Search

Predicting early treatment response in schizophrenia is pivotal for selecting the best therapeutic approach. Utilizing machine learning (ML) technique, we aimed to formulate a model predicting antipsychotic treatment outcomes. Data were obtained from 299 patients with schizophrenia from three multicenter, open-label, non-comparative clinical trials. For prediction of treatment response at weeks 4, 8, and 24, psychopathology (both objective and subjective symptoms), sociodemographic and clinical factors, functional outcomes, attitude toward medication, and metabolic characteristics were evaluated. Various ML techniques were applied. The highest area under the curve (AUC) at weeks 4, 8 and 24 was 0.711, 0.664 and 0.678 with extreme gradient boosting, respectively. Notably, our findings indicate that BMI and attitude toward medication play a pivotal role in predicting treatment responses at all-time points. Other salient features for weeks 4 and 8 included psychosocial functioning, negative symptoms, subjective symptoms like psychoticism and hostility, and the level of prolactin. For week 24, positive symptoms, depression, education level and duration of illness were also important. This study introduced a precise clinical model for predicting schizophrenia treatment outcomes using multiple readily accessible predictors. The findings underscore the significance of metabolic parameters and subjective traits., Competing Interests: Declaration of competing interest The authors have no potential conflicts of interest to disclose., (Copyright © 2024 Elsevier B.V. All rights reserved.)

Pneumonia is a significant adverse drug reaction (ADR) associated with clozapine, characterized by high mortality and potential linkage with other inflammatory responses. Despite the critical nature, research regarding the development of pneumonia during initial clozapine titration remains limited. This retrospective study included 1408 Korean inpatients with schizophrenia spectrum disorders. Data were collected from January 2000 to January 2023. Pneumonia developed in 3.5 % of patients within 8 weeks of clozapine initiation. Patients who developed pneumonia were taking a greater number and higher dose of antipsychotics at baseline (2.14 vs. 1.58, p < 0.001; 25.64 vs. 19.34, p = 0.012). The average onset occurred 17.24 days after initiation, on an average dose of 151.28 mg/day. Titration was either paused or slowed in most of these patients, with no reported fatalities. The types of pneumonia included aspiration pneumonia, mycoplasma pneumonia, bronchopneumonia, and COVID-19 pneumonia. Myocarditis, drug reaction with eosinophilia and systemic symptoms (DRESS) syndrome, and urinary tract infections were also identified. Logistic regression analysis revealed that a greater number of concomitant antipsychotics (odds ratio [OR] = 1.59, p = 0.027) and concomitant benzodiazepine use (OR = 2.33, p = 0.005) at baseline were associated with an increased risk of pneumonia. Overall, pneumonia development during clozapine titration is linked with other inflammatory ADRs, suggesting a shared immunological mechanism. Close monitoring is recommended, especially for patients taking multiple antipsychotics and benzodiazepines. Further studies involving repeated measures of clozapine concentrations at trough and steady state, along with a more detailed description of pneumonia types, are warranted., Competing Interests: Declaration of competing interest The authors have no conflicts of interest concerning the subject of the study., (Copyright © 2023 Elsevier B.V. All rights reserved.)

Safe and effective administration of clozapine requires careful monitoring for inflammatory reactions during the initial titration. The concentration-to-dose (C/D) ratio must be taken into account, which may vary among ethnicities. In this retrospective study, 1408 Korean schizophrenia inpatients were examined for during the first 8 weeks of clozapine titration. The average doses of clozapine administered during weeks 1, 2, 4, and 8 were 77.37, 137.73, 193.20, and 212.83 mg/day, with significantly lower doses for females than males. The average C/D ratio was significantly higher in females (1.75 ± 1.04 and 1.11 ± 0.67 ng/mL per mg/day). Patients with higher C/D ratios were more likely to experience fever and were prescribed lower doses of clozapine starting from week 4. In total, 22.1 % of patients developed a fever at an average of 15.74 days after initiating clozapine. Patients who developed a fever were younger, used more antipsychotics at baseline, had a higher C/D ratio, and had a higher incidence of an elevated C-reactive protein level. A higher C/D ratio, use of a greater number of antipsychotics at baseline, and concomitant olanzapine use were risk factors for the development of inflammatory reactions. The incidence of pneumonia, agranulocytosis, and myocarditis within 8 weeks were 3.7 %, 0.3 %, and 0.1 %. In summary, the target dose of clozapine titration is lower for Korean schizophrenia patients, with a higher C/D ratio and more frequent fever compared to Western patients; however, myocarditis occurs rarely. Our findings may contribute to the titration methods for clozapine for the East Asian population., Competing Interests: Declaration of competing interest The authors have no conflicts of interest concerning the subject of the study., (Copyright © 2023 Elsevier B.V. All rights reserved.)

