Abstract
Background and purpose We evaluated two methods to identify mesial temporal sclerosis (MTS): visual inspection by experienced epilepsy neuroradiologists based on structural magnetic resonance imaging sequences and automated hippocampal volumetry provided by a processing pipeline based on the FMRIB Software Library. Methods This retrospective study included patients from the epilepsy monitoring unit database of our institution. All patients underwent brain magnetic resonance imaging in 1.5T and 3T scanners with protocols that included thin coronal T2, T1 and fluid-attenuated inversion recovery and isometric T1 acquisitions. Two neuroradiologists with experience in epilepsy and blinded to clinical data evaluated magnetic resonance images for the diagnosis of MTS. The diagnosis of MTS based on an automated method included the calculation of a volumetric asymmetry index between the two hippocampi of each patient and a threshold value to define the presence of MTS obtained through statistical tests (receiver operating characteristics curve). Hippocampi were segmented for volumetric quantification using the FIRST tool and fslstats from the FMRIB Software Library. Results The final cohort included 19 patients with unilateral MTS (14 left side): 14 women and a mean age of 43.4 ± 10.4 years. Neuroradiologists had a sensitivity of 100% and specificity of 73.3% to detect MTS (gold standard, k = 0.755). Automated hippocampal volumetry had a sensitivity of 84.2% and specificity of 86.7% (k = 0.704). Combined, these methods had a sensitivity of 84.2% and a specificity of 100% (k = 0.825). Conclusions Automated volumetry of the hippocampus could play an important role in temporal lobe epilepsy evaluation, namely on confirmation of unilateral MTS diagnosis in patients with radiological suggestive findings. © SAGE Publications.

Abstract
Gene expression is dependent on genetic and environmental factors. In the last decade, a large body of research has significantly improved our understanding of the genetic architecture of gene expression. However, it remains unclear whether genetic effects on gene expression remain stable over time. Here, we show, using longitudinal whole-blood gene expression data from a twin cohort, that the genetic architecture of a subset of genes is unstable over time. In addition, we identified 2213 genes differentially expressed across time points that we linked with aging within and across studies. Interestingly, we discovered that most differentially expressed genes were affected by a subset of 77 putative causal genes. Finally, we observed that putative causal genes and down-regulated genes were affected by a loss of genetic control between time points. Taken together, our data suggest that instability in the genetic architecture of a subset of genes could lead to widespread effects on the transcriptome with an aging signature. ©2017 Bryois et al.

Abstract
Given a retinal image, can we automatically determine whether it is of high quality (suitable for medical diagnosis)? Can we also explain our decision, pinpointing the region or regions that led to our decision? Images from human retinas are vital for the diagnosis of multiple health issues, like hypertension, diabetes, and Alzheimer's; low quality images may force the patient to come back again for a second scanning, wasting time and possibly delaying treatment. However, existing retinal image quality assessment methods are either black boxes without explanations of the results or depend heavily on feature engineering or on complex and error-prone anatomical structures' segmentation. Therefore, we propose EyeQual, that solves exactly this problem. EyeQual is novel, fast for inference, accurate and explainable, pinpointing low-quality regions on the image. We evaluated EyeQual on two real datasets where it achieved 100% accuracy taking just 36 milliseconds for each image.

Abstract
This paper describes a methodology for Diabetic Retinopathy detection from eye fundus images using a generalization of the Bag-of-Visual-Words (BoVW) method. We formulate the BoVW as two neural networks that can be trained jointly. Unlike the BoVW, our model is able to learn how to perform feature extraction, feature encoding and classification guided by the classification error. The model achieves 0.97 Area Under the Curve (AUC) on the DR2 dataset while the standard BoVW approach achieves 0.94 AUC. Also, it performs at the same level of the state-of-the-art on the Messidor dataset with 0.90 AUC.

Abstract
Colorectal carcinoma is a major health concern worldwide and its high incidence and mortality require accurate screening methods. Following endoscopic examination, polyps must be removed for histopathological characterization. Aiming to contribute to the improvement of current endoscopy methods of colorectal carcinoma screening or even for future development of laser treatment procedures, we studied the diffusion properties of glucose and water in colorectal healthy and pathological mucosa. These parameters characterize the tissue dehydration and the refractive index matching mechanisms of optical clearing (OC). We used ex vivo tissues to measure the collimated transmittance spectra and thickness during treatments with OC solutions containing glucose in different concentrations. These time dependencies allowed for estimating the diffusion time and diffusion coefficient values of glucose and water in both types of tissues. The measured diffusion times for glucose in healthy and pathological mucosa samples were 299.2 +/- 4.7 s and 320.6 +/- 10.6 s for 40% and 35% glucose concentrations, respectively. Such a difference indicates a slower glucose diffusion in cancer tissues, which originate from their ability to trap far more glucose than healthy tissues. We have also found a higher free water content in cancerous tissue that is estimated as 64.4% instead of 59.4% for healthy mucosa. (C) 2017 Society of Photo-Optical Instrumentation Engineers (SPIE)

Abstract
A complete analysis of the vascular system is a complex task since a large number of parameters are involved. In the research herein reported we present a novel medical framework called web-based integration for vascular expert research networks (Wivern) to be used in a multi-clinical department environment for the analysis of micro and macrocirculation. This tool can manage clinical information of several specialties, such as Neurology or Ophthalmology, and provides computer-aided tools to automatically analyze retinographies, carotid ultrasounds and blood pressure monitor signals, and to automatically compute cardiovascular risk stratification. Wivern is a web-based application with a user friendly interface that provides cross-platform compatibility and device independence. Several automated procedures are integrated within the framework, as a service on the web, to extract relevant parameters from clinical data, physiological signals and medical images. The application is planned for collecting and analyzing data in several clinical studies in different hospital centers to test their behavior and practical use of the different tools of the platform. The usefulness and validation of the system was achieved after the inclusion, by the different medical units, of 800 patients to analyze their hypertensive profile. Moreover, 800 retinal images were processed as well as 400 carotid were analyzed. Wivern provides a unique opportunity for vascular research since it enables an interdisciplinary and integrated study of the vascular network, allowing a more comprehensive evaluation of the consequences of any abnormality. The application also includes automated methods to process patient data in order to simplify the physician tasks.