Real-time, simultaneous myoelectric control using a convolutional neural network

The evolution of deep learning techniques has been transformative as they have allowed complex mappings to be trained between control inputs and outputs without the need for feature engineering. In this work, a myoelectric control system based on convolutional neural networks (CNN) is proposed as a possible alternative to traditional approaches that rely on specifically designed features. This CNN-based system is validated using a real-time Fitts' law style target acquisition test requiring single and combined wrist motions. The performance of the proposed system is then compared to that of a standard support vector machine (SVM) based myoelectric system using a set of time-domain features. Despite the prevalence and demonstrated performance of these well-known features, no significant difference (p>0.05) was found between the two methods for any of the computed control metrics. This demonstrates the potential for automated learning approaches to extract complex and rich information from stochastic biological signals. This first evaluation of the usability of a CNN in a real-time myoelectric control environment provides a basis for further exploration.

Detection and Discrimination of Bacterial Colonies with Mueller Matrix Imaging

The polarization imaging technique is a powerful approach to probe microstructural and optical information of biological structures (e.g., tissue samples). Here, we have studied the polarization properties of different bacterial colonies in order to evaluate the possibility of bacterial detection and discrimination. In this regard, we have taken the backscattering Mueller matrix images of four different bacteria colonies (i.e., Escherichia coli, Lactobacillus rhamnosus, Rhodococcus erythropolis, and Staphylococcus aureus). Although the images have the potential to distinguish qualitatively different bacterial colonies, we explored more accurate and quantitative parameters criteria for discrimination of bacterial samples; more specifically, we have exploited the Mueller matrix polar decomposition (MMPD),frequency distribution histogram (FDH), and central moment analysis method. The outcomes demonstrated a superior capacity of Mueller matrix imaging, MMPD, and FDH in bacterial colonies identification and discrimination. This approach might pave the way for a reliable, efficient, and cheap way of identification of infectious diseases.

A high-throughput flow cytometry-on-a-CMOS platform for single-cell dielectric spectroscopy at microwave frequencies

This work presents a microfluidics-integrated label-free flow cytometry-on-a-CMOS platform for the characterization of the cytoplasm dielectric properties at microwave frequencies. Compared with MHz impedance cytometers, operating at GHz frequencies offers direct intracellular permittivity probing due to electric fields penetrating through the cellular membrane. To overcome the detection challenges at high frequencies, the spectrometer employs on-chip oscillator-based sensors, which embeds simultaneous frequency generation, electrode excitation, and signal detection capabilities. By employing an injection-locking phase-detection technique, the spectrometer offers state-of-the-art sensitivity, achieving a less than 1 aFrms capacitance detection limit (or 5 ppm in frequency-shift) at a 100 kHz noise filtering bandwidth, enabling high throughput (>1k cells per s), with a measured cellular SNR of more than 28 dB. With CMOS/microfluidics co-design, we distribute four sensing channels at 6.5, 11, 17.5, and 30 GHz in an arrayed format whereas the frequencies are selected to center around the water relaxation frequency at 18 GHz. An issue in the integration of CMOS and microfluidics due to size mismatch is also addressed through introducing a cost-efficient epoxy-molding technique. With 3-D hydrodynamic focusing microfluidics, we perform characterization on four different cell lines including two breast cell lines (MCF-10A and MDA-MB-231) and two leukocyte cell lines (K-562 and THP-1). After normalizing the higher frequency signals to the 6.5 GHz ones, the size-independent dielectric opacity shows a differentiable distribution at 17.5 GHz between normal (0.905 ± 0.160, mean ± std.) and highly metastatic (1.033 ± 0.107) breast cells with p ≪ 0.001.

High strain rate and high temperature response of two armour steels: Experimental testing and constitutive modelling

Under ballistic impact or blast loading, the high strain rate and high temperature behaviour of armour steels is key to their response to a given threat. This experimental and numerical investigation examines the tensile response of a class 4a improved rolled homogenous armour steel (IRHA) and a high hardness armour steel (HHA). Cylindrical tensile specimens were tested at a range of strain rates from 0.001 s-1 to 2700 s-1. Quasi-static, elevated temperature tests were performed from room temperature up to 300° C. While the HHA is strain rate insensitive, the IRHA displays a significant increase in strength across the range of loading rates reducing the ultimate strength difference between the materials from 19% at 0.001s-1 to 4.6% at 2700s-1. An inverse numerical modelling approach for constitutive model calibration is presented, which accurately captured the dynamic material behaviour. The modified Johnson-Cook strength and Cockcroft-Latham (C-L) fracture models were capable of predicting the ballistic limit of each material to within 5% of the experimental result and to within 10% for deformation under blast loading. The blast rupture threshold of both materials was significantly over-estimated by the C-L model suggesting stress state or strain rate effects may be reducing the ductility of armour steel under localised blast loading.

Phenytoin mouthwash to treat cancer therapy-induced oral mucositis: A pilot studyPrimary neuroendocrine carcinoma of breast: A rare tumor

BACKGROUND

Oral mucositis is one of the most common side effects of cancer therapy with no definite treatment. Phenytoin has positive effects on healing of mucosal and dermal wounds. In this study efficacy of 1% phenytoin mouthwash on severity of mucositis (on the basis of WHO scale), pain relief (based on Visual Analogue Scale), and improvement of patients' quality of life (on the basis of EORTC-QLQ-H and N35 questionnaire) was evaluated.

MATERIALS AND METHODS

In a pilot -double-blind randomized clinical trial, eight patients in study group were given 1% phenytoin mouthwash while eight patients in control group used normal saline. Data analysis was performed by Mann-Whitney and Repeated Measured ANOVA tests.

