Depression detection from sMRI and rs-fMRI images using machine learning

Major Depression Disorder (MDD) is a common mental disorder that negatively affects many people’s lives worldwide. Developing an automated method to find useful diagnostic biomarkers from brain imaging data would help clinicians to detect MDD in its early stages. Depression is known to be a brain connectivity disorder problem. In this paper, we present a brain connectivity-based machine learning (ML) workflow that utilizes similarity/dissimilarity of spatial cubes in brain MRI images as features for depression detection. The proposed workflow provides a unified framework applicable to both structural MRI images and resting-state functional MRI images. Several cube similarity measures have been explored, including Pearson or Spearman correlations, Minimum Distance Covariance, or inverse of Minimum Distance Covariance. Discriminative features from the cube similarity matrix are chosen with the Wilcoxon rank-sum test. The extracted features are fed into machine learning classifiers to train MDD prediction models. To address the challenge of data imbalance in MDD detection, oversampling is performed to balance the training data. The proposed workflow is evaluated through experiments on three independent public datasets, all imbalanced, of structural MRI and resting-state fMRI images with depression labels. Experimental results show good performance on all three datasets in terms of prediction accuracy, specificity, sensitivity, and area under the Receiver Operating Characteristic (ROC) curve. The use of features from both structured MRI and resting state functional MRI is also investigated.

     

Acquiring reusable skills in intrinsically motivated reinforcement learning

Journal of Intelligent Manufacturing - This paper proposes a novel incremental model for acquiring skills and using them in Intrinsically Motivated Reinforcement Learning (IMRL). In this model, the...     

A robust spread spectrum based image watermarking in ridgelet domain

A new robust method of spread spectrum based image watermarking is proposed in this article. Spread spectrum technique and scrambling are used for increasing robustness and invisibility of the algorithm. Our suggested method is carried out using ridgelet transform as an efficient transform for representing images with line singularities. In embedding part, the host image is partitioned into non-overlapping blocks and ridgelet transform is applied to each single block. In this way, a curved edge is divided into some straight edges so that ridgelet transform shows optimal performance even for complicated images with curve edges. To embed the watermark bits, the best directions of ridgelet coefficients are selected with respect to their variance intensity. In extraction part, a computationally efficient detection method is used for detecting watermark logo blindly from distorted watermarked image. To achieve more robust algorithm firstly, we find the best place to insert the watermark bits and secondly, we encode the scrambled watermark bits by pseudo random sequences with an authentication key. Robustness of our proposed method is tested against different kinds of attacks. According to the experimental results, proposed method shows much improved performance in comparison to other published works.     

Multi‐“Color” Delineation of Bone Microdamages Using Ligand‐Directed Sub‐5 nm Hafnia Nanodots and Photon Counting CT Imaging

The early detection of bone microdamages is crucial to make informed decisions about the therapy and taking precautionary treatments to avoid catastrophic fractures. Conventional computed tomography (CT) imaging faces obstacles in detecting bone microdamages due to the strong self‐attenuation of photons from bone and poor spatial resolution. Recent advances in CT technology as well as novel imaging probes can address this problem effectively. Herein, the bone microdamage imaging is demonstrated using ligand‐directed nanoparticles in conjunction with photon counting spectral CT. For the first time, Gram‐scale synthesis of hafnia (HfO₂) nanoparticles is reported with surface modification by a chelator moiety. The feasibility of delineating these nanoparticles from bone and soft tissue of muscle is demonstrated with photon counting spectral CT equipped with advanced detector technology. The ex vivo and in vivo studies point to the accumulation of hafnia nanoparticles at microdamage site featuring distinct spectral signal. Due to their small sub‐5 nm size, hafnia nanoparticles are excreted through reticuloendothelial system organs without noticeable aggregation while not triggering any adverse side effects based on histological and liver enzyme function assessments. These preclinical studies highlight the potential of HfO₂‐based nanoparticle contrast agents for skeletal system diseases due to their well‐placed K‐edge binding energy.     

