Rashes, ulcers and skin lesions are well suited for telemedicine. We have developed a smartphone app, the first of its kind in Norway, where a referring physician can write a short medical history and take clinical and dermatoscopic photographs with a smartphone, which is then sent to and evaluated by a dermatologist. In the period from June 1 st , 2017, to September 1 st , 2019, clinical information and photographs of rash and skin lesions from 171 patients were sent by 40 primary care and nursing home physicians via the smartphone app to four dermatologists for diagnosis and therapeutic advice. A wide range of dermatological conditions were diagnosed, most commonly chronic ulcers (17%), eczema (15%) and pigmented lesions (13%). Assessed later by a dermatologist, referral for regular consultations with a specialist was avoided in 119 patients (70%). Sixteen patients (9%) were recommended a regular consultation with a dermatologist; information for prioritization in the specialist healthcare service was then provided. In 36 patients (21%), further measures by the referring physician were recommended. Our experience indicates that many ordinary consultations on rash, ulcers and skin lesions in the specialist healthcare services can be avoided when using the smartphone app.

Assume that we are given a coaction δ of a locally compact group G on a C * -algebra A and a T-valued Borel 2-cocycle ω on G. Motivated by the approach of Kasprzak to Rieffel's deformation we define a deformation Aω of A. Among other properties of Aω we show that Aω ⊗ K(L 2 (G)) is canonically isomorphic to A ⋊ δĜ ⋊δ ,ω G. This, together with a slight extension of a result of Echterhoff et al., implies that for groups satisfying the Baum-Connes conjecture the K-theory of Aω remains invariant under homotopies of ω.

We train a machine learning model on large data set for predicting property values in the Norwegian real estate market. Our model is a gradient boosted regression tree. The data set is the largest market data set of properties in Norway considered in the research literature. We achieve state of the art accuracy. A large scale market data set of real estate properties is collected from sales and rental ads on publicly accessible internet sites. The property advertisements show property features and appraisal values made by real estate brokers. We train a gradient boosted regression tree model on selected features of the data set. This is a multivariate regression model built with supervised learning. We do 5-fold cross validation to assess the accuracy and robustness of the model. The gradient boosted regression tree models are already known to give the best prediction accuracy on real estate price valuations. We achieve state of the art pre- diction accuracy using a minimal feature set and only publicly and freely available sales advertisement data. The novelty of our work lies in the fact that we use a minimal feature set in our model, and we have the largest data set in the research literature, and moreover we have used only freely and publicly accessible data which are simple to obtain. This shows that useful estimation models with high accuracy can be built with quite simple resources.

No abstract

ImportanceState-of-the art performance is achieved with a deep learning object detection model for acne detection. There is little current research on object detection in dermatology and acne in particular. As such, this work is early in this field and achieves state of the art performance.ObjectiveTrain an object detection model on a publicly available data set of acne photos.Design, Setting, and ParticipantsA deep learning model is trained with cross validation on a data set of facial acne photos.Main Outcomes and MeasuresObject detection models for detecting acne for single-class (acne) and multi-class (four severity levels). We train and evaluate the models using standard metrics such as mean average precision (mAP). Then we manually evaluate the model predictions on the test set, and calculate accuracy in terms of precision, recall, F1, true and false positive and negative detections.ResultsWe achieve state-of-the art mean average precision mAP@0.5 value of 37.97 for the single class acne detection task, and 26.50 for the 4-class acne detection task. Moreover, our manual evaluation shows that the single class detection model performs well on the validation set, achieving true positive 93.59 %, precision 96.45 % and recall 94.73 %.Conclusions and RelevanceWe are able to train a high-accuracy acne detection model using only a small publicly available data set of facial acne. Transfer learning on the pre-trained deep learning model yields good accuracy and high degree of transferability to patient submitted photographs. We also note that the training of standard architecture object detection models has given significantly better accuracy than more intricate and bespoke neural network architectures in the existing research literature.Key PointsQuestionCan deep learning-based acne detection models trained on a small data set of publicly available photos of patients with acne achieve high prediction accuracy?FindingsWe find that it is possible to train a reasonably good object detection model on a small, annotated data set of acne photos using standard deep learning architectures.MeaningDeep learning-based object detection models for acne detection can be a useful decision support tools for dermatologists treating acne patients in a digital clinical practice. It can prove a particularly useful tool for monitoring the time evolution of the acne disease state over prolonged time during follow-ups, as the model predictions give a quantifiable and comparable output for photographs over time. This is particularly helpful in teledermatological consultations, as a prediction model can be integrated in the patient-doctor remote communication.