Search

The clinical differentiation of face and scalp lesions (FSLs) is challenging even for trained dermatologists. Studies comparing the diagnostic performance of a convolutional neural network (CNN) with dermatologists in FSL are lacking.A market-approved CNN (Moleanalyzer-Pro, FotoFinder Systems) was used for binary classifications of 100 dermoscopic images of FSL. The same lesions were used in a two-level reader study including 64 dermatologists (level I: dermoscopy only; level II: dermoscopy, clinical close-up images, textual information). Primary endpoints were the CNN's sensitivity and specificity in comparison with the dermatologists' management decisions in level II. Generalizability of the CNN results was tested by using four additional external data sets.The CNN's sensitivity, specificity and ROC AUC were 96.2% [87.0%-98.9%], 68.8% [54.7%-80.1%] and 0.929 [0.880-0.978], respectively. In level II, the dermatologists' management decisions showed a mean sensitivity of 84.2% [82.2%-86.2%] and specificity of 69.4% [66.0%-72.8%]. When fixing the CNN's specificity at the dermatologists' mean specificity (69.4%), the CNN's sensitivity (96.2% [87.0%-98.9%]) was significantly higher than that of dermatologists (84.2% [82.2%-86.2%]; p 0.001). Dermatologists of all training levels were outperformed by the CNN (all p 0.001). In confirmation, the CNN's accuracy (83.0%) was significantly higher than dermatologists' accuracies in level II management decisions (all p 0.001). The CNN's performance was largely confirmed in three additional external data sets but particularly showed a reduced specificity in one Australian data set including FSL on severely sun-damaged skin.When applied as an assistant system, the CNN's higher sensitivity at an equivalent specificity may result in an improved early detection of face and scalp skin cancers.

In Deutschland leiden schatzungsweise 1-2% der Bevolkerung unter einer chronischen Wunde. Hierbei gilt es nicht nur an die haufigen Entitaten, sondern auch an seltenere ulzerierende Erkrankungen zu denken. Ein grundlegendes Verstandnis der Pathophysiologie bezuglich der Wundentstehung und Wundheilung hilft den Wundzustand richtig einschatzen zu konnen, um erforderliche diagnostische Schritte und eine phasengerechte Therapie einzuleiten. Die ABCDE-Regel stellt hierbei einen praktischen Leitfaden fur die Wunddiagnostik dar. Bei der klinischen Untersuchung interpretiert der Wundtherapeut entsprechend des Wunddreiecks den Wundgrund, den Wundrand und die Wundumgebung. Dieser Ansatz kann praktische therapeutische Entscheidungen, z.B. die Wahl der Wundauflage, vereinfachen. Nicht immer ist ein kausaler Therapieansatz moglich oder ausreichend. Vielmehr muss auf begleitende Faktoren geachtet und ein ganzheitliches und interdisziplinares Konzept erarbeitet werden. Hierbei kommt dem Hausarzt eine zentrale Schlusselrolle zu.

Background Birt-Hogg-Dube syndrome (BHDS) is a genetic tumor syndrome characterized by lung cysts, spontaneous pneumothorax, fibrofolliculomas, and renal cell cancer. Because of its rarity and clinical heterogeneity, much is still unknown regarding the course of the disease and individual risk assessment. Therefore, we studied nonenvironmental risk factors for pneumothorax in a large sample of patients with BHDS. Methods Clinical data were available from 197 patients with BHDS (male patients, 103; female patients, 94) belonging to 63 unrelated families. The FLCN coding region including adjacent intronic sequences was analyzed by PCR and subsequent Sanger sequencing as well as by multiplex ligation-dependent probe amplification. Statistical analyses were performed, using adequate methods to account for familial clustering. Results Patients who had only a single spontaneous pneumothorax were significantly older at the time of occurrence than those with multiple pneumothoraces (mean, 38.93 vs 29.74 years; P value, .010). The risk for three or more pneumothoraces drastically increased after the second event. Significantly increased pneumothorax risks were found for mutations c.1300G>C (59%) and c.250-2A>G (77%), compared with FLCN hotspot mutation c.1285dup (37% risk) (P value, .02). Conclusions We observed significant differences for the spontaneous pneumothorax risk regarding both age and sex in patients with BHDS. Furthermore, two FLCN mutations were identified that are associated with significantly increased pneumothorax risk. Thus, formerly unknown individual predictors have been identified that provide improved risk stratification for patients with BHDS.

