Turbocharging protein binding site prediction with geometric attention, inter-resolution transfer learning, and homology-based augmentation.

Background: Locating small molecule binding sites in target proteins, in the resolution of either pocket or residue, is critical in many drug-discovery scenarios. Since it is not always easy to find such binding sites using conventional methods, different deep learning methods to predict binding sites out of protein structures have been developed in recent years. The existing deep learning based methods have several limitations, including (1) the inefficiency of the CNN-only architecture, (2) loss of information due to excessive post-processing, and (3) the under-utilization of available data sources.

Obesity and periodontitis: A systematic review and updated meta-analysis.

Background: A previous 2014 meta-analysis reported a positive association between obesity and periodontitis. It was considered necessary to update the recently published papers and to analyse subgroups on important clinical variables that could affect the association between obesity and periodontitis. Therefore, we updated the latest studies and attempted to derive more refined results.

Colored Point Cloud Registration by Depth Filtering.

In the last stage of colored point cloud registration, depth measurement errors hinder the achievement of accurate and visually plausible alignments. Recently, an algorithm has been proposed to extend the Iterative Closest Point (ICP) algorithm to refine the measured depth values instead of the pose between point clouds. However, the algorithm suffers from numerical instability, so a postprocessing step is needed to restrict erroneous output depth values.

CTF-based soft touch actuator for playing electronic piano.

In the field of bioinspired soft robotics, to accomplish sophisticated tasks in human fingers, electroactive artificial muscles are under development. However, most existing actuators show a lack of high bending displacement and irregular response characteristics under low input voltages. Here, based on metal free covalent triazine frameworks (CTFs), we report an electro-ionic soft actuator that shows high bending deformation under ultralow input voltages that can be implemented as a soft robotic touch finger on fragile displays.

An Efficient Three-Dimensional Convolutional Neural Network for Inferring Physical Interaction Force from Video.

Interaction forces are traditionally predicted by a contact type haptic sensor. In this paper, we propose a novel and practical method for inferring the interaction forces between two objects based only on video data-one of the non-contact type camera sensors-without the use of common haptic sensors. In detail, we could predict the interaction force by observing the texture changes of the target object by an external force.

Real-Time Gas Mixture Analysis Using Mid-Infrared Membrane Microcavities.

Real-time gas analysis on-a-chip was demonstrated using a mid-infrared (mid-IR) microcavity. Optical apertures for the microcavity were made of ultrathin silicate membranes embedded in a silicon chip using the complementary metal-oxide-semiconductor (CMOS) process. Fourier transform infrared spectroscopy (FTIR) shows that the silicate membrane is transparent in the range of 2.

Inferring Interaction Force from Visual Information without Using Physical Force Sensors.

In this paper, we present an interaction force estimation method that uses visual information rather than that of a force sensor. Specifically, we propose a novel deep learning-based method utilizing only sequential images for estimating the interaction force against a target object, where the shape of the object is changed by an external force. The force applied to the target can be estimated by means of the visual shape changes.

In vitro performance of a shape memory polymer foam-coated coil embolization device.

Intracranial saccular aneurysm treatment using endovascular embolization devices are limited by aneurysm recurrence that can lead to aneurysm rupture. A shape memory polymer (SMP) foam-coated coil (FCC) embolization device was designed to increase packing density and improve tissue healing compared to current commercial devices. FCC devices were fabricated and tested using in vitro models to assess feasibility for clinical treatment of intracranial saccular aneurysms.

Particulate Release From Nanoparticle-Loaded Shape Memory Polymer Foams.

Highly porous, open-celled shape memory polymer (SMP) foams are being developed for a number of vascular occlusion devices. Applications include abdominal aortic and neurovascular aneurysm or peripheral vascular occlusion. A major concern with implanting these high surface area materials in the vasculature is the potential to generate unacceptable particulate burden, in terms of number, size, and composition.

Comparison of shape memory polymer foam versus bare metal coil treatments in an in vivo porcine sidewall aneurysm model.

The endovascular delivery of platinum alloy bare metal coils has been widely adapted to treat intracranial aneurysms. Despite the widespread clinical use of this technique, numerous suboptimal outcomes are possible. These may include chronic inflammation, low volume filling, coil compaction, and recanalization, all of which can lead to aneurysm recurrence, need for retreatment, and/or potential rupture.

In vitro and in vivo evaluation of a shape memory polymer foam-over-wire embolization device delivered in saccular aneurysm models.

Current endovascular therapies for intracranial saccular aneurysms result in high recurrence rates due to poor tissue healing, coil compaction, and aneurysm growth. We propose treatment of saccular aneurysms using shape memory polymer (SMP) foam to improve clinical outcomes. SMP foam-over-wire (FOW) embolization devices were delivered to in vitro and in vivo porcine saccular aneurysm models to evaluate device efficacy, aneurysm occlusion, and acute clotting.

Markov Network-Based Unified Classifier for Face Recognition.

In this paper, we propose a novel unifying framework using a Markov network to learn the relationships among multiple classifiers. In face recognition, we assume that we have several complementary classifiers available, and assign observation nodes to the features of a query image and hidden nodes to those of gallery images. Under the Markov assumption, we connect each hidden node to its corresponding observation node and the hidden nodes of neighboring classifiers.

Design and biocompatibility of endovascular aneurysm filling devices.

The rupture of an intracranial aneurysm, which can result in severe mental disabilities or death, affects approximately 30,000 people in the United States annually. The traditional surgical method of treating these arterial malformations involves a full craniotomy procedure, wherein a clip is placed around the aneurysm neck. In recent decades, research and device development have focused on new endovascular treatment methods to occlude the aneurysm void space.

Estimation of aneurysm wall stresses created by treatment with a shape memory polymer foam device.

In this study, compliant latex thin-walled aneurysm models are fabricated to investigate the effects of expansion of shape memory polymer foam. A simplified cylindrical model is selected for the in-vitro aneurysm, which is a simplification of a real, saccular aneurysm. The studies are performed by crimping shape memory polymer foams, originally 6 and 8 mm in diameter, and monitoring the resulting deformation when deployed into 4-mm-diameter thin-walled latex tubes.

Face recognition system using multiple face model of hybrid Fourier feature under uncontrolled illumination variation.

The authors present a robust face recognition system for large-scale data sets taken under uncontrolled illumination variations. The proposed face recognition system consists of a novel illumination-insensitive preprocessing method, a hybrid Fourier-based facial feature extraction, and a score fusion scheme. First, in the preprocessing stage, a face image is transformed into an illumination-insensitive image, called an "integral normalized gradient image," by normalizing and integrating the smoothed gradients of a facial image.