Search

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: 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/).

Introduction In a recent epidemiologic study 5.3% of patients with parkinson disease (PD; mean age 70.5years) and up to 11% of PD patients with self-reported vertigo and dizziness had a benign paroxysmal positional vertigo (BPPV) (VanWensen et al., 2013). The prevalence of PD in industrialized countries is generally estimated at 0.1– 0.3% of the entire population (mean age German population ∼44years). It is more common in the elderly, where prevalences rise from 1.8% in those over 65years of age to 2–3% in the population over 80years (Official guidelines of the German Neurological Society, 2012; De Lau and Breteler, 2006). Thus, two questions arise: first, whether the prevalence of PD is increased in patients with diagnosed BPPV, and secondly, whether the prevalence of parkinson symptoms differs in patients with BPPV compared to those with other peripheral vestibular syndromes such as an unilateral peripheral vestibular loss (vestibular neuritis, VN) or Meniere‘s disease (MD). Methods All patients with diagnosed BPPV ( n =2347), MD ( n =1481) and VN ( n =713) seen in an interdisciplinary dizziness unit as well as in the German Center for Vertigo and Balance disorders of the Munich University between 2004 and 2013 were retrospectively analysed concerning their signs and symptoms of PD. All patients routinely underwent a detailed neurological examination and detailed vestibular testing. The parkinson symptoms were retrospectively classified according to the modified Hoehn and Jahr scale (Goetz, 2004). Results The prevalence for parkinson symptoms in the total BPPV group (mean age 59years) was 1.1%. In the subgroups of BPPV patients >65years of age it raised to 2.6% ( n =25), >70years to 3.5% ( n =23; mean age 77years), and >80years to 4.5% ( n =8). The parkinson symptoms were only mild (Hoehn and Jahr stages ⩽ 2 in the great majority of patients; only 4 patients at stage 3). In VN (mean age 55years) and MD (mean age 59years) the prevalences lay within the range of the normal population: 0.1% in all VN patients, respectively 0.5% in those >65years ( n =1); 0.3% in MD patients, respectively 0.7% in those >65years ( n =4). Conclusions Among patients with peripheral vestibular disorders only the BPPV patients showed a slightly increased prevalence of parkinson symptoms compared to the general population (2.6% vs. 1.8% in patients >65years). However, the prevalence for PD in BPPV patients (3.5% >70years, mean age 77years) seems to be lower than the prevalence for BPPV in PD (5.3% at a mean age of 70.5years) at least in less severely affected patients that can visit an outpatient clinic.