Background In recent studies, convolutional neural networks (CNNs) outperformed dermatologists in distinguishing dermoscopic images of melanoma and nevi. In these studies, dermatologists and artificial intelligence were considered as opponents. However, the combination of classifiers frequently yields superior results, both in machine learning and among humans. In this study, we investigated the potential benefit of combining human and artificial intelligence for skin cancer classification. Methods Using 11,444 dermoscopic images, which were divided into five diagnostic categories, novel deep learning techniques were used to train a single CNN. Then, both 112 dermatologists of 13 German university hospitals and the trained CNN independently classified a set of 300 biopsy-verified skin lesions into those five classes. Taking into account the certainty of the decisions, the two independently determined diagnoses were combined to a new classifier with the help of a gradient boosting method. The primary end-point of the study was the correct classification of the images into five designated categories, whereas the secondary end-point was the correct classification of lesions as either benign or malignant (binary classification). Findings Regarding the multiclass task, the combination of man and machine achieved an accuracy of 82.95%. This was 1.36% higher than the best of the two individual classifiers (81.59% achieved by the CNN). Owing to the class imbalance in the binary problem, sensitivity, but not accuracy, was examined and demonstrated to be superior (89%) to the best individual classifier (CNN with 86.1%). The specificity in the combined classifier decreased from 89.2% to 84%. However, at an equal sensitivity of 89%, the CNN achieved a specificity of only 81.5% Interpretation Our findings indicate that the combination of human and artificial intelligence achieves superior results over the independent results of both of these systems.

Die Shiitake-Dermatitis tritt ca. einen bis 2 Tage nach dem Verzehr roher oder unzureichend gekochter Shiitake-Pilze auf und ist durch striar angeordnete, juckende, erythematose Papulovesikel gekennzeichnet. Ursachlich ist Lentinan, ein in der Zellwand der Shiitake-Pilze vorkommendes Polysaccharid. In der histologischen Untersuchung stellt sich ein ekzemartiges Bild mit Spongiose und lymphohistiozytarem Infiltrat dar. Die Ergebnisse der konfokalen Lasermikroskopie (KLM) korrelieren mit den Aspekten der histologischen Untersuchung; deshalb konnte die KLM zukunftig als nichtinvasives Verfahren zur Diagnostik einer Shiitake-Dermatitis dienen.

Ex-vivo fluorescence confocal microscopy (FCM) has been used on fresh tissue, but there is little experience on frozen sections. We evaluated the applicability of FCM on frozen sections of basal cell carcinomas (BCCs), stained with acridine orange and digitally colored to simulate hematoxylin and eosin (H&E) dyes. We compared our diagnostic accuracy in detecting and subtyping BCCs with FCM to our gold standard (H&E stained frozen sections used in 3D horizontal micrographic surgery). Fourty-six primary BCCs were analyzed for free margins as well as histological subtype with all FCM modes and conventional H&E staining. Adnexa, artifacts and diagnostic confidence were evaluated. Free margins were identified with a sensitivity and specificity of 92% and 91%. Concordance for tumor subtype was 88%. FCM may be used on both fresh tissue and frozen samples, although with reduced performance and different artifacts. The device is useful for the intraoperative diagnosis, subtyping and margin-mapping of BCCs.

Objective Birt-Hogg-Dube syndrome (BHDS) is an inherited tumour syndrome characterised by three major symptoms: lung cysts with spontaneous pneumothorax, fibrofolliculoma and renal cell cancer. The first family with this syndrome was described in 1975 and one of its members presented with adenomatous colon polyps and colorectal cancer. Since then, it has been a matter of debate whether colorectal cancer is indeed part of the BHDS spectrum and if regular screening should be recommended. Design We analysed the frequency of colorectal cancer in a large sample of BHDS families. Clinical data were available from 256 BHDS patients (male 130, female 126) belonging to 83 unrelated families. For controls, 83 index patients who attended our outpatient clinic for non–malignancy-related genetic counselling and their family members (total of 519 controls) were used. Results The patients with BHDS showed a moderately but significantly increased rate of colorectal cancer (5.1% versus 1.5%, p-value .0068). Unexpectedly, 35% of patients with colorectal cancer corresponding to eight of 82 BHDS families fulfilled the revised Bethesda criteria for HNPCC, either because colorectal cancer occurred before age 50 years or because three family members were affected by colorectal cancer. Apart from colorectal cancer, no other HNPCC-associated tumours occurred within the Bethesda criteria–positive families, an observation that argues against a concurrence of BHDS and HNPCC in these families. Conclusion The results suggest that BHDS is associated with early-onset colorectal cancer, a hypothesis that might have a marked impact on preventive screening recommendations.

