Search

Image detection deep learning models are being used in various fields such as autonomous vehicles, medical, and agriculture. In the defense field, various efforts are being made to apply deep learning models to surveillance systems. Since the defense field deals with important issues such as combat, the reliability of the model is important. Therefore, this study intends to propose a new "strong supervised learning method" to ensure explainability of the defense image classification model. The proposed method is using the knowledge of military experts and the HITL (Human In The Loop) method in which humans and AI interact. This is the learning that provides the data of important parts of objects in addition to the general supervised learning which provides input data and labeling data. The learning process for the additionally provided data is carried out through a loss function of SSIM (Structural Similarity Index Map), which measures the similarity between data of important parts and a feature map of a convolutional neural network. In order to evaluate the proposed methodology, the model learned by general supervised learning and the model learned by the proposed method were compared. The comparison criteria were as follows: first, evaluation of similarity between the results of Grad CAM, a Visual Explain method, along with the data the important parts designated by experts; and second, the classification accuracy of objects. As a result of the experiment using the VGG16, ResNet50, InceptionV3, and Xception models, the similarity between the data of the Grad CAM and the region of interest was improved by an average of 15.52% based on the entire image area. In addition, the classification accuracy showed an average improvement of 4.72% based on Test Set. [ABSTRACT FROM AUTHOR]

The increasing utilization of artificial intelligence algorithms in drug development has proven to be highly efficient and effective. One area where deep learning-based approaches have made significant contributions is in drug repositioning, enabling the identification of new therapeutic applications for existing drugs. In the present study, a trained deep-learning model was employed to screen a library of FDA-approved drugs to discover novel inhibitors targeting JAK2. To accomplish this, reference datasets containing active and decoy compounds specific to JAK2 were obtained from the DUD-E database. RDKit, a cheminformatic toolkit, was utilized to extract molecular features from the compounds. The DeepChem framework's GraphConvMol, based on graph convolutional network models, was applied to build a predictive model using the DUD-E datasets. Subsequently, the trained deep-learning model was used to predict the JAK2 inhibitory potential of FDA-approved drugs. Based on these predictions, ribociclib, topiroxostat, amodiaquine, and gefitinib were identified as potential JAK2 inhibitors. Notably, several known JAK2 inhibitors demonstrated high potential according to the prediction results, validating the reliability of our prediction model. To further validate these findings and confirm their JAK2 inhibitory activity, molecular docking experiments were conducted using tofacitinib—an FDA-approved drug for JAK2 inhibition. Experimental validation successfully confirmed our computational analysis results by demonstrating that these novel drugs exhibited comparable inhibitory activity against JAK2 compared to tofacitinib. In conclusion, our study highlights how deep learning models can significantly enhance virtual screening efforts in drug discovery by efficiently identifying potential candidates for specific targets such as JAK2. These newly discovered drugs hold promises as novel JAK2 inhibitors deserving further exploration and investigation. [ABSTRACT FROM AUTHOR]

Proving the "expectation-threshold" conjecture of Kahn and Kalai [Combin. Probab. Comput. 16 (2007), pp. 495–502], we show that for any increasing property \mathcal {F} on a finite set X, \[ p_c(\mathcal {F})=O(q(\mathcal {F})\log \ell (\mathcal {F})), \] where p_c(\mathcal {F}) and q(\mathcal {F}) are the threshold and "expectation threshold" of \mathcal {F}, and \ell (\mathcal {F}) is the maximum of 2 and the maximum size of a minimal member of \mathcal {F}. [ABSTRACT FROM AUTHOR]

Introduction: Adolescents are experiencing high rates of depressive symptoms, with negative consequences to their long‐term health. Group‐based, mindful self‐compassion programs show promise in mitigating the development of more significant depression in at‐risk adolescents. However, the lack of well‐designed, active control conditions has limited the ability to examine the efficacy of such interventions. Methods: Fifty‐nine adolescents (Mage = 15.81, 70% female) with subsyndromal depressive symptoms from the Southeastern US were randomized to group‐based Mindful Self‐Compassion for Teens (N = 30) or a newly developed active control Healthy Lifestyles group (N = 29) during 2018 and 2019. Participants attended 8 weekly "main" sessions followed by 6 monthly continuation sessions. The feasibility and acceptability of participation in both groups were measured using attrition, attendance, credibility, and satisfaction data. Depression scores were collected weekly, and self‐compassion scores were collected five times across 36 weeks. Results: Both groups were equally feasible and acceptable during the 8‐week program period; however, monthly continuation sessions were poorly attended in both groups. The risk of developing clinically significant depression was 2.6 times higher in the control group compared with the self‐compassion group (p =.037) across 36 weeks. Depression significantly decreased in the self‐compassion group, while it significantly increased in the control group. Both groups increased significantly in reports of self‐compassion. These findings are on par with results noting the efficacy of cognitive‐based interventions for high‐risk adolescents; follow‐up studies with larger sample sizes should be conducted to confirm these findings. Conclusions: Initial examination suggests Mindful Self‐Compassion for Teens programming is feasible, acceptable, and efficacious in preventing the development of clinically significant depression in adolescents with subsyndromal depression. Future studies may benefit from refinements to the self‐compassion measurement and/or the attention control condition; moreover, larger sample sizes are needed to confirm results. [ABSTRACT FROM AUTHOR]

