Discovery and validation of molecular patterns and immune characteristics in the peripheral blood of ischemic stroke patients

Background

Stroke is a disease with high morbidity, disability, and mortality. Immune factors play a crucial role in the occurrence of ischemic stroke (IS), but their exact mechanism is not clear. This study aims to identify possible immunological mechanisms by recognizing immune-related biomarkers and evaluating the infiltration pattern of immune cells.

Methods

We downloaded datasets of IS patients from GEO, applied R language to discover differentially expressed genes, and elucidated their biological functions using GO, KEGG analysis, and GSEA analysis. The hub genes were then obtained using two machine learning algorithms (least absolute shrinkage and selection operator (LASSO) and support vector machine-recursive feature elimination (SVM-RFE)) and the immune cell infiltration pattern was revealed by CIBERSORT. Gene-drug target networks and mRNA-miRNA-lncRNA regulatory networks were constructed using Cytoscape. Finally, we used RT-qPCR to validate the hub genes and applied logistic regression methods to build diagnostic models validated with ROC curves.

Results

We screened 188 differentially expressed genes whose functional analysis was enriched to multiple immune-related pathways. Six hub genes (ANTXR2, BAZ2B, C5AR1, PDK4, PPIH, and STK3) were identified using LASSO and SVM-RFE. ANTXR2, BAZ2B, C5AR1, PDK4, and STK3 were positively correlated with neutrophils and gamma delta T cells, and negatively correlated with T follicular helper cells and CD8, while PPIH showed the exact opposite trend. Immune infiltration indicated increased activity of monocytes, macrophages M0, neutrophils, and mast cells, and decreased infiltration of T follicular helper cells and CD8 in the IS group. The ceRNA network consisted of 306 miRNA-mRNA interacting pairs and 285 miRNA-lncRNA interacting pairs. RT-qPCR results indicated that the expression levels of BAZ2B, C5AR1, PDK4, and STK3 were significantly increased in patients with IS. Finally, we developed a diagnostic model based on these four genes. The AUC value of the model was verified to be 0.999 in the training set and 0.940 in the validation set.

Conclusion

Our research explored the immune-related gene expression modules and provided a specific basis for further study of immunomodulatory therapy of IS.

Acute exacerbation of idiopathic pulmonary fibrosis disease: a diagnosis model in China

OBJECTIVE

To develop and validate a diagnosis model to inform risk stratified decisions for idiopathic pulmonary fibrosis patients experiencing acute exacerbations (AE-IPF).

METHODS

In this retrospective cohort study performed from 1 January 2016 to 31 December 2022, we used data from the West China Hospital of Sichuan University for model development and validation. Blood test results and the underlying diseases of patients were collected through the HIS system and LIS system. An algorithm for filtering candidate variables based on least absolute shrinkage and selection operator (LASSO) regression. Logistic regression was performed to develop the risk model. Multiple imputation handled missing predictor data. Model performance was assessed through calibration and diagnostic odds ratio.

RESULTS

311 and 133 participants were included in the development and validation cohorts, respectively. 3 candidate predictors (29 parameters) were included. A logistic regression analysis revealed that dyspnea, percentage of CD4+  T-lymphocytes, and percentage of monocytes are independent risk factors for AE-IPF. Nomographic model was constructed using these independent risk factors, and the C-index was 0.69. For internal validation, the C-index was 0.69, and that indicated good accuracy. Diagnostic odds ratio was 5.40. Meanwhile, in mild, moderate, and severe subgroups, AE positivity rates were 0.37, 0.47, and 0.81, respectively. The diagnostic model can classify patients with AE-IPF into different risk classes based on dyspnea, percentage of CD4+  T-lymphocytes, and percentage of monocytes.

CONCLUSION

A diagnosis model was developed and validated that used information collected from HIS system and LIS system and may be used to risk stratify idiopathic pulmonary fibrosis patients experiencing acute exacerbations.

New approach for early and specific Alzheimer disease diagnosis from different plasma biomarkers

BACKGROUND

Alzheimer Disease (AD) is a complex pathology, in which several biochemical pathways could be involved. Therefore, the development of clinical studies combining different nature biomarkers in an AD diagnosis approach is required. Specifically, the present study evaluated blood biomarkers from different molecular pathways (epigenomics, lipid metabolism, lipid peroxidation), to obtain an early and specific AD diagnosis approach.

