Deep Transfer Learning Using Real-World Image Features for Medical Image Classification, with a Case Study on Pneumonia X-ray Images.

Deep learning has profoundly influenced various domains, particularly medical image analysis. Traditional transfer learning approaches in this field rely on models pretrained on domain-specific medical datasets, which limits their generalizability and accessibility. In this study, we propose a novel framework called real-world feature transfer learning, which utilizes backbone models initially trained on large-scale general-purpose datasets such as ImageNet. We evaluate the effectiveness and robustness of this approach compared to models trained from scratch, focusing on the task of classifying pneumonia in X-ray images. Our experiments, which included converting grayscale images to RGB format, demonstrate that real-world-feature transfer learning consistently outperforms conventional training approaches across various performance metrics. This advancement has the potential to accelerate deep learning applications in medical imaging by leveraging the rich feature representations learned from general-purpose pretrained models. The proposed methodology overcomes the limitations of domain-specific pretrained models, thereby enabling accelerated innovation in medical diagnostics and healthcare. From a mathematical perspective, we formalize the concept of real-world feature transfer learning and provide a rigorous mathematical formulation of the problem. Our experimental results provide empirical evidence supporting the effectiveness of this approach, laying the foundation for further theoretical analysis and exploration. This work contributes to the broader understanding of feature transferability across domains and has significant implications for the development of accurate and efficient models for medical image analysis, even in resource-constrained settings.

Ceramide promotes lytic reactivation of Epstein-Barr virus in gastric carcinoma.

Epstein-Barr virus (EBV) has a lifelong latency period after initial infection. Rarely, however, when the EBV immediate early gene BZLF1 is expressed by a specific stimulus, the virus switches to the lytic cycle to produce progeny viruses. We found that EBV infection reduced levels of various ceramide species in gastric cancer cells. As ceramide is a bioactive lipid implicated in the infection of various viruses, we assessed the effect of ceramide on the EBV lytic cycle. Treatment with C6-ceramide (C6-Cer) induced an increase in the endogenous ceramide pool and increased production of the viral product as well as BZLF1 expression. Treatment with the ceramidase inhibitor ceranib-2 induced EBV lytic replication with an increase in the endogenous ceramide pool. The glucosylceramide synthase inhibitor Genz-123346 inhibited C6-Cer-induced lytic replication. C6-Cer induced extracellular signal-regulated kinase 1/2 (ERK1/2) and CREB phosphorylation, c-JUN expression, and accumulation of the autophagosome marker LC3B. Treatment with MEK1/2 inhibitor U0126, siERK1&2, or siCREB suppressed C6-Cer-induced EBV lytic replication and autophagy initiation. In contrast, siJUN transfection had no impact on BZLF1 expression. The use of 3-methyladenine (3-MA), an inhibitor targeting class III phosphoinositide 3-kinases (PI3Ks) to inhibit autophagy initiation, resulted in reduced beclin-1 expression, along with suppressed C6-Cer-induced BZLF1 expression and LC3B accumulation. Chloroquine, an inhibitor of autophagosome-lysosome fusion, increased BZLF1 protein intensity and LC3B accumulation. However, siLC3B transfection had minimal effect on BZLF1 expression. The results suggest the significance of ceramide-related sphingolipid metabolism in controlling EBV latency, highlighting the potential use of drugs targeting sphingolipid metabolism for treating EBV-positive gastric cancer.IMPORTANCEEpstein-Barr virus remains dormant in the host cell but occasionally switches to the lytic cycle when stimulated. However, the exact molecular mechanism of this lytic induction is not well understood. In this study, we demonstrate that Epstein-Barr virus infection leads to a reduction in ceramide levels. Additionally, the restoration of ceramide levels triggers lytic replication of Epstein-Barr virus with increase in phosphorylation of extracellular signal-regulated kinase 1/2 (ERK1/2) and CREB. Our study suggests that the Epstein-Barr virus can inhibit lytic replication and remain latent through reduction of host cell ceramide levels. This study reports the regulation of lytic replication by ceramide in Epstein-Barr virus-positive gastric cancer.