Simple Summary Actinic keratoses (AKs) are extremely common in the elderly population; they are universally recognized as precursors of invasive squamous cell carcinoma, and their risk of progression relates to the basal growth pattern of keratinocytes in the histological slides, based on a model called “PRO” classification. Since AKs can be investigated at bedside with non-invasive devices such the new line-field confocal optical coherence tomography (LC-OCT), we hypothesized that it was also possible to use such devices for reproducing the PRO classification and assessing the progression risk of AK without an invasive biopsy. In this pilot study, we demonstrated the feasibility of the LC-OCT severity grading of AKs based on the histological PRO classification, obtaining a good correlation between the two models and strong interobserver agreement. Abstract It is known that actinic keratoses (AKs) can progress to invasive squamous cell carcinoma (SCC). The histological PRO grading of AKs is based on the growth pattern of basal keratinocytes and relates to their progression risk. AKs can be non-invasively characterized by line-field confocal optical coherence tomography (LC-OCT). The aim of the study was to define criteria for an LC-OCT grading of AKs based on the PRO classification and to correlate it with its histological counterpart. To evaluate the interobserver agreement for the LC-OCT PRO classification, fifty AKs were imaged by LC-OCT and biopsied for histopathology. PRO histological grading was assessed by an expert consensus, while two evaluator groups separately performed LC-OCT grading on vertical sections. The agreement between LC-OCT and histological PRO grading was 75% for all lesions (weighted kappa 0.66, 95% CI 0.48–0.83, p ≤ 0.001) and 85.4% when comparing the subgroups PRO I vs. PRO II/III (weighted kappa 0.64, 95% CI 0.40–0.88, p ≤ 0.001). The interobserver agreement for LC-OCT was 90% (Cohen’s kappa 0.84, 95% CI 0.71–0.91, p ≤ 0.001). In this pilot study, we demonstrated that LC-OCT is potentially able to classify AKs based on the basal growth pattern of keratinocytes, in-vivo reproducing the PRO classification, with strong interobserver agreement and a good correlation with histopathology.

Skin is our barrier against environmental damage. Moisturizers are widely used to increase hydration and barrier integrity of the skin; however, there are contrasting observations on their in vivo effects in real-life settings. In cosmetic studies, corneometers and tewameters are traditionally used to assess skin hydration. In this study, two novel noninvasive diagnostic techniques, optical coherence tomography (OCT) and confocal Raman spectroscopy, were used to analyze stratum corneum and epidermal thickness (ET), water content, blood flow in function of depth, skin roughness, attenuation coefficient, natural moisturizing factor, ceramides and free fatty acids, cholesterol, urea, and lactates in 20 female subjects aged between 30 and 45 before and after 2 weeks application of a commercially available moisturizing lotion on one forearm. The untreated forearm served as control. A third measurement was conducted 1 week after cessation of moisturizing to verify whether the changes in the analyzed parameters persisted. We noticed a reduction in skin roughness, an increase in ceramides and free fatty acids and a not statistically significant increase in ET. As a conclusion, short time moisturizing appears insufficient to provide significant changes in skin morphology and composition, as assessed by OCT and RS. Novel noninvasive imaging methods are suitable for the evaluation of skin response to topical moisturizers. Further studies on larger sample size and longer treatment schedules are needed to analyze changes under treatment with moisturizers and to standardize the use of novel noninvasive diagnostic techniques.

Optical coherence tomography (OCT) has been able to establish itself in recent years not only in academic-scientific, but also in everyday dermatological practice. Its focus lies on epithelial tumors of the skin, which can be diagnosed intuitively and within a few seconds. Thus, basal cell carcinomas, actinic keratoses, and different stages of field cancerization can be diagnosed and monitored for response to therapy or possible recurrence. This often helps to avoid invasive sample extraction. Recently, the field of OCT and its latest advancement, dynamic OCT (D-OCT), has been expanded to include non-oncologic dermatological diseases. This encompasses inflammatory dermatoses and the analysis of physiological skin parameters such as hydration. Thanks to automated vascular imaging and the measurement of objective parameters such as epidermal thickness, blood flow at depth, optical attenuation coefficient, and skin roughness, more and more characteristics of the skin can be studied in a noninvasive and standardized way. New potential areas of application are eczema, contact allergic dermatitis, psoriasis, rosacea, telangiectasia, acute and chronic wounds, melasma and nevus flammeus but also melanocytic lesions.

Optical coherence tomography (OCT) and confocal laser microscopy (CLSM) are established non-invasive methods in clinical dermatological routine diagnosis. Whereas CLSM is especially useful to distinguish between nevi and melanoma, OCT is suitable for the diagnosis and differentiation of non-melanoma skin cancer. Line-field confocal optical coherence tomography (LC-OCT) is a new innovative device, which has better cellular resolution than OCT and a higher penetration depth than CLSM. Similar to CLSM, LC-OCT also allows 3D images in real time to be taken. Therefore LC-OCT is very useful for the examination of skin lesions of all kinds, since it unites the features of CLSM and OCT.

