Graph Convolutional Networks For Disease Mapping and Classification in Healthcare DOI
Rakesh Kumar, Devvret Verma,

J. Relin Francis Raj

и другие.

Опубликована: Дек. 29, 2023

In the context of healthcare, this study investigates use Graph A convolutional Networks (GCNs) for disease mapping along with classification. Based on an interpretivist philosophical thought, a descriptive design alongside secondary data collection is used in deductive manner. The research creates strong framework sickness mapping, assesses how well GCNs adapt to varied health information, and compares their effectiveness more conventional machine learning techniques order determine suitable they are. An investigation conducted into understanding GCN-based diagnosis models, offering valuable perspectives decision-making procedures. findings support improved diagnostic precision, wellinformed treatment planning, precision medical treatments. emphasis when applying results procedures connection systems that provide decision support, ongoing improvement. importance model interpretability, ability be general as realworld integration highlighted by critical analysis. Developing interpretability strategies addressing ethical issues are among recommendations. ensure responsible deployment, future work ought concentrate improving GCN architectures, integrating multi-modal information advocating interdisciplinary collaboration.

Язык: Английский

Exploring Molecular Mechanisms and Biomarkers in COPD: An Overview of Current Advancements and Perspectives DOI Open Access
Chin-Ling Li,

Shih-Feng Liu

International Journal of Molecular Sciences, Год журнала: 2024, Номер 25(13), С. 7347 - 7347

Опубликована: Июль 4, 2024

Chronic obstructive pulmonary disease (COPD) plays a significant role in global morbidity and mortality rates, typified by progressive airflow restriction lingering respiratory symptoms. Recent explorations molecular biology have illuminated the complex mechanisms underpinning COPD pathogenesis, providing critical insights into progression, exacerbations, potential therapeutic interventions. This review delivers thorough examination of latest progress research related to COPD, involving fundamental pathways, biomarkers, targets, cutting-edge technologies. Key areas focus include roles inflammation, oxidative stress, protease-antiprotease imbalances, alongside genetic epigenetic factors contributing susceptibility heterogeneity. Additionally, advancements omics technologies-such as genomics, transcriptomics, proteomics, metabolomics-offer new avenues for comprehensive profiling, aiding discovery novel biomarkers targets. Comprehending foundation carries substantial creation tailored treatment strategies enhancement patient outcomes. By integrating clinical practice, there is promising pathway towards personalized medicine approaches that can improve diagnosis, treatment, overall management ultimately reducing its burden.

Язык: Английский

Процитировано

9
Deep learning for detecting and early predicting chronic obstructive pulmonary disease from spirogram time series DOI Creative Commons
Shuhao Mei, Xin Li, Yuxi Zhou

и другие.

npj Systems Biology and Applications, Год журнала: 2025, Номер 11(1)

Опубликована: Фев. 15, 2025

Chronic Obstructive Pulmonary Disease (COPD) is a chronic lung condition characterized by airflow obstruction. Current diagnostic methods primarily rely on identifying prominent features in spirometry (Volume-Flow time series) to detect COPD, but they are not adept at predicting future COPD risk based subtle data patterns. In this study, we introduce novel deep learning-based approach, DeepSpiro, aimed the early prediction of risk. DeepSpiro consists four key components: SpiroSmoother for stabilizing Volume-Flow curve, SpiroEncoder capturing volume variability-pattern through patches varying lengths, SpiroExplainer integrating heterogeneous and explaining predictions attention, SpiroPredictor disease undiagnosed high-risk patients patch concavity, with horizons 1–5 years, or even longer. Evaluated UK Biobank dataset, achieved an AUC 0.8328 detection demonstrated strong predictive performance (p-value < 0.001). summary, can effectively predict long-term progression disease.

Язык: Английский

Процитировано

1
Applications of digital health technologies and artificial intelligence algorithms in COPD: systematic review DOI Creative Commons

Zhenli Chen,

Jie Hao, Haixia Sun

и другие.

BMC Medical Informatics and Decision Making, Год журнала: 2025, Номер 25(1)

Опубликована: Фев. 13, 2025

Язык: Английский

Процитировано

0
Strategies to include prior knowledge in omics analysis with deep neural networks DOI Creative Commons
Kisan Thapa, Meric Kinali, Shichao Pei

и другие.

Patterns, Год журнала: 2025, Номер 6(3), С. 101203 - 101203

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
Prognostic Biomarkers Based on Proteomic Technology in COPD: A Recent Review DOI Creative Commons
F Hanyu, Ying Liu, Qiwen Yang

и другие.

International Journal of COPD, Год журнала: 2023, Номер Volume 18, С. 1353 - 1365

Опубликована: Июнь 1, 2023

Chronic obstructive pulmonary disease (COPD) is a common heterogeneous respiratory which characterized by persistent and incompletely reversible airflow limitation. Due to the heterogeneity phenotypic complexity of COPD, traditional diagnostic methods provide limited information pose great challenge clinical management. In recent years, with development omics technologies, proteomics, metabolomics, transcriptomics, etc., have been widely used in study providing help discover new biomarkers elucidate complex mechanisms COPD. this review, we summarize prognostic COPD based on proteomic studies years evaluate their association prognosis. Finally, present prospects challenges prognostic-related studies. This review expected cutting-edge evidence evaluation patients inform future

Язык: Английский

Процитировано

5
Comprehensive time-course gene expression evaluation of high-risk beef cattle to establish immunological characteristics associated with undifferentiated bovine respiratory disease DOI Creative Commons
Matthew A. Scott, Robert Valeris-Chacín, Alexis C. Thompson

и другие.