Combination Analysis of PCDHGA12 and CDO1 DNA Methylation in Bronchial Washing Fluid for Lung Cancer Diagnosis.

BACKGROUND: When suspicious lesions are observed on computer-tomography (CT), invasive tests are needed to confirm lung cancer. Compared with other procedures, bronchoscopy has fewer complications. However, the sensitivity of peripheral lesion through bronchoscopy including washing cytology is low. A new test with higher sensitivity through bronchoscopy is needed. In our previous study, DNA methylation of PCDHGA12 in bronchial washing cytology has a diagnostic value for lung cancer. In this study, combination of PCDHGA12 and CDO1 methylation obtained through bronchial washing cytology was evaluated as a diagnostic tool for lung cancer. METHODS: A total of 187 patients who had suspicious lesions in CT were enrolled. PCDHGA12 methylation test, CDO1 methylation test, and cytological examination were performed using 3-plex LTE-qMSP test. RESULTS: Sixty-two patients were diagnosed with benign diseases and 125 patients were diagnosed with lung cancer. The sensitivity of PCDHGA12 was 74.4% and the specificity of PCDHGA12 was 91.9% respectively. CDO1 methylation test had a sensitivity of 57.6% and a specificity of 96.8%. The combination of both PCDHGA12 methylation test and CDO1 methylation test showed a sensitivity of 77.6% and a specificity of 90.3%. The sensitivity of lung cancer diagnosis was increased by combining both PCDHGA12 and CDO1 methylation tests. CONCLUSION: Checking DNA methylation of both PCDHGA12 and CDO1 genes using bronchial washing fluid can reduce the invasive procedure to diagnose lung cancer.

MetaSwin: a unified meta vision transformer model for medical image segmentation.

Transformers have demonstrated significant promise for computer vision tasks. Particularly noteworthy is SwinUNETR, a model that employs vision transformers, which has made remarkable advancements in improving the process of segmenting medical images. Nevertheless, the efficacy of training process of SwinUNETR has been constrained by an extended training duration, a limitation primarily attributable to the integration of the attention mechanism within the architecture. In this article, to address this limitation, we introduce a novel framework, called the MetaSwin model. Drawing inspiration from the MetaFormer concept that uses other token mix operations, we propose a transformative modification by substituting attention-based components within SwinUNETR with a straightforward yet impactful spatial pooling operation. Additionally, we incorporate of Squeeze-and-Excitation (SE) blocks after each MetaSwin block of the encoder and into the decoder, which aims at segmentation performance. We evaluate our proposed MetaSwin model on two distinct medical datasets, namely BraTS 2023 and MICCAI 2015 BTCV, and conduct a comprehensive comparison with the two baselines, i.e., SwinUNETR and SwinUNETR+SE models. Our results emphasize the effectiveness of MetaSwin, showcasing its competitive edge against the baselines, utilizing a simple pooling operation and efficient SE blocks. MetaSwin's consistent and superior performance on the BTCV dataset, in comparison to SwinUNETR, is particularly significant. For instance, with a model size of 24, MetaSwin outperforms SwinUNETR's 76.58% Dice score using fewer parameters (15,407,384 vs 15,703,304) and a substantially reduced training time (300 vs 467 mins), achieving an improved Dice score of 79.12%. This research highlights the essential contribution of a simplified transformer framework, incorporating basic elements such as pooling and SE blocks, thus emphasizing their potential to guide the progression of medical segmentation models, without relying on complex attention-based mechanisms.

Recent Advancements in Deep Learning Using Whole Slide Imaging for Cancer Prognosis.

