Short term studies in controlled environments have shown that user behaviour is consistent enough to predict disruptive smartphone notifications. However, in practice, user behaviour changes over time (concept drift) and individual user preferences need to be considered. There is a lack of research on which methods are best suited for predicting disruptive smartphone notifications longer-term, taking into account varying error costs. In this paper we report on a 16 week field study comparing how well different learners perform at mitigating disruptive incoming phone calls.
In machine learning, concept drift can cause the optimal solution to a given problem to change as time passes, leading to less accurate predictions. Concept drift can be sudden, gradual or reoccuring. Understanding the consequences of concept drift is particularly important in human-centric applications where changes in the underlying data and environment are common and unexpected. In order to gain a better understanding of the adverse effects of different types of concept drift on learners, we propose a novel simulation tool that is able to incrementally generate datasets with customisable concept drift by interacting with a human in a game-like setting. We illustrate our approach by generating and analysing concept drift simulations inspired by body-sensor based long-term activity recognition. Our initial results show that current unsupervised adaptation techniques can be caught in cyclic mislabelling and that a hybrid solution that is selfcalibrating and semi-supervised is more robust than any of the two taken separately for this example.
Sterility restrictions in surgical settings make touch-less interaction an interesting solution for surgeons to interact directly with digital images. In this demo, we present a system for gesture-based interaction with medical images based on a wristband inertial sensor and capacitive floor sensors, allowing for hand and foot gesture input. Hand gesture commands have been designed for interacting with 3D and 2D medical images in two different displays, while foot gestures can enable, disable, and switch interaction between different systems. The gestures are recognized in real time with the help of a neural network classifier, which is based on a given training set and extracts different features of accelerometer and gyroscope. For displaying the medical images a simple image viewer for 2D images is used while 3D images are presented with the open-source software InVesalius.
Introduction
There is ongoing debate over whether to fix asymptomatic contralateral inguinal hernias during repair of the presenting hernia. This study reviewed the practice of one high-volume hernia surgeon, comparing unilateral and bilateral laparoscopic repairs, to establish if bilateral repair is associated with an increased risk of post-operative complications compared to unilateral repair.
Material & Methods
All patients operated on within a 5-year period were audited and subcategorised depending on whether they underwent unilateral or bilateral repair. Data was collected on complication rates, rates of recurrence, readmission within 30 days and duration of operation.
Results
186 patients underwent repair of 265 hernias. 79 patients underwent bilateral repair. 25 patients suffered a complication (13.4%). Complication rates were 11.2% and 16.5% in the unilateral group and bilateral group respectively (p=0.50). These included superficial infection (1.9% vs 1.3%, p=0.75), haematoma (5.6% vs 7.6%, p=0.59), seroma (2.8 vs 7.6%, p=0.13), and chronic pain (0% vs 2.5%, p=0.01). There were no mesh infections, persisting numbness or explantation in either arm. In both groups, median nights stayed was 0. At median length of follow-up of 1.8 years, recurrence rates were 1.9% vs 0% in the unilateral and bilateral arms respectively (p=0.22).
Conclusions
There was no significant difference in rates of complications between the bilateral and unilateral arms. Opportunistic repair may reduce risk of future surgeries and morbidity and does not appear to be associated with increased complications. Surgeons may wish to consider this when consenting patients.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.