METHODS

The participants were classified into early AD (n = 53), and non-AD (healthy controls, other dementias) (n = 83). Blood samples were collected and biochemical determinations (microRNAs, lipids, lipid peroxidation compounds) were carried out by quantitative PCR and liquid chromatography coupled to mass spectrometry, respectively. Then, a logistic regression model with a Bayesian variable selection procedure was developed.

RESULTS

The Bayesian variable selection procedure for microRNAs did not show any relevant variable. Therefore, microRNA biomarkers were excluded. So, the developed model considered only lipids and lipid peroxidation compounds. The corresponding selected variables were age, 18:0 LPC, PGE2, isoprostanes and, isofurans. The validated model (by leave-one-out cross-validation) provided satisfactory diagnosis indexes (AUC 0.83, Sensitivity 87 %, Specificity 79 %).

CONCLUSION

The developed model included biomarkers from different pathways (lipid metabolism, oxidative stress), achieving a promising approach to early, specific and, minimally invasive AD diagnosis. Nevertheless, further work to validate clinically these preliminary results with an external cohort is required. Also, the integration of different compounds coming from several biochemical pathways could constitute a relevant research field for the development of AD therapeutic targets.

Functional and structural MRI based obsessive-compulsive disorder diagnosis using machine learning methods

BACKGROUND

The success of neuroimaging in revealing neural correlates of obsessive-compulsive disorder (OCD) has raised hopes of using magnetic resonance imaging (MRI) indices to discriminate patients with OCD and the healthy. The aim of this study was to explore MRI based OCD diagnosis using machine learning methods.

METHODS

Fifty patients with OCD and fifty healthy subjects were allocated into training and testing set by eight to two. Functional MRI (fMRI) indices, including amplitude of low-frequency fluctuation (ALFF), fractional ALFF (fALFF), regional homogeneity (ReHo), degree of centrality (DC), and structural MRI (sMRI) indices, including volume of gray matter, cortical thickness and sulcal depth, were extracted in each brain region as features. The features were reduced using least absolute shrinkage and selection operator regression on training set. Diagnosis models based on single MRI index / combined MRI indices were established on training set using support vector machine (SVM), logistic regression and random forest, and validated on testing set.

RESULTS

SVM model based on combined fMRI indices, including ALFF, fALFF, ReHo and DC, achieved the optimal performance, with a cross-validation accuracy of 94%; on testing set, the area under the receiver operating characteristic curve was 0.90 and the validation accuracy was 85%. The selected features were located both within and outside the cortico-striato-thalamo-cortical (CSTC) circuit of OCD. Models based on single MRI index / combined fMRI and sMRI indices underperformed on the classification, with a largest validation accuracy of 75% from SVM model of ALFF on testing set.

CONCLUSION

SVM model of combined fMRI indices has the greatest potential to discriminate patients with OCD and the healthy, suggesting a complementary effect of fMRI indices on the classification; the features were located within and outside the CSTC circuit, indicating an importance of including various brain regions in the model.

An end-end arrhythmia diagnosis model based on deep learning neural network with multi-scale feature extraction

This study presents an innovative end-to-end deep learning arrhythmia diagnosis model that aims to address the problems in arrhythmia diagnosis. The model performs pre-processing of the heartbeat signal by automatically and efficiently extracting time-domain, time-frequency-domain and multi-scale features at different scales. These features are imported into an adaptive online convolutional network-based classification inference module for arrhythmia diagnosis. Experimental results show that the AOCT-based deep learning neural network diagnostic module has excellent parallel computing and classification inference capabilities, and the overall performance of the model improves with increasing scales. In particular, when multi-scale features are used as inputs, the model is able to learn both time-frequency domain information and other rich information, thus significantly improving the performance of the end-to-end diagnostic model. The final results show that the AOCT-based deep learning neural network model has an average accuracy of 99.72%, a recall of 99.62%, and an F1 score of 99.3% in diagnosing four common heart diseases.