This review furnishes an exhaustive analysis of the latest advancements in deep learning techniques applied to whole slide images (WSIs) in the context of cancer prognosis, focusing specifically on publications from 2019 through 2023. The swiftly maturing field of deep learning, in combination with the burgeoning availability of WSIs, manifests significant potential in revolutionizing the predictive modeling of cancer prognosis. In light of the swift evolution and profound complexity of the field, it is essential to systematically review contemporary methodologies and critically appraise their ramifications. This review elucidates the prevailing landscape of this intersection, cataloging major developments, evaluating their strengths and weaknesses, and providing discerning insights into prospective directions. In this paper, a comprehensive overview of the field aims to be presented, which can serve as a critical resource for researchers and clinicians, ultimately enhancing the quality of cancer care outcomes. This review's findings accentuate the need for ongoing scrutiny of recent studies in this rapidly progressing field to discern patterns, understand breakthroughs, and navigate future research trajectories.

Machine Learning Big Data Analysis of the Impact of Air Pollutants on Rhinitis-Related Hospital Visits.

This study seeks to elucidate the intricate relationship between various air pollutants and the incidence of rhinitis in Seoul, South Korea, wherein it leveraged a vast repository of data and machine learning techniques. The dataset comprised more than 93 million hospital visits (n = 93,530,064) by rhinitis patients between 2013 and 2017. Daily atmospheric measurements were captured for six major pollutants: PM10, PM2.5, O3, NO2, CO, and SO2. We employed traditional correlation analyses alongside machine learning models, including the least absolute shrinkage and selection operator (LASSO), random forest (RF), and gradient boosting machine (GBM), to dissect the effects of these pollutants and the potential time lag in their symptom manifestation. Our analyses revealed that CO showed the strongest positive correlation with hospital visits across all three categories, with a notable significance in the 4-day lag analysis. NO2 also exhibited a substantial positive association, particularly with outpatient visits and hospital admissions and especially in the 4-day lag analysis. Interestingly, O3 demonstrated mixed results. Both PM10 and PM2.5 showed significant correlations with the different types of hospital visits, thus underlining their potential to exacerbate rhinitis symptoms. This study thus underscores the deleterious impacts of air pollution on respiratory health, thereby highlighting the importance of reducing pollutant levels and developing strategies to minimize rhinitis-related hospital visits. Further research considering other environmental factors and individual patient characteristics will enhance our understanding of these intricate dynamics.

Discovering potential pathways between type 2 diabetes mellitus and diabetic retinopathy: A big data analysis of the South Korean National Sample Cohort.

Diabetes mellitus, a prevalent metabolic disorder, is associated with a multitude of complications that necessitate vigilant management post-diagnosis. A notable complication, diabetic retinopathy, could lead to intense ocular injury, including vision impairment and blindness, due to the impact of the disease. Studying the transition from diabetes to diabetic retinopathy is paramount for grasping and halting the progression of complications. In this study, we examine the statistical correlation between type 2 diabetes mellitus and retinal disorders classified elsewhere, ultimately proposing a comprehensive disease network. The National Sample Cohort of South Korea, containing approximately 1 million samples and primary diagnoses based on the International Statistical Classification of Diseases and Related Health Problems 10th Revision classification, was utilized for this retrospective analysis. The diagnoses of both conditions displayed a statistically significant correlation with a chi-square test value of P < .001, and the t test for the initial diagnosis date also yielded a P < .001 value. The devised network, comprising 27 diseases and 142 connections, was established through statistical evaluations. This network offers insight into potential pathways leading to diabetic retinopathy and intermediary diseases, encouraging medical researchers to further examine various risk factors associated with these connections.

Deep Learning Techniques with Genomic Data in Cancer Prognosis: A Comprehensive Review of the 2021-2023 Literature.

Deep learning has brought about a significant transformation in machine learning, leading to an array of novel methodologies and consequently broadening its influence. The application of deep learning in various sectors, especially biomedical data analysis, has initiated a period filled with noteworthy scientific developments. This trend has majorly influenced cancer prognosis, where the interpretation of genomic data for survival analysis has become a central research focus. The capacity of deep learning to decode intricate patterns embedded within high-dimensional genomic data has provoked a paradigm shift in our understanding of cancer survival. Given the swift progression in this field, there is an urgent need for a comprehensive review that focuses on the most influential studies from 2021 to 2023. This review, through its careful selection and thorough exploration of dominant trends and methodologies, strives to fulfill this need. The paper aims to enhance our existing understanding of applications of deep learning in cancer survival analysis, while also highlighting promising directions for future research. This paper undertakes aims to enrich our existing grasp of the application of deep learning in cancer survival analysis, while concurrently shedding light on promising directions for future research in this vibrant and rapidly proliferating field.

