Background: The definitive diagnosis of melanocytic neoplasia using solely histopathologic evaluation can be challenging. Novel techniques that objectively confirm diagnoses are needed. This study details the development and validation of a melanoma prediction model from spatially resolved multivariate protein expression profiles generated by imaging mass spectrometry (IMS). Methods: Three board-certified dermatopathologists blindly evaluated 333 samples. Samples with triply concordant diagnoses were included in this study, divided into a training set (n = 241) and a test set (n = 92). Both the training and test sets included various representative subclasses of unambiguous nevi and melanomas. A prediction model was developed from the training set using a linear support vector machine classification model. Results:We validated the prediction model on the independent test set of 92 specimens (75 classified correctly, 2 misclassified, and 15 indeterminate). IMS detects melanoma with a sensitivity of 97.6% and a specificity of 96.4% when evaluating each unique spot. IMS predicts melanoma at the sample level with a sensitivity of 97.3% and a specificity of 97.5%. Indeterminate results were excluded from sensitivity and specificity calculations. Conclusion:This study provides evidence that IMS-based proteomics results are highly concordant to diagnostic results obtained by careful histopathologic evaluation from a panel of expert dermatopathologists.

The ability of wound dressings to hydrate the stratum corneum has been tested experimentally using attenuated total reflectance (ATR) infrared spectroscopy. A silicone gel dressing was able to hydrate the stratum corneum to the extent that its water content within the sampled volume was >60% after 5 h of contact compared with about 15% of normal exposed skin and about 30% for a highly permeable hydrophilic polyurethane dressing. An ATR method was devised to determine the presence and depth of penetration of silicone in the stratum corneum as a result of skin coverage by the polymeric dressing.

As part of the Accessible Routes from Crowdsourced Cloud Services project (ARCCS) we conducted a series of experiments using the ARCCS sensor to identify push style of wheelchair users. The aim of ARCCS is to make use of a set of well-calibrated sensors to establish a processing chain that then provides ground truth of known accuracy about location, the nature of the environment, and physiological effort. In this paper we focus on two classification problems 1) The push style employed by people as they push themselves and 2) Whether the person is being pushed by an attendant or pushing themselves (independent of push style). Solving the first enables us to develop a level of granularity to pushing classification which transcends rehabilitation and accessibility. The first problem was solved using a wrist-mounted ARCCS sensor, and the second using a wheel-mounted ARCCS sensor. Push styles were classified between semi-circular and arc styles in both indoor and outdoor environments with a high-decrees of precision and recall (>95%). The ARCCS sensor also proved capable of discerning attendant from self-propulsion with near perfect accuracy and recall, without the need for a body-worn sensor.

No abstract