1
|
Huang T, Huang Z, Peng X, Pang L, Sun J, Wu J, He J, Fu K, Wu J, Sun X. Construction and validation of risk prediction models for pulmonary embolism in hospitalized patients based on different machine learning methods. Front Cardiovasc Med 2024; 11:1308017. [PMID: 38984357 PMCID: PMC11232034 DOI: 10.3389/fcvm.2024.1308017] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/06/2023] [Accepted: 06/11/2024] [Indexed: 07/11/2024] Open
Abstract
Objective This study aims to apply different machine learning (ML) methods to construct risk prediction models for pulmonary embolism (PE) in hospitalized patients, and to evaluate and compare the predictive efficacy and clinical benefit of each model. Methods We conducted a retrospective study involving 332 participants (172 PE positive cases and 160 PE negative cases) recruited from Guangdong Medical University. Participants were randomly divided into a training group (70%) and a validation group (30%). Baseline data were analyzed using univariate analysis, and potential independent risk factors associated with PE were further identified through univariate and multivariate logistic regression analysis. Six ML models, namely Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Naive Bayes (NB), Support Vector Machine (SVM), and AdaBoost were developed. The predictive efficacy of each model was compared using the receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC). Clinical benefit was assessed using decision curve analysis (DCA). Results Logistic regression analysis identified lower extremity deep venous thrombosis, elevated D-dimer, shortened activated partial prothrombin time, and increased red blood cell distribution width as potential independent risk factors for PE. Among the six ML models, the RF model achieved the highest AUC of 0.778. Additionally, DCA consistently indicated that the RF model offered the greatest clinical benefit. Conclusion This study developed six ML models, with the RF model exhibiting the highest predictive efficacy and clinical benefit in the identification and prediction of PE occurrence in hospitalized patients.
Collapse
Affiliation(s)
- Tao Huang
- Emergency Medicine Center, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Zhihai Huang
- Emergency Medicine Center, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Xiaodong Peng
- Emergency Medicine Center, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Lingpin Pang
- Emergency Medicine Center, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Jie Sun
- Emergency Medicine Center, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Jinbo Wu
- Emergency Medicine Center, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Jinman He
- Emergency Medicine Center, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Kaili Fu
- Respiratory and Critical Care Medicine, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Jun Wu
- Respiratory and Critical Care Medicine, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| | - Xishi Sun
- Emergency Medicine Center, Affiliated Hospital of Guangdong Medical University, Zhanjiang, Guangdong, China
| |
Collapse
|
2
|
Clinical Characteristics of Patients with Pulmonary Thromboembolism Based on Computer Statistical Analysis. JOURNAL OF HEALTHCARE ENGINEERING 2022; 2022:1112960. [PMID: 35242294 PMCID: PMC8888095 DOI: 10.1155/2022/1112960] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 12/11/2021] [Revised: 01/21/2022] [Accepted: 01/27/2022] [Indexed: 02/02/2023]
Abstract
In order to explore the reliable clinical characteristics of patients with pulmonary thromboembolism, this paper applies computer statistical analysis methods to the analysis of clinical characteristics of patients with pulmonary thromboembolism. Moreover, starting from the actual situation, this paper combines experiments to study the case, uses sample screening and sample processing to group samples freely, and conducts grouping reliability research through data statistics methods. After verifying the reliability of the grouping, this paper combines the samples to test the effectiveness of the computer statistical analysis method applied to the clinical characteristic analysis of patients with pulmonary thromboembolism and conducts numerical analysis in combination with the comparative analysis method. The results of the research show that the computer statistical method proposed in this paper has a good effect on the clinical characteristics of patients with pulmonary thromboembolism and meets the actual needs of clinical analysis.