Transformer Architecture and Attention Mechanisms in Genome Data Analysis: A Comprehensive Review.

The emergence and rapid development of deep learning, specifically transformer-based architectures and attention mechanisms, have had transformative implications across several domains, including bioinformatics and genome data analysis. The analogous nature of genome sequences to language texts has enabled the application of techniques that have exhibited success in fields ranging from natural language processing to genomic data. This review provides a comprehensive analysis of the most recent advancements in the application of transformer architectures and attention mechanisms to genome and transcriptome data. The focus of this review is on the critical evaluation of these techniques, discussing their advantages and limitations in the context of genome data analysis. With the swift pace of development in deep learning methodologies, it becomes vital to continually assess and reflect on the current standing and future direction of the research. Therefore, this review aims to serve as a timely resource for both seasoned researchers and newcomers, offering a panoramic view of the recent advancements and elucidating the state-of-the-art applications in the field. Furthermore, this review paper serves to highlight potential areas of future investigation by critically evaluating studies from 2019 to 2023, thereby acting as a stepping-stone for further research endeavors.

Machine learning for small interfering RNAs: a concise review of recent developments.

The advent of machine learning and its subsequent integration into small interfering RNA (siRNA) research heralds a new epoch in the field of RNA interference (RNAi). This review emphasizes the urgency and relevance of assimilating the plethora of contributions and advancements in this domain, particularly focusing on the period of 2019-2023. Given the rapid progression of deep learning technologies, our synthesis of recent research is paramount to staying apprised of the state-of-the-art methods being utilized. It not only offers a comprehensive insight into the confluence of machine learning and siRNA but also serves as a beacon, guiding future explorations in this intersectional research field. Our rigorous examination of studies promises a discerning perspective on the contemporary landscape of machine learning applications in siRNA design and function. This review is an effort to foster further discourse and propel academic inquiry in this multifaceted domain.

Recent Advances in Deep Learning for Protein-Protein Interaction Analysis: A Comprehensive Review.

Deep learning, a potent branch of artificial intelligence, is steadily leaving its transformative imprint across multiple disciplines. Within computational biology, it is expediting progress in the understanding of Protein-Protein Interactions (PPIs), key components governing a wide array of biological functionalities. Hence, an in-depth exploration of PPIs is crucial for decoding the intricate biological system dynamics and unveiling potential avenues for therapeutic interventions. As the deployment of deep learning techniques in PPI analysis proliferates at an accelerated pace, there exists an immediate demand for an exhaustive review that encapsulates and critically assesses these novel developments. Addressing this requirement, this review offers a detailed analysis of the literature from 2021 to 2023, highlighting the cutting-edge deep learning methodologies harnessed for PPI analysis. Thus, this review stands as a crucial reference for researchers in the discipline, presenting an overview of the recent studies in the field. This consolidation helps elucidate the dynamic paradigm of PPI analysis, the evolution of deep learning techniques, and their interdependent dynamics. This scrutiny is expected to serve as a vital aid for researchers, both well-established and newcomers, assisting them in maneuvering the rapidly shifting terrain of deep learning applications in PPI analysis.

Deep learning in CRISPR-Cas systems: a review of recent studies.