Background The treatment of keratinocyte cancers (KC) strictly depends on their differentiation and invasiveness. Non-invasive diagnostic techniques can support the diagnosis in real time, avoiding unnecessary biopsies. This study aimed to preliminarily define main imaging criteria and histological correlations of actinic keratosis (AK), Bowen's disease (BD) and squamous cell carcinoma (SCC) using the novel device line-field confocal optical coherence tomography (LC-OCT). Methods Dermoscopy and LC-OCT images of 73 histopathologically confirmed lesions (46 AKs, 11 BD and 16 SCCs) were included in the study. Exemplary lesions (10 AKs, 5 BD and 5 SCCs) were additionally investigated with optical coherence tomography and reflectance confocal microscopy. Results Most common LC-OCT findings of KC in the descriptive statistics were hyperkeratosis/parakeratosis, disruption of stratum corneum, broadened epidermis, basal and suprabasal keratinocyte atypia, dilated vessels/neoangiogenesis and elastosis/collagen alterations. In the univariate multinomial logistic regression, a preserved DEJ was less common in SCC compared with AK and BD, BD displayed marked keratinocyte atypia involving all epidermal layers (bowenoid pattern), while SCC showed ulceration, increased epidermal thickness, keratin plugs, acantholysis, not visible/interrupted DEJ and epidermal bright particles. LC-OCT increased the diagnostic confidence by 24.7% compared with dermoscopy alone. Conclusions Our study describes for the first time specific LC-OCT features of different stages of KC and their histopathological correlates, focusing on keratinocyte morphology and architecture of the epidermis and DEJ. LC-OCT may open new scenarios in the bedside diagnosis, treatment planning and follow-up of KC.

Background Early detection of melanoma can be lifesaving but this remains a challenge. Recent diagnostic studies have revealed the superiority of artificial intelligence (AI) in classifying dermoscopic images of melanoma and nevi, concluding that these algorithms should assist a dermatologist’s diagnoses. Objective The aim of this study was to investigate whether AI support improves the accuracy and overall diagnostic performance of dermatologists in the dichotomous image–based discrimination between melanoma and nevus. Methods Twelve board-certified dermatologists were presented disjoint sets of 100 unique dermoscopic images of melanomas and nevi (total of 1200 unique images), and they had to classify the images based on personal experience alone (part I) and with the support of a trained convolutional neural network (CNN, part II). Additionally, dermatologists were asked to rate their confidence in their final decision for each image. Results While the mean specificity of the dermatologists based on personal experience alone remained almost unchanged (70.6% vs 72.4%; P=.54) with AI support, the mean sensitivity and mean accuracy increased significantly (59.4% vs 74.6%; P=.003 and 65.0% vs 73.6%; P=.002, respectively) with AI support. Out of the 10% (10/94; 95% CI 8.4%-11.8%) of cases where dermatologists were correct and AI was incorrect, dermatologists on average changed to the incorrect answer for 39% (4/10; 95% CI 23.2%-55.6%) of cases. When dermatologists were incorrect and AI was correct (25/94, 27%; 95% CI 24.0%-30.1%), dermatologists changed their answers to the correct answer for 46% (11/25; 95% CI 33.1%-58.4%) of cases. Additionally, the dermatologists’ average confidence in their decisions increased when the CNN confirmed their decision and decreased when the CNN disagreed, even when the dermatologists were correct. Reported values are based on the mean of all participants. Whenever absolute values are shown, the denominator and numerator are approximations as every dermatologist ended up rating a varying number of images due to a quality control step. Conclusions The findings of our study show that AI support can improve the overall accuracy of the dermatologists in the dichotomous image–based discrimination between melanoma and nevus. This supports the argument for AI-based tools to aid clinicians in skin lesion classification and provides a rationale for studies of such classifiers in real-life settings, wherein clinicians can integrate additional information such as patient age and medical history into their decisions.

BACKGROUND The diagnosis of allergic contact dermatitis should be confirmed by skin patch tests. Distinguishing between irritant and allergic reactions is sometimes difficult. OBJECTIVES To analyse the in vivo morphological changes in patch test reactions compared to healthy skin, and to detect subclinical changes in doubtful reactions using optical coherence tomography (OCT). To develop an OCT-based algorithm to support patch-test grading. METHODS One hundred twenty-nine skin patch-test areas were scanned with OCT to evaluate the following features: architectural and vascular morphology, epidermal thickness, optical attenuation coefficient (AC), and blood flow at 0.1, 0.2, and 0.35 mm depth. RESULTS Most common OCT features of acute contact allergic reactions in patch tests were spongiosis with microvesicles (94.8%), macrovesicles (60.3%), and coalescing vesicles (46.6%), the latter useful in differentiating acute allergic from irritant dermatitis (P-value < .05). Objective quantitative parameters correlated well with the severity grade: epidermal thickness due to spongiosis, AC (P-value < .05) and blood flow at 0.2 and 0.35 mm (P-value < .01). CONCLUSIONS OCT as a noninvasive diagnostic tool, established for skin cancer diagnosis, is useful for evaluating contact allergic patch-test reactions. Not only morphological but also objective features such as blood flow and AC correlate with the reaction severity. Further studies are needed to explore the differences in irritant and allergic contact dermatitis.

