Recognizing Emotions From an Ensemble of Features.

This paper details the authors' efforts to push the baseline of emotion recognition performance on the Geneva Multimodal Emotion Portrayals (GEMEP) Facial Expression Recognition and Analysis database. Both subject-dependent and subject-independent emotion recognition scenarios are addressed in this paper. The approach toward solving this problem involves face detection, followed by key-point identification, then feature generation, and then, finally, classification. An ensemble of features consisting of hierarchical Gaussianization, scale-invariant feature transform, and some coarse motion features have been used. In the classification stage, we used support vector machines. The classification task has been divided into person-specific and person-independent emotion recognitions using face recognition with either manual labels or automatic algorithms. We achieve 100% performance for the person-specific one, 66% performance for the person-independent one, and 80% performance for overall results, in terms of classification rate, for emotion recognition with manual identification of subjects.

[In vivo osteogenic evaluation of titanium implants with strontium loaded nanotubes].

Objective: To evaluate the in vivo osteogenic activity of titanium implants with strontium loaded TiO2 nanotubes (NTSr). Methods: The strontium loaded titanium nanotubes were formed on pure titanium implants through anodization and hydrothermal treatment, and the unmodified titanium (Control) and sheer TiO2 nanotubes (NT) were set to be control groups and treatment group. Inductively coupled plasma mass spectrometry (ICP-MS) was used to evaluate the Sr release at 28 days. Field emission scanning electron microscopes (FE-SEM) was used to view the micro-topography, atomic force microscope was used to exam the surface roughness, and nano-indenter was used to evaluate the hardness of three groups (n=3). Three groups of implant samples were inserted into the distal femoral metaphysis of New Zealand rabbits (n=4 at each time point). After 4 weeks and 12 weeks, samples were harvested. Micro-CT scanning, immunofluorescent and histological examinations were carried out. Results: The strontium ions could be released slowly for at least 28 days [the Sr concentration at 28 Day was (2.6±1.5) ng/ml]. NTSr coating exhibited a nanoscale tube array (the diameter was about 70 nm), and the surface roughness of implant was increased with the nanobube coating [Control (34.8±5.3) nm, NT (66.2±4.3) nm, NTSr (85.7±10.6) nm, F=37.59, P<0.001]. The surface roughness (Ra) of NT and NTSr groups was higher than the control group (P<0.05). Comparing to Control implants, NTSr implants exhibited a better osteogenic ability [the bone volume/total volume (BV/TV) value was Control (24.7±1.1)% vs. NTSr (37.7±1.9)% at 4 weeks (P<0.05), and Control (40.7±0.9)% vs. NTSr (51.9±2.1)% at 12 weeks (P<0.05)]. The fluorescent examination revealed that NTSr coating can also accelerated the generation of new bone tissue (bone tissue area% labelled by alizarin red at day 7 was Control (19.2±2.9)% vs. NT (35.4±3.7)% vs. NTSr (40.9±0.9)% (F=42.74, P<0.01). The results in the NT and NTSr group were statistically higher than that in the control group (P<0.05). Conclusions: The strontium loaded TiO2 nanotubes can enhance new bone formation around titanium implants.

[Activation of endoplasmic reticulum stress and its effect on osteogenic differentiation induced by micropit/nanotube topography].

Objective: To explore the activation of endoplasmic reticulum stress (ERS) in bone marrow mesenchymal stem cell (BMMSC) and its effect on osteogenic differentiation induced by micropit/nanotube topography (MNT), so as to provide guidance for the topography design of biomaterials. Methods: Four sample groups were fabricated: polishing control group (polished titanium, PT, no treatment), thapsigargin treatment (TG, 0.1 µmol/L TG treated for 9 h), MNT5 and MNT20 (anodized at 5 V and 20 V after acid etching). Scanning electron microscope (SEM) was used to observe the topography of Ti samples. The alkaline phosphatase (ALP) production, collagen secretion and extracellular matrix (ECM) mineralization of BMMSC (osteogenic induced for 7, 14 and 21 d) on Ti samples were detected to evaluate the osteogenic differentiation. After 12 h incubation, the shape and size of ER was examined using a transmission electron microscope (TEM), and ERS-related genes including immunoglobulin heavy chain binding protein (BiP), protein kinase RNA-like endoplasmic reticulum kinase (PERK) and activating transcription factor 4 (ATF4) were detected by quantitative real-time PCR (qRT-PCR). Results: After 7, 14 and 21 d of induction, the ALP production, collagen secretion and ECM mineralization in TG and MNT20 all significantly increased compared to PT (P<0.05). The cells grown on TG, MNT5 and MNT20 surfaces displayed gross distortions of the ER. Compared to PT, BiP, PERK, ATF4 mRNA expression in TG was respectively 1.87±0.10, 2.24±0.35, 1.85±0.14; BiP, ATF4 mRNA expression in MNT5 were respectively 1.27±0.09, 1.25±0.04; BiP, PERK, ATF4 mRNA expression in MNT20 were respectively 1.44±0.09, 2.40±0.60, 1.48±0.05 (P<0.05). Conclusions: MNT triggered different degree of ERS, and the activated ERS may promote MNT-induced osteogenic differentiation.