In genetic engineering, the revolutionary CRISPR-Cas system has proven to be a vital tool for precise genome editing. Simultaneously, the emergence and rapid evolution of deep learning methodologies has provided an impetus to the scientific exploration of genomic data. These concurrent advancements mandate regular investigation of the state-of-the-art, particularly given the pace of recent developments. This review focuses on the significant progress achieved during 2019-2023 in the utilization of deep learning for predicting guide RNA (gRNA) activity in the CRISPR-Cas system, a key element determining the effectiveness and specificity of genome editing procedures. In this paper, an analytical overview of contemporary research is provided, with emphasis placed on the amalgamation of artificial intelligence and genetic engineering. The importance of our review is underscored by the necessity to comprehend the rapidly evolving deep learning methodologies and their potential impact on the effectiveness of the CRISPR-Cas system. By analyzing recent literature, this review highlights the achievements and emerging trends in the integration of deep learning with the CRISPR-Cas systems, thus contributing to the future direction of this essential interdisciplinary research area.

Deep Learning Approaches for lncRNA-Mediated Mechanisms: A Comprehensive Review of Recent Developments.

This review paper provides an extensive analysis of the rapidly evolving convergence of deep learning and long non-coding RNAs (lncRNAs). Considering the recent advancements in deep learning and the increasing recognition of lncRNAs as crucial components in various biological processes, this review aims to offer a comprehensive examination of these intertwined research areas. The remarkable progress in deep learning necessitates thoroughly exploring its latest applications in the study of lncRNAs. Therefore, this review provides insights into the growing significance of incorporating deep learning methodologies to unravel the intricate roles of lncRNAs. By scrutinizing the most recent research spanning from 2021 to 2023, this paper provides a comprehensive understanding of how deep learning techniques are employed in investigating lncRNAs, thereby contributing valuable insights to this rapidly evolving field. The review is aimed at researchers and practitioners looking to integrate deep learning advancements into their lncRNA studies.

SuperstarGAN: Generative adversarial networks for image-to-image translation in large-scale domains.

Image-to-image translation with generative adversarial networks (GANs) has been extensively studied in recent years. Among the models, StarGAN has achieved image-to-image translation for multiple domains with a single generator, whereas conventional models require multiple generators. However, StarGAN has several limitations, including the lack of capacity to learn mappings among large-scale domains; furthermore, StarGAN can barely express small feature changes. To address the limitations, we propose an improved StarGAN, namely SuperstarGAN. We adopted the idea, first proposed in controllable GAN (ControlGAN), of training an independent classifier with the data augmentation techniques to handle the overfitting problem in the classification of StarGAN structures. Since the generator with a well-trained classifier can express small features belonging to the target domain, SuperstarGAN achieves image-to-image translation in large-scale domains. Evaluated with a face image dataset, SuperstarGAN demonstrated improved performance in terms of Fréchet Inception distance (FID) and learned perceptual image patch similarity (LPIPS). Specifically, compared to StarGAN, SuperstarGAN exhibited decreased FID and LPIPS by 18.1% and 42.5%, respectively. Furthermore, we conducted an additional experiment with interpolated and extrapolated label values, indicating the ability of SuperstarGAN to control the degree of expression of the target domain features in generated images. Additionally, SuperstarGAN was successfully adapted to an animal face dataset and a painting dataset, where it can translate styles of animal faces (i.e., a cat to a tiger) and styles of painters (i.e., Hassam to Picasso), respectively, which explains the generality of SuperstarGAN regardless of datasets.

Regulating POLR3G by MicroRNA-26a-5p as a promising therapeutic target of lung cancer stemness and chemosensitivity.

Cancer stem cells (CSCs) identified in lung cancer exhibit resistance to chemotherapy, radiotherapy, and targeted therapy. Therefore, a technology for controlling CSCs is needed to overcome such resistance to cancer therapy. Various evidences about the association between epithelial-mesenchymal transition related transcriptomic alteration and acquisition of CSC phenotype have been proposed recently. Down-regulated miR-26a-5p is closely related to mesenchymal-like lung cancer cell lines. These findings suggest that miR-26a-5p might be involved in lung cancer stemness. RNA polymerase III subunit G (POLR3G) was selected as a candidate target of miR-26a-5p related to cancer stemness. It was found that miR-26a-5p directly regulates the expression of POLR3G.Overexpression of miR-26a-5p induced a marked reduction of colony formation and sphere formation. Co-treatment of miR-26a-5p and paclitaxel decreased cell growth, suggesting that miR-26a-5p might play a role as a chemotherapy sensitizer. In the cancer genome atlas data, high miR-26a-5p and low POLR3G expression were also related to higher survival rate of patients with lung adenocarcinoma. These results suggest that miR-26a-5p can suppress lung cancer stemness and make cancer cell become sensitive to chemotherapy. This finding provides a novel insight into a potential lung cancer treatment by regulating stemness.