Protein Structural Changes During Preparation and Storage of Surimi

The changes in protein structure associated with the preparation and frozen storage of surimi were investigated. Raw surimi was prepared by repeatedly washing Alaska pollock flesh with chilled water. The product was either slowly frozen or underwent rapid freezing using liquid air; in either case it was then subjected to frozen storage at ‐20 °C for 24 mo. Fourier transform infrared/attenuated total reflectance (FTIR/ATR) spectroscopy showed that during preparation of surimi, the a‐helix content increased with increased number of washing cycles. Differential scanning calorimetry (DSC) revealed a shift in the thermal transition of actin to a higher temperature during surimi preparation. Electrophoresis, FTIR/ATR spectroscopy, and DSC results revealed a loss of myofibrillar proteins from surimi after 3 washing cycles, suggesting that 3 washing cycles were adequate to prepare surimi. Sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) showed relatively minor changes in protein subunit structure with some loss of the myosin light chains (MLC); myosin heavy chain (MHC), actin, and tropomyosin were found to be relatively stable. Native‐PAGE showed no major changes in surimi after 24 mo storage at ‐20 °C. FTIR/ ATR spectroscopy indicated a significant decrease in a‐helix relative to p‐sheet structure in surimi after 2 y of storage at ‐20 °C. The loss of α‐helical content was more significant in slowly frozen surimi compared with rapid‐frozen surimi samples. DSC results revealed a shift in the thermal transition of actin to lower temperatures during frozen storage of surimi.     

Hydroxymethylfurfural in fruit puree and juice: preconcentration and determination using microextraction method coupled with high-performance liquid chromatography and optimization by Box–Behnken design

In the present study, a sensitive and rapid method for separation and determination of hydroxymethylfurfural (HMF) in fruit puree and juices was proposed. Dispersive liquid–liquid microextraction (DLLME) coupled with high-performance liquid chromatography (HPLC) was used for extraction and quantitative determination of HMF in fruit puree and juices. The effective parameters such as the type and volume of extraction and dispersive solvents, pH and salt amount (NaCl) were studied and optimized with the aid of response surface methodology based on Box–Behnken design to obtain the best condition for HMF extraction. At the optimized conditions, parameter values were 60 µL extracting solvent, 600 µL dispersive solvent, 2 g NaCl and pH 5. Repeatability of the method, described as the relative standard deviation, was 3.1% (n?=?6) and the recovery was 98.4%. The limit of detection and limit of quantitation were 1.47 and 5.28 µg L^?1, respectively. The merit figures of DLLME–HPLC–UV method showed that the proposed method can be noticed as a new, fast and good alternative method for investigation of HMF in various fruit puree and juice samples.     

Numerical simulation of the wall shear stress distribution in a carotid artery bifurcation

Journal of Mechanical Science and Technology - The effect of stenosis for a carotid artery bifurcation with elastic and rigid walls is investigated numerically. In the present study, the blood flow...     

Energy provisioning in wireless rechargeable sensor networks with limited knowledge

Internet of Things in many applications depends on Wireless Sensor Networks where the sensors are battery powered. Recent advances in wireless energy transfer and rechargeable batteries provide a new chance for Wireless Rechargeable Sensor Networks when the mobile chargers (MCs) patrol the network field and replenish the power of sensors. We consider multiple MCs and a few charging stations (CSs) in the network. The MCs lose their power too, so they move toward CSs to replenish the energy of themselves. We propose an approach named Limited Knowledge Charging (LKC) where each CS makes a virtual area by using grid cells. Based on the cell’s information, CSs coordinate among themselves to direct MCs in the network. The main design goal of LKC is to prolong the network lifetime, by using many techniques such as balancing the energy of network areas. LKC reduces movements of MCs too as a second goal. LKC is an online approach that adapts itself with situation changes of the network. Many related studies use global knowledge, which is not always satisfied in practice. Instead, LKC is a local knowledge approach. Using exhaustive simulation, the satisfaction of the design goals of LKC is demonstrated.

     

Characterization of the produced electrospun fish gelatin nanofiber containing fucoxanthin

Food Science and Biotechnology - This study aims to prepare fish gelatin nanofibers extracted from fish waste by using electrospinning method and its encapsulation with fucoxanthin extracted from...