University of Lincoln

About: University of Lincoln is a education organization based out in Lincoln, Lincolnshire, United Kingdom. It is known for research contribution in the topics: Population & Context (language use). The organization has 2341 authors who have published 7025 publications receiving 124797 citations.

Siamese neural networks for continuous disease severity evaluation and change detection in medical imaging.

Abstract: Using medical images to evaluate disease severity and change over time is a routine and important task in clinical decision making. Grading systems are often used, but are unreliable as domain experts disagree on disease severity category thresholds. These discrete categories also do not reflect the underlying continuous spectrum of disease severity. To address these issues, we developed a convolutional Siamese neural network approach to evaluate disease severity at single time points and change between longitudinal patient visits on a continuous spectrum. We demonstrate this in two medical imaging domains: retinopathy of prematurity (ROP) in retinal photographs and osteoarthritis in knee radiographs. Our patient cohorts consist of 4861 images from 870 patients in the Imaging and Informatics in Retinopathy of Prematurity (i-ROP) cohort study and 10,012 images from 3021 patients in the Multicenter Osteoarthritis Study (MOST), both of which feature longitudinal imaging data. Multiple expert clinician raters ranked 100 retinal images and 100 knee radiographs from excluded test sets for severity of ROP and osteoarthritis, respectively. The Siamese neural network output for each image in comparison to a pool of normal reference images correlates with disease severity rank (ρ = 0.87 for ROP and ρ = 0.89 for osteoarthritis), both within and between the clinical grading categories. Thus, this output can represent the continuous spectrum of disease severity at any single time point. The difference in these outputs can be used to show change over time. Alternatively, paired images from the same patient at two time points can be directly compared using the Siamese neural network, resulting in an additional continuous measure of change between images. Importantly, our approach does not require manual localization of the pathology of interest and requires only a binary label for training (same versus different). The location of disease and site of change detected by the algorithm can be visualized using an occlusion sensitivity map-based approach. For a longitudinal binary change detection task, our Siamese neural networks achieve test set receiving operator characteristic area under the curves (AUCs) of up to 0.90 in evaluating ROP or knee osteoarthritis change, depending on the change detection strategy. The overall performance on this binary task is similar compared to a conventional convolutional deep-neural network trained for multi-class classification. Our results demonstrate that convolutional Siamese neural networks can be a powerful tool for evaluating the continuous spectrum of disease severity and change in medical imaging.

Constructionism, Realism and Housing Theory

Abstract: This paper considers the theoretical postulates of the discourses of 'objectivist realism' and social constructionism and discusses their relevance to housing studies. In place of such theories, the paper puts forward a theory of contextualised rational action. This theory is then applied in the contexts of 'tenant participation', 'housing management' and 'comparative housing policy'.

Extinct, obscure or imaginary: The lizard species with the smallest ranges

Abstract: Aim Small geographic ranges make species especially prone to extinction from anthropogenic disturbances or natural stochastic events. We assemble and analyse a comprehensive dataset of all the world's lizard species and identify the species with the smallest ranges—those known only from their type localities. We compare them to wide-ranging species to infer whether specific geographic regions or biological traits predispose species to have small ranges. Location Global. Methods We extensively surveyed museum collections, the primary literature and our own field records to identify all the species of lizards with a maximum linear geographic extent of <10 km. We compared their biogeography, key biological traits and threat status to those of all other lizards. Results One in seven lizards (927 of the 6,568 currently recognized species) are known only from their type localities. These include 213 species known only from a single specimen. Compared to more wide-ranging taxa, they mostly inhabit relatively inaccessible regions at lower, mostly tropical, latitudes. Surprisingly, we found that burrowing lifestyle is a relatively unimportant driver of small range size. Geckos are especially prone to having tiny ranges, and skinks dominate lists of such species not seen for over 50 years, as well as of species known only from their holotype. Two-thirds of these species have no IUCN assessments, and at least 20 are extinct. Main conclusions Fourteen per cent of lizard diversity is restricted to a single location, often in inaccessible regions. These species are elusive, usually poorly known and little studied. Many face severe extinction risk, but current knowledge is inadequate to properly assess this for all of them. We recommend that such species become the focus of taxonomic, ecological and survey efforts.

SimpleDS: A Simple Deep Reinforcement Learning Dialogue System

Abstract: This article presents SimpleDS, a simple and publicly available dialogue system trained with deep reinforcement learning. In contrast to previous reinforcement learning dialogue systems, this system avoids manual feature engineering by performing action selection directly from raw text of the last system and (noisy) user responses. Our initial results, in the restaurant domain, report that it is indeed possible to induce reasonable behaviours with such an approach that aims for higher levels of automation in dialogue control for intelligent interactive systems and robots.