[Display omitted] • A new affinity peptide against β-LG was successfully screened, designed, synthesized, and characterized. • A conductive TA–PEI network was incorporated into electrode for electrochemical sensing of β-LG. • The developed sensor demonstrated a linear correlation range of 50 to 750 ng with a K d value of 213.9 ng. • The sensor could detect β-LG in cow and goat milk, with mean recoveries of 88.5% and 92.2%. Harmful substances that cause food allergies can pose a significant threat to consumers along with food safety. According to the World Health Organization (WHO), approximately 10 % of the global population is currently affected by food allergies. Therefore, there is an urgent need for the development of more accurate and precise biosensors capable of detecting these hazardous substances including beta-lactoglobulin. Although numerous detection and analysis methods have been developed, they still suffer from various limitations. In this study, a tannic acid–polyethyleneimine (TA-PEI) network modified screen-printed electrodes (SPE) are newly developed and the binding sequence of peptide against β-LG was successfully screened using random peptide library. A novel affinity peptide with the desired sequence of S-L-S-P-S-L-W-Q-V-S-M-L-G-G-G-G-E-P-L-Q-L-K-M against β-lactoglobulin (β-LG) is designed and synthesized. The synthesized affinity peptide was immobilized on TA-PEI modified SPE to develop peptide-based sensor against β-LG for the first time. Under successful optimization, the developed sensor exhibited a linear relationship between 50 and 750 ng, with a K d of 213.9 ng. In addition, the sensor was able to detect β-LG in cow and goat milk, with average recoveries of 88.5 % and 92.2 %, respectively. [ABSTRACT FROM AUTHOR]

Piezoelectric biomaterials can generate piezoelectrical charges in response to mechanical activation. These generated charges can directly stimulate bone regeneration by triggering signaling pathway that is important for regulating osteogenesis of cells seeded on the materials. On the other hand, mechanical forces applied to the biomaterials play an important role in bone regeneration through the process called mechanotransduction. While mechanical force and electrical charges are both important contributing factors to bone tissue regeneration, they operate through different underlying mechanisms. The utilizations of piezoelectric biomaterials have been explored to serve as self-charged scaffolds which can promote stem cell differentiation and the formation of functional bone tissues. However, it is still not clear how mechanical activation and electrical charge act together on such a scaffold and which factors play more important role in the piezoelectric stimulation to induce osteogenesis. In our study, we found Poly(l-lactic acid) (PLLA)-based piezoelectric scaffolds with higher piezoelectric charges had a more pronounced osteoinductive effect than those with lower charges. This provided a new mechanistic insight that the observed osteoinductive effect of the piezoelectric PLLA scaffolds is likely due to the piezoelectric stimulation they provide, rather than mechanical stimulation alone. Our findings provide a crucial guide for the optimization of piezoelectric material design and usage. [ABSTRACT FROM AUTHOR]

Saxitoxin (STX), which is produced by certain dinoflagellate species, is a type of paralytic shellfish poisoning toxin that poses a serious threat to human health and the environment. Therefore, developing a technology for the convenient and cost-effective detection of STX is imperative. In this study, we developed an affinity peptide-imprinted polymer-based indirect competitive ELISA (ic-ELISA) without using enzyme–toxin conjugates. AuNP/Co 3 O 4 @Mg/Al cLDH was synthesized by calcining AuNP/ZIF-67@Mg/Al LDH, which was obtained by combining AuNPs, ZIF-67, and flower-like Mg/Al LDH. This synthesized nanozyme exhibited high catalytic activity (K m = 0.24 mM for TMB and 132.5 mM for H 2 O 2). The affinity peptide-imprinted polymer (MIP) was imprinted with an STX-specific template peptide (STX MIP) on a multi-well microplate and then reacted with an STX-specific signal peptide (STX SP). The interaction between the STX SP and MIP was detected using a streptavidin-coated nanozyme (SA-AuNP/Co 3 O 4 @Mg/Al cLDH). The developed MIP-based ic-ELISA exhibited excellent selectivity and sensitivity, with a limit of detection of 3.17 ng/mL (equivalent: 0.317 μg/g). Furthermore, the system was validated using a commercial ELISA kit and mussel tissue samples, and it demonstrated a high STX recovery with a low coefficient of variation. These results imply that the developed ic-ELISA can be used to detect STX in real samples. [ABSTRACT FROM AUTHOR]

[Display omitted] • USP2a, a member of the ubiquitin-specific protease (USP) family, is a promising target for cancer therapy. • This study optimized the structure of ML364, leading to the identification of compound 8v as a potent USP2a inhibitor. • Compound 8v selectively reduced USP2a substrate levels and inhibited cancer cell growth. Ubiquitination is a representative post-translational modification that tags target proteins with ubiquitin to induce protein degradation or modify their functions. Deubiquitinating enzymes (DUBs) play a crucial role in reversing this process by removing ubiquitin from target proteins. Among them, USP2a has emerged as a promising target for cancer therapy due to its oncogenic properties in various cancer types, but its inhibitors have been limited. In this study, our aim was to optimize the structure of ML364, a USP2a inhibitor, and synthesize a series of its derivatives to develop potent USP2a inhibitors. Compound 8v emerged as a potential USP2a inhibitor with lower cytotoxicity compared to ML364. Cellular assays demonstrated that compound 8v effectively reduced the levels of USP2a substrates and attenuated cancer cell growth. We confirmed its direct interaction with the catalytic domain of USP2a and its selective inhibitory activity against USP2a over other USP subfamily proteins (USP7, 8, or 15). In conclusion, compound 8v has been identified as a potent USP2a inhibitor with substantial potential for cancer therapy. [ABSTRACT FROM AUTHOR]