Horizontal Arrays of One-Dimensional van der Waals Heterostructures as Transistor Channels.

The nanotube/dielectric interface plays an essential role in achieving superb switching characteristics of carbon nanotube-based transistors for energy-efficient computation. Formation of van der Waals heterostructures with hexagonal boron nitride nanotubes could be an effective means to reduce interface state density, but the need for isolating nanotubes during the formation of coaxial outer layers has hindered the fabrication of their horizontal arrays. Here, we develop a strategy to create isolated heterostructure arrays using aligned carbon nanotubes grown on a quartz substrate as starting materials. Air-suspended arrays of carbon nanotubes are prepared by a dry transfer technique and then used as templates for the coaxial wrapping of boron nitride nanotubes. We then fabricate the transistors, where boron nitride serves as interfacial layers between carbon nanotube channels and conventional gate dielectrics, showing hysteresis-free characteristics owing to the improved interfaces. We have also gained a deeper understanding of the strain applied on inner carbon nanotubes, as well as the inhomogeneity of the outer coating, by characterizing individual heterostructures over trenches and on a substrate surface. The device fabrication and characterization presented here essentially do not require elaborate electron microscopy, thus paving the way for the practical use of one-dimensional van der Waals heterostructures for nanoelectronics.

Heterogeneous Structure Omnidirectional Strain Sensor Arrays With Cognitively Learned Neural Networks.

Mechanically stretchable strain sensors gain tremendous attention for bioinspired skin sensation systems and artificially intelligent tactile sensors. However, high-accuracy detection of both strain intensity and direction with simple device/array structures is still insufficient. To overcome this limitation, an omnidirectional strain perception platform utilizing a stretchable strain sensor array with triangular-sensor-assembly (three sensors tilted by 45°) coupled with machine learning (ML) -based neural network classification algorithm, is proposed. The strain sensor, which is constructed with strain-insensitive electrode regions and strain-sensitive channel region, can minimize the undesirable electrical intrusion from the electrodes by strain, leading to a heterogeneous surface structure for more reliable strain sensing characteristics. The strain sensor exhibits decent sensitivity with gauge factor (GF) of ≈8, a moderate sensing range (≈0-35%), and relatively good reliability (3000 stretching cycles). More importantly, by employing a multiclass-multioutput behavior-learned cognition algorithm, the stretchable sensor array with triangular-sensor-assembly exhibits highly accurate recognition of both direction and intensity of an arbitrary strain by interpretating the correlated signals from the three-unit sensors. The omnidirectional strain perception platform with its neural network algorithm exhibits overall strain intensity and direction accuracy around 98% ± 2% over a strain range of ≈0-30% in various surface stimuli environments.

Low-to-moderate atmospheric ozone levels are negatively correlated with hospital visits by asthma patients.

Asthma is a chronic illness of the airways that affects approximately 300 million individuals worldwide. While it is commonly accepted that high ozone levels exacerbate asthma symptoms, the impact of low to moderate ozone levels on asthma symptoms has received little attention. The purpose of this research was to determine the relationship between hospital visits by asthma patients showing the severity of their symptoms and moderate ozone levels. Statistical analyses were performed on hospital visit big data for asthma patients in Seoul, Korea, collected between 2013 and 2017. The data set includes outpatient hospital visits (n = 17,787,982), hospital admissions (n = 215,696), and emergency department visits (n = 85,482). The frequency of hospital visits by asthma patients was evaluated in relation to low ozone levels (< 0.03 ppm) and moderate ozone levels (0.03-0.06 ppm) in the Seoul environment. In comparison to low ozone levels, moderate ozone levels resulted in a reduction in outpatient hospital visits (t = 7.052, P < .001). When ozone levels were low to moderate, there was a negative correlation between ozone levels and outpatient visits (r = -0.281, 95% CI: -0.331 to -0.228). Negative associations were also identified between ozone levels and new hospital admissions (t = 2.909, P < .01; r = -0.125, 95% CI: -0.179 to -0.070) and emergency treatments (t = 2.679, P < .01; r = -0.132, 95% CI: -0.186 to -0.076). Additionally, it was verified that moderate ozone levels one day before the visits resulted in a reduction in outpatient visits (t = 5.614, P < .001; r = -0.207, 95% CI: -0.259 to -0.153). A strong relationship was identified between moderate atmospheric ozone levels and a reduction in asthma patient hospital visits.