Collapse
|
3
|
Li C, Jin M, Luo Y, Jin Z, Pi L. Integrated bioinformatics analysis of core regulatory elements involved in keloid formation. BMC Med Genomics 2021; 14:239. [PMID: 34600545 PMCID: PMC8487518 DOI: 10.1186/s12920-021-01087-7] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/12/2021] [Accepted: 09/17/2021] [Indexed: 12/18/2022] Open
Abstract
BACKGROUND Keloid is a benign fibro-proliferative dermal tumor formed by an abnormal scarring response to injury and characterized by excessive collagen accumulation and invasive growth. The mechanism of keloid formation has not been fully elucidated, especially during abnormal scarring. Here, we investigated the regulatory genes, micro-RNAs (miRNAs) and transcription factors (TFs) that influence keloid development by comparing keloid and normal scar as well as keloid and normal skin. METHODS Gene expression profiles (GSE7890, GSE92566, GSE44270 and GSE3189) of 5 normal scar samples, 10 normal skin samples and 18 keloid samples from the Gene Expression Omnibus (GEO) database were interrogated. Differentially expressed genes (DEGs) were identified between keloid and normal skin samples as well as keloid and normal scar samples with R Project for Statistical Computing. Gene Ontology (GO) functional enrichment analysis was also performed with R software. DEG-associated protein-protein interaction (PPI) network was constructed by STRING, followed by module selection from the PPI network based on the MCODE analysis. Regulatory relationships between TF/miRNA and target genes were predicted with miRnet and cytoscape. Core regulatory genes were verified by RT-qPCR. RESULTS We identified 628 DEGs, of which 626 were up-regulated and 2 were down-regulated. Seven core genes [neuropeptide Y(NPY), 5-hydroxytryptamine receptor 1A(HTR1A), somatostatin (SST), adenylate cyclase 8 (ADCY8), neuromedin U receptor 1 (NMUR1), G protein subunit gamma 3 (GNG3), and G protein subunit gamma 13 (GNG13)] all belong to MCODE1 and were enriched in the "G protein coupled receptor signaling pathway" of the GO biological process category. Furthermore, nine core miRNAs (hsa-mir-124, hsa-let-7, hsa-mir-155, hsa-mir-26a, hsa-mir-941, hsa-mir-10b, hsa-mir-20, hsa-mir-31 and hsa-mir-372), and two core TFs (SP1 and TERT) were identified to play important roles in keloid formation. In the TF/miRNA-target gene network, both hsa-mir-372 and hsa-mir-20 had a regulatory effect on GNG13, ADCY8 was predicted to be target by hsa-mir-10b, and HTR1A and NPY were potentially by SP1. Furthermore, the expression of core regulatory genes (GNG13, ADCY8, HTR1A and NPY) was validated in clinical samples. CONCLUSIONS GNG13, ADCY8, NPY and HTR1A may act as core genes in keloid formation and these core genes establish relationship with SP1 and miRNA (hsa-mir-372, hsa-mir-20, hsa-mir-10b), which may influence multiple signaling pathways in the pathogenesis of keloid.
Collapse
Affiliation(s)
- Chuying Li
- Klebs Research Center, Department of Dermatology, Yanbian University Hospital, Yanji, 133000, China
| | - Meitong Jin
- Klebs Research Center, Department of Dermatology, Yanbian University Hospital, Yanji, 133000, China
| | - Yinli Luo
- Klebs Research Center, Department of Dermatology, Yanbian University Hospital, Yanji, 133000, China
| | - Zhehu Jin
- Klebs Research Center, Department of Dermatology, Yanbian University Hospital, Yanji, 133000, China.
| | - Longquan Pi
- Klebs Research Center, Department of Dermatology, Yanbian University Hospital, Yanji, 133000, China.
| |
Collapse
|
4
|
Xia Q, Ling X, Wang Z, Shen T, Chen M, Mao D, Ma X, Ning J, Zhang H, Chen D, Gu Q, Shen H, Yan J. Lateral rectus muscle differentiation potential in paralytic esotropia patients. BMC Ophthalmol 2021; 21:235. [PMID: 34044792 PMCID: PMC8161593 DOI: 10.1186/s12886-021-01994-4] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/01/2021] [Accepted: 05/10/2021] [Indexed: 12/05/2022] Open
Abstract
Purpose and background Recently, we found that maximal medial rectus recession and lateral rectus resection in patients with complete lateral rectus paralysis resulted in a partial restoration of abduction. In an attempt to understand some of the mechanisms involved with this effect we examined gene expression profiles of lateral recti from these patients, with our focus being directed to genes related to myogenesis. Materials and methods Lateral recti resected from patients with complete lateral rectus paralysis and those from concomitant esotropia (controls) were collected. Differences in gene expression profiles between these two groups were examined using microarray analysis and quantitative Reverse-transcription PCR (qRT-PCR). Results A total of 3056 differentially expressed genes (DEGs) were identified between these two groups. Within the paralytic esotropia group, 2081 genes were up-regulated and 975 down-regulated. The results of RT-PCR revealed that PAX7, MYOG, PITX1, SIX1 and SIX4 showed higher levels of expression, while that of MYOD a lower level of expression within the paralytic esotropia group as compared with that in the control group (p < 0.05). Conclusion The decreased expression of MYOD in the paralytic esotropia group suggested that extraocular muscle satellite cell (EOMSCs) differentiation processes were inhibited. Whereas the high expression levels of PAX7, SIX1/4 and MYOG, suggested that the EOMSCs were showing an effective potential for differentiation. The stimulation resulting from muscle surgery may induce EOMSCs to differentiate and thus restore abduction function. Supplementary Information The online version contains supplementary material available at 10.1186/s12886-021-01994-4.
Collapse
Affiliation(s)
- Qing Xia
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Xiangtian Ling
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Zhonghao Wang
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Tao Shen
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Minghao Chen
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Danyi Mao
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Xinqi Ma
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Jie Ning
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Han Zhang
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Dongli Chen
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China
| | - Qiong Gu
- Research Center for Drug Discovery, School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, 510006, People's Republic of China
| | - Huangxuan Shen
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China. .,Biobank of Eye, State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China.
| | - Jianhua Yan
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou, 510060, China.
| |
Collapse
|