Frontiers in Immunology, Год журнала: 2024, Номер 15

Опубликована: Сен. 13, 2024

Bovine respiratory disease (BRD) remains the leading infectious in beef cattle production systems. Host gene expression upon facility arrival may indicate risk of BRD development and severity. However, a time-course approach would better define how influences immunological inflammatory responses after occurrences. Here, we evaluated whole blood transcriptomes high-risk at three time points to elucidate BRD-associated host response. Sequenced jugular mRNA from 36 (2015:

Язык: Английский

Процитировано

0
Advances on the role of lung macrophages in the pathogenesis of chronic obstructive pulmonary disease in the Era of Single-Cell Genomics DOI
Xiaohua Li, Hui Zhang, Xuebin Chi

и другие.

International Journal of Medical Sciences, Год журнала: 2024, Номер 22(2), С. 298 - 308

Опубликована: Дек. 11, 2024

Chronic Obstructive Pulmonary Disease (COPD) is a heterogeneous respiratory disorder characterized by persistent airflow limitation. The diverse pathogenic mechanisms underlying COPD progression remain incompletely understood. Macrophages, serving as the most representative immune cells in tract, constitute first line of innate defense and maintain pulmonary immunological homeostasis. Recent advances have provided deeper insights into phenotypic functional alterations macrophages their role pathogenesis. Notably, advent single-cell RNA sequencing has revolutionized our understanding macrophage molecular heterogeneity COPD. Herein, we review principal investigations concerning sophisticated through which influence COPD, encompassing inflammatory mediator production, protease/antiprotease release, phagocytic activity. Additionally, synthesize findings from available literature regarding all identified sub-populations thereby advancing comprehension heterogeneity's significance complex pathophysiological

Язык: Английский

Процитировано

0
The influence of service performance in China's sci-tech commissioner system: Using social network analysis and interpretable machine learning DOI Creative Commons
Jinghao Chen,

Wensi Li,

Q. Liu

и другие.

Heliyon, Год журнала: 2024, Номер 10(12), С. e32968 - e32968

Опубликована: Июнь 1, 2024

The Sci-Tech Commissioner System (SCS) is a result of exploratory efforts by the Chinese government to use science and technology strengthen agricultural sector. Social network analysis (SNA) machine learning (ML) techniques make it feasible assess service performance in China's SCS using indicators such as group types structure features. In this study, SNA clustering algorithm were employed categorize sci-tech commissioners. By comparing accuracy different classification algorithms predicting results, LightGBM was finally select determine features commissioners establish an interpretable ML model. Then, SHAP used analyze influences affecting performance. Results show that forms are group-oriented, include small groups young with close cooperation, larger middle-aged commissioners, old isolated points highly-influential Furthermore, while size not determinant commissioner's average performance, coordination ability found be more critical. Moreover, differences distinct caused various factors, but good structures extensive social contacts essential for high

Язык: Английский

Процитировано

0
Machine Learning-based Approaches to Integrate Heterogeneous Data for Biological Knowledge Transfer DOI Creative Commons
Youngjun Park

Опубликована: Янв. 1, 2024

Recent developments in high-throughput data generation methodologies, such as next-generation sequencing or MALDI-TOF mass spectrometry, are creating a strong necessity for science to transform the field of biomedical research. Over past decade, these technologies have facilitated accumulation extensive omics data. Although this advancement has greatly contributed knowledge expansion research, studies still limited due heterogeneity: batch effects, heterogeneity types, and biological different species. These challenges complicate applicability statistical methods machine learning models complex analysis scenarios with various datasets. Consequently, there is growing demand methodologies handle heterogeneous In thesis, I investigated aforementioned three challenges, namely type heterogeneity, developed novel address them. The first challenge systematic non-biological variation added datasets during acquisition. Batch effects one factors hindering integrative same types spectrometry single-cell RNA were investigated. Different hospitals generated large scale from patient samples. Due procedures protocols, exist on levels each dataset impede analysis. examined using models, logistic regression, lightGBM, neural network. With recent advancements sequencing, it become widely employed diverse produce large-scale cell populations within tissues. However, presence necessitates appropriate pre-processing when integrating multiple studies. Initially, impact single Following that, simple approach low-dimensional embedding transformation effect mitigation was proposed. Data transformations significant subsequent downstream by altering distribution. normalization step often neglected. Therefore, distinct evaluated regarding their dimensionality reduction clustering. This result shows that proportion can be mitigated transformation, showed comparable results already published deep network models. next integration research presents challenge, given differences formats underlying hypotheses. To multi-modal methodology capable effectively handling required. my meta-transfer based few-shot model proposed integrate bulk- highly effective small sample sizes. It able mitigate predict suggests new utilize amount bulk-cell available public databases. By leveraging existing data, researchers overcome study size constraints last related originates variety species unique genome. poses challenging task may require transfer learning. Transfer classified into homogeneous categories features' characteristics source target approaches two datasets, potential value cross-species antimicrobial resistance prediction. clinical practices higher species, training aggregated proved beneficial predicting unknown Furthermore, introduced data-driven way. conventional relies gene homology. dependence severely limits wide applications non-model organisms. Thus, designed independent homology exploiting shared labels, experimental conditions, among species-agnostic successfully integrates thesis thoroughly explores posed corresponding challenges. offers comprehensive perspective issue field. conducting additional analyses leverage enhance robustness generalizability, thus contributing addressing reproducibility crisis.

Язык: Английский

Процитировано

0
Exploring the Unknown: The Application and Prospects of Artificial Intelligence in Genomics and Bioinformatics DOI Open Access

Qigang Feng,

Jie Li, Qing Zhang

и другие.

Health, Год журнала: 2024, Номер 16(09), С. 837 - 848

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0