BACKGROUND Early detection of melanoma can be lifesaving but this remains a challenge. Recent diagnostic studies have revealed the superiority of artificial intelligence (AI) in classifying dermoscopic images of melanoma and nevi, concluding that these algorithms should assist a dermatologist’s diagnoses. OBJECTIVE The aim of this study was to investigate whether AI support improves the accuracy and overall diagnostic performance of dermatologists in the dichotomous image–based discrimination between melanoma and nevus. METHODS Twelve board-certified dermatologists were presented disjoint sets of 100 unique dermoscopic images of melanomas and nevi (total of 1200 unique images), and they had to classify the images based on personal experience alone (part I) and with the support of a trained convolutional neural network (CNN, part II). Additionally, dermatologists were asked to rate their confidence in their final decision for each image. RESULTS While the mean specificity of the dermatologists based on personal experience alone remained almost unchanged (70.6% vs 72.4%; P=.54) with AI support, the mean sensitivity and mean accuracy increased significantly (59.4% vs 74.6%; P=.003 and 65.0% vs 73.6%; P=.002, respectively) with AI support. Out of the 10% (10/94; 95% CI 8.4%-11.8%) of cases where dermatologists were correct and AI was incorrect, dermatologists on average changed to the incorrect answer for 39% (4/10; 95% CI 23.2%-55.6%) of cases. When dermatologists were incorrect and AI was correct (25/94, 27%; 95% CI 24.0%-30.1%), dermatologists changed their answers to the correct answer for 46% (11/25; 95% CI 33.1%-58.4%) of cases. Additionally, the dermatologists’ average confidence in their decisions increased when the CNN confirmed their decision and decreased when the CNN disagreed, even when the dermatologists were correct. Reported values are based on the mean of all participants. Whenever absolute values are shown, the denominator and numerator are approximations as every dermatologist ended up rating a varying number of images due to a quality control step. CONCLUSIONS The findings of our study show that AI support can improve the overall accuracy of the dermatologists in the dichotomous image–based discrimination between melanoma and nevus. This supports the argument for AI-based tools to aid clinicians in skin lesion classification and provides a rationale for studies of such classifiers in real-life settings, wherein clinicians can integrate additional information such as patient age and medical history into their decisions.

Birt-Hogg-Dube syndrome (BHD, also referred to as Hornstein-Knickenberg syndrome) is an autosomal dominant tumor syndrome caused by mutations in the FLCN gene located on chromosome 17. Depending on their age, patients with BHD may exhibit various clinical signs and symptoms. Disease severity can vary greatly among members of the same family. Early symptoms include basal lung cysts, which can lead to recurrent spontaneous pneumothoraces. The majority of patients (> 90 %) develop multiple fibrofolliculomas, especially on the face and upper trunk, in the second or third decade of life. Given the 12-34 % lifetime risk of developing benign or malignant renal tumors, targeted screening programs are prognostically crucial. While these renal tumors may belong to various histological subtypes, common variants include multifocal - sometimes bilateral - chromophobe and oncocytic hybrid tumors. Early diagnosis and adequate long-term care of families with BHD require interdisciplinary cooperation.

Das Birt-Hogg-Dube-Syndrom (BHD-Syndrom, eigentlich Hornstein-Knickenberg- Syndrom) ist ein autosomal dominant erbliches Tumorsyndrom, welches durch Mutationen im FLCN-Gen auf Chromosom 17 verursacht wird. Patienten mit BHD-Syndrom konnen altersabhangig verschiedene Symptome zeigen, deren Auspragung auch innerhalb einer Familie unterschiedlich schwer sein kann. Ein fruhes Symptom sind basal betonte Lungenzysten, welche Ursache wiederholter Spontanpneumothoraces sein konnen. Die Mehrheit der Patienten (> 90 %) entwickelt im mittleren Lebensalter zahlreiche Fibrofollikulome vor allem im Gesicht und am Oberkorper. Fur die Prognose entscheidend ist eine gezielte Tumorvorsorge, da ein Lebenszeitrisiko von 12-34 % fur benigne und maligne Nierentumoren besteht. Die Nierentumoren beim BHD-Syndrom konnen verschiedenen histologischen Subgruppen angehoren, wobei multifokale, auch bilaterale Hybridtumoren mit chromophoben und onkozytaren Anteilen haufig sind. Die fruhzeitige Diagnosestellung ebenso wie die langfristige Betreuung von Familien mit BHD-Syndrom erfordern eine interdisziplinare Zusammenarbeit.

Frontal fibrosing alopecia (FFA) describes the scarring, band-like recession of the frontotemporal hairline. Treatment is difficult, and currently, no evidence-based therapy exists. The purpose of this study is to report clinical features and treatment responses in a large cohort of patients with FFA. The authors analyzed a series of 72 patients with a clinical or histologic diagnosis of FFA. A total of 70 patients were female (97.2%), and 2 were male (2.8%). In females, the first onset of FFA was postmenopausal in 81.4% (n = 57). Documented eyebrow loss was present in 61.1% (n = 44), whereas involvement of eyelashes and body hair was reported in only 4.2% (n = 3) and 5.6% (n = 4), respectively. Clinical symptoms were itching (40.3%, n = 29) and trichodynia (4.2%, n = 3) in the alopecic area. Virtually all patients were treated with topical high-potency steroids. Additional treatments were topical tacrolimus, systemic retinoids, and hydroxychloroquine. A total of 48 patients (66.7%) received a combination of high-potency steroids with topical pimecrolimus. In this subgroup, subjective improvement or disease stabilization was reported by 64.6% (n = 31), and the hairline was stabilized on average after 9 to 12 months of therapy. The combination therapy of topical high-potency steroids with pimecrolimus may be an effective and steroid-saving treatment for FFA.

