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.

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.

Severe hypokalaemia refractory to potassium therapy may occur during therapeutic thiopentone coma. Severe rebound hyperkalaemia may occur after cessation of thiopentone infusion. Protocols for the management of patients with therapeutic barbiturate coma should recognise this potentially serious complication.

Background: Terminal delirium, specifically the hyperactive delirium subtype at the end of life, is common in palliative care patients. Standard care often involves sedation to alleviate distress. The alpha2-adrenoreceptor agonist dexmedetomidine may have promise in terminal delirium, due to its properties of decreasing delirium and permitting rousable sedation. Aim: This study aimed to describe the effect of dexmedetomidine on delirium and sedation, when delivered via continuous subcutaneous infusion (CSCI) in patients with terminal delirium. Design: The trial was prospectively registered in the ANZCTR database (ACTRN12618000658213) and conducted in accordance with CONSORT (pilot study extension). Twenty-two adult patients were treated with a CSCI of dexmedetomidine with a two-tier dose schedule, low and high dose. Delirium severity was measured by the Memorial Delirium Assessment Scale (MDAS, target <13), and sedation by the Richmond Agitation-Sedation Scale, Palliative Version (RASS-PAL, target −1 to −3). Results: All patients had a response to dexmedetomidine as measured by decrease in MDAS after initiation; 59% required escalation to high dose to maintain control of delirium. All responses to high dose were sustained. RASS-PAL scores showed significant variability, however mean scores remained within target range on both doses, and the majority of patients were rousable. Fifty percent of patients treated crossed over to standard care; no patients who crossed over were experiencing moderate-severe delirium. Predominant reason for crossover was family request for deeper sedation. Conclusion: Dexmedetomidine shows potential for the management of terminal delirium with improved interactivity. Further research is needed to determine efficacy compared to current standard care.

No abstract