosis of sclerodermiform basal cell carcinomas.

Correlation Between Incisional Biopsy Histological Subtype and a Mohs Surgery Specimen for Nonmelanoma Skin Cancer.

BACKGROUND: Histological diagnosis of a clinically suspected nonmelanoma skin cancer (NMSC) is recommended before treatment. For NMSC, concordance between the histological subtype of the preoperative biopsy and the excision specimen of basal cell carcinoma (BCC) has been reported to range from 10% to 81%. No large study on the concordance between NMSC histology seen in a preoperative biopsy with the following tumour specimen from Mohs micrographic surgery (MMS) has been performed in a Latin American population. OBJECTIVE: The aim of this study was to analyse and compare the histological subtype of the incisional biopsies reviewed by the dermatopathologist with the histological subtype of the tumour specimen obtained during MMS interpreted by the dermatopathologist and the Mohs surgeon. METHODS: A retrospective analysis of 320 NMSC was performed. The interobserver correlation was based on kappa values. RESULTS: The mean weighted kappa value between the preoperative NMSC biopsy and intraoperative histological subtype of the tumour specimen from MMS analysed by the Mohs surgeon and the dermatopathologist was 0.22 and 0.24, respectively. The correlation in the histologic subtype of the intraoperative tumour specimen from MMS that was interpreted by the dermatopathologist and Mohs surgeon was 0.58. CONCLUSIONS: Dermatologists need to be aware of the limited value of incisional biopsies to accurately diagnose the histological subtype of a NMSC. The concordance rate in the histological diagnosis of the tumour specimens that were obtained from MMS between the Mohs surgeon and the dermatopathologist is moderate. However, the correlation is low compared with incisional biopsy subtypes.

Utilidad de la ecografía cutánea en la clasificación de subtipos de los carcinomas basocelulares primarios / Usefulness of high-frequency ultrasound in the classification of histologic subtypes of primary basal cell carcinoma

INTRODUCCIÓN: La biopsia incisional puede fallar en la correcta catalogación de subtipos histológicos de carcinoma basocelular (CBC). La ecografía (ECO) cutánea es una herramienta diagnóstica útil en el diagnóstico y manejo de este tumor. OBJETIVOS: El objetivo principal fue evaluar la utilidad diagnóstica de la ECO frente a la biopsia punch en la correcta clasificación del patrón histológico de infiltración de los CBC primarios. Los objetivos secundarios fueron: evaluar si el rendimiento diagnóstico de la ECO frente a la biopsia incisional guardaba relación con el tamaño tumoral y con formas de CBC simples frente a formas mixtas. MÉTODOS: Estudio observacional prospectivo de los casos de CBC primarios atendidos en el Servicio de Dermatología del Hospital Costa del Sol (Marbella) entre octubre de 2013 y mayo de 2014. Previamente a la extirpación quirúrgica se realizó una ECO cutánea (Dermascan C©, sonda lineal, 20 Mhz) y una biopsia punch. Se valoró el porcentaje de acuerdo absoluto y rendimiento diagnóstico (sensibilidad, especificidad, valor predictivo positivo [VPP] y valor predictivo negativo [VPN]) para resultados globales y parciales entre ECO y punch frente al gold estándar (biopsia escisional por cortes seriados). RESULTADOS: Se incluyeron 156 casos. La tasa de concordancia diagnóstica global de la ECO fue del 73,7% (sensibilidad: 74,5%, especificidad: 73%) vs. 79,9% (sensibilidad: 76%, especificidad: 82%) para el punch. En el análisis individual destaca para el CBC superficial un VPP para la ECO del 93,3% frente al 92% para el punch. En el análisis por tamaño tumoral la ECO incrementó el porcentaje de acuerdo absoluto del 70,4 al 77,3% (área 40 mm2 vs. área > 40 mm2) manteniendo el VPN constante para ambos subgrupos (82%). Para la biopsia punch, el porcentaje de acuerdo absoluto pasó del 86,4 al 72,6%. CONCLUSIONES: La ECO cutánea muestra una especial utilidad para descartar la presencia de invasividad, para el diagnóstico de formas superficiales simples y para la correcta catalogación de CBC de área mayor a 40 mm2

INTRODUCTION: Incisional biopsy may not always provide a correct classification of histologic subtypes of basal cell carcinoma (BCC). High-frequency ultrasound (HFUS) imaging of the skin is useful for the diagnosis and management of this tumor. OBJECTIVES: The main aim of this study was to compare the diagnostic value of HFUS compared with punch biopsy for the correct classification of histologic subtypes of primary BCC. We also analyzed the influence of tumor size and histologic subtype (single subtype vs. mixed) on the diagnostic yield of HFUS and punch biopsy. METHODS: Retrospective observational study of primary BCCs treated by the Dermatology Department of Hospital Costa del Sol in Marbella, Spain, between october 2013 and may 2014. Surgical excision was preceded by HFUS imaging (Dermascan C©, 20-MHz linear probe) and a punch biopsy in all cases. We compared the overall diagnostic yield and accuracy (sensitivity, specificity, positive predictive value [PPV], and negative predictive value [NPV]) of HFUS and punch biopsy against the gold standard (excisional biopsy with serial sections) for overall and subgroup results. RESULTS: We studied 156 cases. The overall diagnostic yield was 73.7% for HFUS (sensitivity, 74.5%; specificity, 73%) and 79.9% for punch biopsy (sensitivity, 76%; specificity, 82%). In the subgroup analyses, HFUS had a PPV of 93.3% for superficial BCC (vs. 92% for punch biopsy). In the analysis by tumor size, HFUS achieved an overall diagnostic yield of 70.4% for tumors measuring 40 mm2 or less and 77.3% for larger tumors; the NPV was 82% in both size groups. Punch biopsy performed better in the diagnosis of small lesions (overall diagnostic yield of 86.4% for lesions 40 mm2 vs. 72.6% for lesions > 40 mm2). CONCLUSIONS: HFUS imaging was particularly useful for ruling out infiltrating BCCs, diagnosing simple, superficial BCCs, and correctly classifying BCCs larger than 40 mm2

Thresholding methods for lesion segmentation of basal cell carcinoma in dermoscopy images.

PURPOSE: Algorithms employed for pigmented lesion segmentation perform poorly on dermoscopy images of basal cell carcinoma (BCC), the most common skin cancer. The main objective was to develop better methods for BCC segmentation. METHODS: Fifteen thresholding methods were implemented for BCC lesion segmentation. We propose two error metrics that better measure the type II error: Relative XOR Error and Lesion Capture Ratio. RESULTS: On training/test sets of 305 and 34 BCC images, respectively, five new techniques outperform two state-of-the-art methods used in segmentation of melanomas, based on the new error metrics. CONCLUSION: The proposed algorithms, which include solutions for image vignetting correction and border expansion to achieve dermatologist-like borders, provide more inclusive and feature-preserving border detection, favoring better BCC classification accuracy, in future work.