Germline mutations in the Folliculin (FLCN) tumour suppressor gene result in fibrofolliculomas, lung cysts and renal cancers, but the precise mechanisms of tumour suppression by FLCN remain elusive. Here we identify Rab7A, a small GTPase important for endocytic trafficking, as a novel FLCN interacting protein and demonstrate that FLCN acts as a Rab7A GTPase-activating protein. FLCN−/− cells display slower trafficking of epidermal growth factor receptors (EGFR) from early to late endosomes and enhanced activation of EGFR signalling upon ligand stimulation. Reintroduction of wild-type FLCN, but not tumour-associated FLCN mutants, suppresses EGFR signalling in a Rab7A-dependent manner. EGFR signalling is elevated in FLCN−/− tumours and the EGFR inhibitor afatinib suppresses the growth of human FLCN−/− cells as tumour xenografts. The functional interaction between FLCN and Rab7A appears conserved across species. Our work highlights a mechanism explaining, at least in part, the tumour suppressor function of FLCN., Folliculin is a known tumour suppressor but the molecular mechanisms behind this function are unclear. Here the authors show that Folliculin regulates EGFR signalling by modulating its Rab7a-dependent trafficking.

Convolutional neural networks (CNNs) efficiently differentiate skin lesions by image analysis. Studies comparing a market-approved CNN in a broad range of diagnoses to dermatologists working under less artificial conditions are lacking. One hundred cases of pigmented/non-pigmented skin cancers and benign lesions were used for a two-level reader study in 96 dermatologists (level I: dermoscopy only; level II: clinical close-up images, dermoscopy, and textual information). Additionally, dermoscopic images were classified by a CNN approved for the European market as a medical device (Moleanalyzer Pro, FotoFinder Systems, Bad Birnbach, Germany). Primary endpoints were the sensitivity and specificity of the CNN's dichotomous classification in comparison with the dermatologists' management decisions. Secondary endpoints included the dermatologists' diagnostic decisions, their performance according to their level of experience, and the CNN's area under the curve (AUC) of receiver operating characteristics (ROC). The CNN revealed a sensitivity, specificity, and ROC AUC with corresponding 95% confidence intervals (CI) of 95.0% (95% CI 83.5% to 98.6%), 76.7% (95% CI 64.6% to 85.6%), and 0.918 (95% CI 0.866-0.970), respectively. In level I, the dermatologists' management decisions showed a mean sensitivity and specificity of 89.0% (95% CI 87.4% to 90.6%) and 80.7% (95% CI 78.8% to 82.6%). With level II information, the sensitivity significantly improved to 94.1% (95% CI 93.1% to 95.1%; P < 0.001), while the specificity remained unchanged at 80.4% (95% CI 78.4% to 82.4%; P = 0.97). When fixing the CNN's specificity at the mean specificity of the dermatologists' management decision in level II (80.4%), the CNN's sensitivity was almost equal to that of human raters, at 95% (95% CI 83.5% to 98.6%) versus 94.1% (95% CI 93.1% to 95.1%); P = 0.1. In contrast, dermatologists were outperformed by the CNN in their level I management decisions and level I and II diagnostic decisions. More experienced dermatologists frequently surpassed the CNN's performance. Under less artificial conditions and in a broader spectrum of diagnoses, the CNN and most dermatologists performed on the same level. Dermatologists are trained to integrate information from a range of sources rendering comparative studies that are solely based on one single case image inadequate.

Background: Recently, convolutional neural networks (CNNs) systematically outperformed dermatologists in distinguishing dermoscopic melanoma and nevi images. However, such a binary classification does not reflect the clinical reality of skin cancer screenings in which multiple diagnoses need to be taken into account. Methods: Using 11,444 dermoscopic images, which covered dermatologic diagnoses comprising the majority of commonly pigmented skin lesions commonly faced in skin cancer screenings, a CNN was trained through novel deep learning techniques. A test set of 300 biopsy-verified images was used to compare the classifier's performance with that of 112 dermatologists from 13 German university hospitals. The primary end-point was the correct classification of the different lesions into benign and malignant. The secondary end-point was the correct classification of the images into one of the five diagnostic categories. Findings: Sensitivity and specificity of dermatologists for the primary end-point were 74.4% (95% confidence interval [CI]: 67.0–81.8%) and 59.8% (95% CI: 49.8–69.8%), respectively. At equal sensitivity, the algorithm achieved a specificity of 91.3% (95% CI: 85.5–97.1%). For the secondary end-point, the mean sensitivity and specificity of the dermatologists were at 56.5% (95% CI: 42.8–70.2%) and 89.2% (95% CI: 85.0–93.3%), respectively. At equal sensitivity, the algorithm achieved a specificity of 98.8%. Two-sided McNemar tests revealed significance for the primary end-point (p < 0.001). For the secondary end-point, outperformance (p < 0.001) was achieved except for basal cell carcinoma (on-par performance). Interpretation: Our findings show that automated classification of dermoscopic melanoma and nevi images is extendable to a multiclass classification problem, thus better reflecting clinical differential diagnoses, while still outperforming dermatologists at a significant level (p < 0.001).