Increasing the number of satellites in a global navigation satellite system (GNSS) improves the positioning accuracy and increases availability. However, it reduces the positioning accuracy improvement rate and increases the calculation loads, which can cause battery usage problems in mobile devices using a GNSS. An appropriate satellite selection method is required. One current method entails the use of ideal satellite placement with respect to the minimum geometric dilution of precision (GDOP). In this study, the described ideal satellite placement with the minimum GDOP were divided in terms of the horizontal dilution of precision (HDOP) and vertical dilution of precision (VDOP). HDOP and VDOP were mathematically derived and analyzed. The derived formula was verified using simulations. The analysis was performed with actual dual GNSS satellite data. The satellites adjacent to the ideal placement were selected and the DOP was calculated. Simply selecting satellites closest to the ideal placement afforded large values for HDOP and VDOP. This issue was addressed using a satellite changing algorithm considering the dual GNSS, resulting in reduced values of the HDOP and VDOP.

In this study, we present a neural network that consists of nodes with heterogeneous sensitivity. Each node in a network is assigned a variable that determines the sensitivity with which it learns to perform a given task. The network is trained via a constrained optimization that maximizes the sparsity of the sensitivity variables while ensuring optimal network performance. As a result, the network learns to perform a given task using only a few sensitive nodes. Insensitive nodes, which are nodes with zero sensitivity, can be removed from a trained network to obtain a computationally efficient network. Removing zero-sensitivity nodes has no effect on the performance of the network because the network has already been trained to perform the task without them. The regularization parameter used to solve the optimization problem was simultaneously found during the training of the networks. To validate our approach, we designed networks with computationally efficient architectures for various tasks such as autoregression, object recognition, facial expression recognition, and object detection using various datasets. In our experiments, the networks designed by our proposed method provided the same or higher performances but with far less computational complexity., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

In this study, we present deep neural networks with a set of node-wise varying activation functions. The feature-learning abilities of the nodes are affected by the selected activation functions, where the nodes with smaller indices become increasingly more sensitive during training. As a result, the features learned by the nodes are sorted by the node indices in order of their importance such that more sensitive nodes are related to more important features. The proposed networks learn input features but also the importance of the features. Nodes with lower importance in the proposed networks can be pruned to reduce the complexity of the networks, and the pruned networks can be retrained without incurring performance losses. We validated the feature-sorting property of the proposed method using both shallow and deep networks as well as deep networks transferred from existing networks., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

This paper presents a recurrent learning-based facial attribute recognition method that mimics human observers' visual fixation. The concentrated views of a human observer while focusing and exploring parts of a facial image over time are generated and fed into a recurrent network. The network makes a decision concerning facial attributes based on the features gleaned from the observer's visual fixations. Experiments on facial expression, gender, and age datasets show that applying visual fixation to recurrent networks improves recognition rates significantly. The proposed method not only outperforms state-of-the-art recognition methods based on static facial features, but also those based on dynamic facial features.

Electrically tunable colloidal photonic crystals (ETPCs) have been investigated because of several merits such as easy color tunability, no discoloration, and clear color. The coloration mechanism of ETPCs has been explained in terms of only the electric field. Herein, we report on a new mechanism: electric field plus redox reaction. Specifically, the coloration behavior of ETPCs was investigated under electrically conductive or insulated conditions using current-voltage, cyclic voltammetry, and zeta potential measurements, as well as scanning electron microscopy. Electrophoretic movement of ETPC particles toward the positive electrode was caused by the electric field due to the particles' negative surface charge. At the positive electrode, ETPC particles lost their electrons and formed a colloidal crystal structure. Finally, an ETPC transparent tube device was constructed to demonstrate the coloration mechanism.

Due to the outstanding mechanical properties of individual carbon nanotubes (CNTs) at the nanoscale, CNT yarns are expected to demonstrate high strength at the macroscale. In this study, a predictable model was developed to predict the tensile strength of twisted CNT yarns. First, the failure mechanism of twisted CNT yarns was investigated using in situ tensile tests and ex situ observations. It was revealed that CNT bundles, which are groups of CNTs that are tightly bound together, formed during tensile loading, leaving some voids around the bundles. Failure of the CNT yarns occurred as the CNT bundles were pulled out of the yarns. Two stresses that determined the tensile strength of the CNT yarns were identified: interfacial shear and frictional stresses originating from van der Waals interactions, and the lateral pressure generated by the twisted yarn structure. Molecular dynamics and yarn mechanics were used to calculate these two stresses. Finally, the tensile strength of CNT yarns was predicted and compared with experimental data, showing reasonable agreement.

Images acquired by a camera show lens blur due to imperfection in the optical system even when images are properly focused. Lens blur is non-stationary in a sense that the amount of blur depends on pixel locations in a sensor. Lens blur is also asymmetric in a sense that the amount of blur is different in the radial and tangential directions, and also in the inward and outward radial directions. This paper presents parametric blur kernel models based on the normal sinh-arcsinh distribution function. The proposed models can provide flexible shapes of blur kernels with a different symmetry and skewness to model complicated lens blur due to optical aberration in a properly focused images accurately. Blur of single focal length lenses is estimated, and the accuracy of the models is compared with the existing parametric blur models. An advantage of the proposed models is demonstrated through deblurring experiments.