Machine Learning-Based Analyses of the Effects of Various Types of Air Pollutants on Hospital Visits by Asthma Patients.

Asthma is a chronic respiratory disorder defined by airway inflammation, chest pains, wheezing, coughing, and difficulty breathing that affects an estimated 300 million individuals globally. Although various studies have shown an association between air pollution and asthma, few studies have used statistical and machine learning algorithms to investigate the effect of each individual air pollutant on asthma. The purpose of this research was to assess the association between air pollutants and the frequency of hospital visits by asthma patients using three analysis methods: linear correlation analyses were performed by Pearson correlation coefficients, and least absolute shrinkage and selection operator (LASSO) and random forest (RF) models were used for machine learning-based analyses to investigate the effect of air pollutants. This research studied asthma patients using the hospital visit database in Seoul, South Korea, collected between 2013 and 2017. The data set included outpatient hospital visits (n = 17,787,982), hospital admissions (n = 215,696), and emergency department visits (n = 85,482). The daily atmospheric environmental information from 2013 to 2017 at 25 locations in Seoul was evaluated. The three analysis models revealed that NO2 was the most significant pollutant on average in outpatient hospital visits by asthma patients. For example, NO2 had the greatest impact on outpatient hospital visits, resulting in a positive association (r=0.331). In hospital admissions of asthma patients, CO was the most significant pollutant on average. It was observed that CO exhibited the most positive association with hospital admissions (I = 3.329). Additionally, a significant time lag was found between both NO2 and CO and outpatient hospital visits and hospital admissions of asthma patients in the linear correlation analysis. In particular, NO2 and CO were shown to increase hospital admissions at lag 4 in the linear correlation analysis. This study provides evidence that PM2.5, PM10, NO2, CO, SO2, and O3 are associated with the frequency of hospital visits by asthma patients.

Estimating the Prognosis of Low-Grade Glioma with Gene Attention Using Multi-Omics and Multi-Modal Schemes.

The prognosis estimation of low-grade glioma (LGG) patients with deep learning models using gene expression data has been extensively studied in recent years. However, the deep learning models used in these studies do not utilize the latest deep learning techniques, such as residual learning and ensemble learning. To address this limitation, in this study, a deep learning model using multi-omics and multi-modal schemes, namely the Multi-Prognosis Estimation Network (Multi-PEN), is proposed. When using Multi-PEN, gene attention layers are employed for each datatype, including mRNA and miRNA, thereby allowing us to identify prognostic genes. Additionally, recent developments in deep learning, such as residual learning and layer normalization, are utilized. As a result, Multi-PEN demonstrates competitive performance compared to conventional models for prognosis estimation. Furthermore, the most significant prognostic mRNA and miRNA were identified using the attention layers in Multi-PEN. For instance, MYBL1 was identified as the most significant prognostic mRNA. Such a result accords with the findings in existing studies that have demonstrated that MYBL1 regulates cell survival, proliferation, and differentiation. Additionally, hsa-mir-421 was identified as the most significant prognostic miRNA, and it has been extensively reported that hsa-mir-421 is highly associated with various cancers. These results indicate that the estimations of Multi-PEN are valid and reliable and showcase Multi-PEN's capacity to present hypotheses regarding prognostic mRNAs and miRNAs.