Stratum Corneum Hydration Measurements with a Bluetooth Wireless Probe: A Real-Life Study at Home Compared to Measurements under Laboratory Conditions.

INTRODUCTION: Non-invasive measurement of the stratum corneum hydration (SCH) with capacitance-based instrumentation is established in dermatological and cosmetic studies. We wanted to test the reliability of non-invasive self-measurements for SCH performed under real-life conditions by volunteers with a Bluetooth-based (wireless) probe Corneometer® (CM 825i) transmitting the data via a smartphone application to a central server. Probes and smartphones communicated using Bluetooth Low Energy. Data from the smartphone were securely transferred to a remote server in a different country with TLS encryption using HTTPS protocols. CM 825i values were correlated with the established CM 825 under laboratory conditions. The primary endpoint was the correlation of the two probes. Secondary endpoints were the coefficient of variation (CV) and delta values (before and after treatment). METHODS: Eighteen healthy volunteers (f: 8; m: 10) participated in the prospective observational study. The real-world home use of the wireless CM 825i was performed before and after treatments with base cream DAC for 7 days. RESULTS: Both instruments showed a significant and relevant correlation (p < 0.0001; Spearman coefficient of r = 0.8647). CM 825i and CM 825 differentiate significantly between normal and high SCH. Both devices showed comparable robustness in repeated measurements with a CV between 5.6% and 9.2%. CONCLUSION: We could show a significant correlation between both devices and a comparable differentiation between low and high SCH and comparable CVs. The real-life use demonstrated adequate acquiring and transmitting of in vivo data to a smartphone and subsequently transmitting to the secure server with low numbers of missed transmissions (<0.2%) and missed measurements (<5%).

Estimating and abstracting the 3D structure of feline bones using neural networks on X-ray (2D) images.

Computing 3D bone models using traditional Computed Tomography (CT) requires a high-radiation dose, cost and time. We present a fully automated, domain-agnostic method for estimating the 3D structure of a bone from a pair of 2D X-ray images. Our triplet loss-trained neural network extracts a 128-dimensional embedding of the 2D X-ray images. A classifier then finds the most closely matching 3D bone shape from a predefined set of shapes. Our predictions have an average root mean square (RMS) distance of 1.08 mm between the predicted and true shapes, making our approach more accurate than the average achieved by eight other examined 3D bone reconstruction approaches. Each embedding extracted from a 2D bone image is optimized to uniquely identify the 3D bone CT from which the 2D image originated and can serve as a kind of fingerprint of each bone; possible applications include faster, image content-based bone database searches for forensic purposes.