Background Recent studies have successfully demonstrated the use of deep-learning algorithms for dermatologist-level classification of suspicious lesions by the use of excessive proprietary image databases and limited numbers of dermatologists. For the first time, the performance of a deep-learning algorithm trained by open-source images exclusively is compared to a large number of dermatologists covering all levels within the clinical hierarchy. Methods We used methods from enhanced deep learning to train a convolutional neural network (CNN) with 12,378 open-source dermoscopic images. We used 100 images to compare the performance of the CNN to that of the 157 dermatologists from 12 university hospitals in Germany. Outperformance of dermatologists by the deep neural network was measured in terms of sensitivity, specificity and receiver operating characteristics. Findings The mean sensitivity and specificity achieved by the dermatologists with dermoscopic images was 74.1% (range 40.0%–100%) and 60% (range 21.3%–91.3%), respectively. At a mean sensitivity of 74.1%, the CNN exhibited a mean specificity of 86.5% (range 70.8%–91.3%). At a mean specificity of 60%, a mean sensitivity of 87.5% (range 80%–95%) was achieved by our algorithm. Among the dermatologists, the chief physicians showed the highest mean specificity of 69.2% at a mean sensitivity of 73.3%. With the same high specificity of 69.2%, the CNN had a mean sensitivity of 84.5%. Interpretation A CNN trained by open-source images exclusively outperformed 136 of the 157 dermatologists and all the different levels of experience (from junior to chief physicians) in terms of average specificity and sensitivity.

Background: Several recent publications have demonstrated the use of convolutional neural networks to classify images of melanoma at par with board-certified dermatologists. However, the non-availability of a public human benchmark restricts the comparability of the performance of these algorithms and thereby the technical progress in this field. Methods: An electronic questionnaire was sent to dermatologists at 12 German university hospitals. Each questionnaire comprised 100 dermoscopic and 100 clinical images (80 nevi images and 20 biopsy-verified melanoma images, each), all open-source. The questionnaire recorded factors such as the years of experience in dermatology, performed skin checks, age, sex and the rank within the university hospital or the status as resident physician. For each image, the dermatologists were asked to provide a management decision (treat/biopsy lesion or reassure the patient). Main outcome measures were sensitivity, specificity and the receiver operating characteristics (ROC). Results: Total 157 dermatologists assessed all 100 dermoscopic images with an overall sensitivity of 74.1%, specificity of 60.0% and an ROC of 0.67 (range = 0.538–0.769); 145 dermatologists assessed all 100 clinical images with an overall sensitivity of 89.4%, specificity of 64.4% and an ROC of 0.769 (range = 0.613–0.9). Results between test-sets were significantly different (P < 0.05) confirming the need for a standardised benchmark. Conclusions: We present the first public melanoma classification benchmark for both non-dermoscopic and dermoscopic images for comparing artificial intelligence algorithms with diagnostic performance of 145 or 157 dermatologists. Melanoma Classification Benchmark should be considered as a reference standard for white-skinned Western populations in the field of binary algorithmic melanoma classification.