RESULTS

Reduction of mucositis severity was observed, but the difference was not significant. On the other hand, patients on phenytoin therapy had better pain relief (VAS# 6.75 ± 1.58 at the beginning of the study reached to # 3.75 ± 1.16 after 3 weeks in phenytoin group) and improvement in quality of life (score of QOL was 70.63 ± 5.5 that reached to 63.61 ± 6.39 in phenytoin group) than normal saline group significantly (P < 0.05).

CONCLUSION

One percent phenytoin mouthwash caused pain relief and improvement of life quality significantly in patients with mucositis due to cancer therapy, but it did not reduce the severity of mucositis in a statistically significant scale.

INVESTIGATION OF SPIRAL AND SWEEPING HOLES

Surface infrared thermography, hotwire anemometry, and thermocouple surveys were performed on two new film cooling hole geometries: spiral/rifled holes and fluidic sweeping holes. The spiral holes attempt to induce large-scale vorticity to the film cooling jet as it exits the hole to prevent the formation of the kidney shaped vortices commonly associated with film cooling jets. The fluidic sweeping hole uses a passive in-hole geometry to induce jet sweeping at frequencies that scale with blowing ratios. The spiral hole performance is compared to that of round holes with and without compound angles. The fluidic hole is of the diffusion class of holes and is therefore compared to a 777 hole and Square holes. A patent-pending spiral hole design showed the highest potential of the non-diffusion type hole configurations. Velocity contours and flow temperature were acquired at discreet cross-sections of the downstream flow field. The passive fluidic sweeping hole shows the most uniform cooling distribution but suffers from low span-averaged effectiveness levels due to enhanced mixing. The data was taken at a Reynolds number of 11,000 based on hole diameter and freestream velocity. Infrared thermography was taken for blowing ratios of 1.0, 1.5, 2.0, and 2.5 at a density ratio of 1.05. The flow inside the fluidic sweeping hole was studied using 3D unsteady RANS.

A survey of Euphorinae (Hymenoptera: Braconidae) of southern Iran, with description of a new species

A faunistic survey of Euphorinae (Hym., Braconidae) in southern Iran, as well as an updated checklist of the genera and species in Iran, are presented. Sampling was performed using Malaise traps at different locations of Hormozgan province and Qeshm Island in Persian Gulf during 2011-2013. In total, 38 species belonging to 10 genera are listed from Iran. The recorded species belong to the following genera: Allurus Forster, 1862 (two species), Chrysopophthorus Goidanich 1948 (one species), Dinocampus Forster, 1862 (one species), Ecclitura Kokujev, 1902 (one species), Leiophron Nees von Esenbeck, 1819 (10 species), Meteorus Haliday, 1835 (12 species), Perilitus Nees von Esenbeck, 1819 (five species), Syntretus Forster 1862 (three species), Wesmaelia Foerster, 1862 (one species) and Zele Curtis, 1832 (two species). Allurus lituratus (Haliday 1835), Dinocampus coccinellae (Schrank, 1802), Leiophron (Peristenus) grandiceps (Thomson 1892), Meteorus rubens (Nees, 1811) and Wesmaelia petiolata (Wollaston, 1858) are new records for Hormozgan province and Leiophron (Peristenus) grandiceps (Thomson 1892) is recorded for the first time from Iran. In addition, Meteorus breviterebratus Ameri, Talebi & Beyarslan sp. n. is newly described and illustrated. 

Study of the genus Opius Wesmael (Hymenoptera: Braconidae: Opiinae) in Southern Iran, with eleven new records

This study was carried out to determine the species of the genus Opius Wesmael, 1835 in Hormozgan province (Southern Iran). Malaise traps and sweep nets were used to obtain adult specimens from various habitats in Hormozgan province during 2011-2013. Fifteen species from the genus Opius belonging to ten subgenera were collected. The subgenera Merotrachys Fischer, 1972 and Opiostomus Fischer, 1972 and eleven species are recorded for the first time from Iran: Opius (Agnopius) nowakowskii Fischer, 1959; Opius (Agnopius) novosimilis Fischer, 1989, Opius (Allophlebus) staryi Fischer, 1958; Opius (Allotypus) damnosus Papp, 1980; Opius (Opiostomus) riphaeus Tobias, 1986; Opius (Opiothorax) minusculae Fischer, 1967; Opius (Pendopius) bajariae Fischer, 1989; Opius (Merotrachys) penetrator Fischer 1966; Opius (Hypocynodus) flavipes Szepligeti, 1898; Opius (Hypocynodus) latidens Fischer, 1990 and Opius (Hypocynodus) latipediformis Fischer 2004. A key for identification of Opius species from southern Iran is provided.

Support vector regression for improved real-time, simultaneous myoelectric control

This study describes the first application of a support vector machine (SVM) based scheme for real-time simultaneous and proportional myoelectric control of multiple degrees of freedom (DOFs). Three DOFs including wrist flexion-extension, abduction-adduction and forearm pronation-supination were investigated with 10 able-bodied subjects and two individuals with transradial limb deficiency (LD). A Fitts' law test involving real-time target acquisition tasks was conducted to compare the usability of the SVM-based control system to that of an artificial neural network (ANN) based method. Performance was assessed using the Fitts' law throughput value as well as additional metrics including completion rate, path efficiency and overshoot. The SVM-based approach outperformed the ANN-based system in every performance measure for able-bodied subjects. The SVM outperformed the ANN in path efficiency and throughput with the first LD subject and in throughput with the second LD subject. The superior performance of the SVM-based system appears to be due to its higher estimation accuracy of all DOFs during inactive and low amplitude segments (these periods were frequent during real-time control). Another advantage of the SVM-based method was that it substantially reduced the processing time for both training and real time control.