Background: Recent studies have demonstrated the use of convolutional neural networks (CNNs) to classify images of melanoma with accuracies comparable to those achieved by board-certified dermatologists. However, the performance of a CNN exclusively trained with dermoscopic images in a clinical image classification task in direct competition with a large number of dermatologists has not been measured to date. This study compares the performance of a convolutional neuronal network trained with dermoscopic images exclusively for identifying melanoma in clinical photographs with the manual grading of the same images by dermatologists. Methods: We compared automatic digital melanoma classification with the performance of 145 dermatologists of 12 German university hospitals. We used methods from enhanced deep learning to train a CNN with 12,378 open-source dermoscopic images. We used 100 clinical images to compare the performance of the CNN to that of the dermatologists. Dermatologists were compared with the deep neural network in terms of sensitivity, specificity and receiver operating characteristics. Findings: The mean sensitivity and specificity achieved by the dermatologists with clinical images was 89.4% (range: 55.0%-100%) and 64.4% (range: 22.5%-92.5%). At the same sensitivity, the CNN exhibited a mean specificity of 68.2% (range 47.5%-86.25%). Among the dermatologists, the attendings showed the highest mean sensitivity of 92.8% at a mean specificity of 57.7%. With the same high sensitivity of 92.8%, the CNN had a mean specificity of 61.1%. Interpretation: For the first time, dermatologist-level image classification was achieved on a clinical image classification task without training on clinical images. The CNN had a smaller variance of results indicating a higher robustness of computer vision compared with human assessment for dermatologic image classification tasks. (C) 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Birt-Hogg-Dubé syndrome (BHDS) is a genetic tumor syndrome characterized by lung cysts, pneumothorax, fibrofolliculomas and renal cell cancer. The diagnosis of BHDS is usually considered if kidney cancer occurs before age 50 years, is multifocal and/or bilateral or of the oncocytoma/hybrid oncocytoma-chromophobe type. Using a sample of 50 BHDS families with a total of 178 patients we analyzed how many kidney cancer patients fulfilled one or more of these criteria. Furthermore, we addressed the question if genotype-phenotype-correlations exist that can be used for risk stratification. Renal cell cancer occurred in 34/178 (19.1%) patients, and the reported male bias was not observed. Furthermore, most kidney malignancies occurred after the age of 50 years. Thus, the majority of tumors did not show the typical hallmarks of BHDS. A below-average tumor frequency (17.2%) was observed for the known mutational hotspot c.1285delC/dupC that was the cause of BHDS in 24% of families. Unexpected was the high tumor frequency (66.7%) associated with mutation c.887C>G within a single family, a finding that merits further exploration.

Trotz der einfachen visuellen Zuganglichkeit des Hautorgans ermoglichen nichtinvasive, diagnostische Verfahren in der Dermatologie zusatzlich, strukturelle und funktionelle Veranderungen der Haut in vivo genauer zu erfassen. Neben der Diagnostik konnen diese Methoden zur Untersuchung dynamischer Prozesse, zur Quantifizierung und zur Verlaufsbeobachtung eingesetzt werden. Zu den Methoden der morphologischen Darstellung gehoren neben der Sonografie, die in Kap. Sonografie von Haut und Lymphknoten abgehandelt wird, die konfokale Lasermikroskopie, die optische Koharenztomografie, die Multiphotonentomografie und die Profilometrie. Parameter der Hautfunktionen – wie Barrierefunktion, Hydratationsstatus, Farbe, pH-Wert, Fettgehalt, Schuppung, Elastizitat, Sauerstoffversorgung und Durchblutung – konnen ebenfalls nichtinvasiv erfasst und quantifiziert werden. Neben dem Einsatz dieser Methoden in Studien zur Quantifizierung von Therapieeffekten haben einige Techniken Eingang in die Routinediagnostik gefunden, zum Beispiel bei der Hautkrebsfruherkennung oder im Rahmen von Begutachtungen bei Berufserkrankungen.

Diagnostic ultrasound has become a standard procedure in clinical dermatology. Devices with intermediate high frequencies of 7.5-15 MHz are used in dermato-oncology for the staging and postoperative care of skin tumor patients and in angiology for improved vessel diagnostics. In contrast, the high frequency ultrasound systems with 20-100 MHz probes offer a much higher resolution, yet with a lower penetration depth of about 1 cm. The main indications are the preoperative measurements of tumor thickness in malignant melanoma and other skin tumors and the assessment of inflammatory and soft tissue diseases, offering information on the course of these dermatoses and allowing therapy monitoring. This article gives an overview on technical principles, devices, mode of examination, influencing factors, interpretation of the images, indications but also limitations of this technique.

Background: Perineural invasion (PNI) in cutaneous squamous cell carcinoma (SCC) is considered to be a negative prognostic factor. A lot of uncertainty remains regarding the classification, diagnosis, treatment and prognosis of SCC with PNI. Objective: To describe typical courses of SCC with PNI and associated findings in order to suggest an optimized diagnostic and therapeutic approach. Methods: We present eight cases of SCC with PNI, considering patient and tumor characteristics, histology, treatment and clinical course regarding local recurrence and metastasization. Results: SCC patients with PNI have a higher rate of local recurrences and greater risk for metastasization than SCC patients without PNI. Age ranged from 68 to 77 years, 6 patients were male and 2 female, with all tumors localized on the head. Three patients had chronic lymphocytic leukemia. Conclusion: Based on the data of this series and the current literature, we make suggestions for better diagnostic and therapeutic management.

Birt-Hogg-Dube syndrome (BHDS) is a rare genetic disorder caused by mutations in the FLCN gene. Clinically, BHDS is highly heterogeneous even within the same family. The most characteristic features of BHDS are fibrofolliculomas, lung cysts that confer the risk of pneumothorax, and benign or malignant renal tumours. Additionally, several other tumours are discussed to be possibly linked to BHDS including colon and breast cancer, thyroid tumors, and parathyroid adenomas. This article is protected by copyright. All rights reserved.