Comprehensive analysis of circulating cell-free RNAs in blood for diagnosing non-small cell lung cancer

Early screening and detection of non-small cell lung cancer (NSCLC) is crucial due to the significantly low survival rate in advanced stages. Blood-based liquid biopsy is non-invasive test to assistant disease diagnosis, while cell-free RNA is one of the promising biomarkers in blood. However, the disease related signatures have not been explored completely for most cell-free RNA transcriptome sequencing (cfRNA-Seq) datasets. To address this gap, we developed a comprehensive cfRNA-Seq pipeline for data analysis and constructed a machine learning model to facilitate noninvasive early diagnosis of NSCLC. The results of our study have demonstrated the identification of differential mRNA, lncRNAs and miRNAs from cfRNA-Seq, which have exhibited significant association with development and progression of lung cancer. The classifier based on gene expression signatures achieved an impressive area under the curve (AUC) of up to 0.9, indicating high specificity and sensitivity in both cross-validation and independent test. Furthermore, the analysis of T cell and B cell immune repertoire extracted from cfRNA-Seq have provided insights into the immune status of cancer patients, while the microbiome analysis has revealed distinct bacterial and viral profiles between NSCLC and normal samples. In our future work, we aim to validate the existence of cancer associated T cell receptors (TCR)/B cell receptors (BCR) and microorganisms, and subsequently integrate all identified signatures into diagnostic model to improve the prediction accuracy. This study not only provided a comprehensive analysis pipeline for cfRNA-Seq dataset but also highlights the potential of cfRNAs as promising biomarkers and models for early NSCLC diagnosis, emphasizing their importance in clinical settings.

A Diagnostic Model for Breast Lesions With Enlarged Enhancement Extent on Contrast-Enhanced Ultrasound Improves Malignancy Prediction

OBJECTIVE

The aim of the work described here was to develop a diagnostic model based on contrast-enhanced ultrasound (CEUS) features to improve performance in predicting the probability of malignancy for breast lesions with an enlarged enhancement extent on CEUS.

METHODS

In total, 299 consecutive patients who underwent CEUS examination and had confirmed pathological results were retrospectively enrolled. Among the 299 patients, an enlarged enhancement extent on CEUS was found in 142 patients. In this special cohort, we analyzed the association of malignant pathologic results with perfusion patterns emphatically by reclassifying the patterns.

RESULTS

A diagnostic model was developed and presented as a nomogram, assessed with discrimination and calibration. Receiver operating characteristic (ROC) curve analysis revealed that the areas under the curves of the conventional perfusion and modified perfusion patterns were 0.58 and 0.76 (p < 0.001), respectively. A diagnostic model was built and exhibited good discrimination with a C-index of 0.95 (95% confidence interval: 0.91-0.98), which was confirmed to be 0.93 via internal bootstrapping validation.

CONCLUSION

The nomogram based on CEUS features provides radiologists with a quantitative tool to predict the probability of malignancy in this special cohort of breast lesions.

Identification and verification of feature biomarkers associated with immune cells in neonatal sepsis

BACKGROUND

Neonatal sepsis (NS), a life-threatening condition, is characterized by organ dysfunction and is the most common cause of neonatal death. However, the pathogenesis of NS is unclear and the clinical inflammatory markers currently used are not ideal for diagnosis of NS. Thus, exploring the link between immune responses in NS pathogenesis, elucidating the molecular mechanisms involved, and identifying potential therapeutic targets is of great significance in clinical practice. Herein, our study aimed to explore immune-related genes in NS and identify potential diagnostic biomarkers. Datasets for patients with NS and healthy controls were downloaded from the GEO database; GSE69686 and GSE25504 were used as the analysis and validation datasets, respectively. Differentially expressed genes (DEGs) were identified and Gene Set Enrichment Analysis (GSEA) was performed to determine their biological functions. Composition of immune cells was determined and immune-related genes (IRGs) between the two clusters were identified and their metabolic pathways were determined. Key genes with correlation coefficient > 0.5 and p < 0.05 were selected as screening biomarkers. Logistic regression models were constructed based on the selected biomarkers, and the diagnostic models were validated.

RESULTS

Fifty-two DEGs were identified, and GSEA indicated involvement in acute inflammatory response, bacterial detection, and regulation of macrophage activation. Most infiltrating immune cells, including activated CD8 + T cells, were significantly different in patients with NS compared to the healthy controls. Fifty-four IRGs were identified, and GSEA indicated involvement in immune response and macrophage activation and regulation of T cell activation. Diagnostic models of DEGs containing five genes (PROS1, TDRD9, RETN, LOC728401, and METTL7B) and IRG with one gene (NSUN7) constructed using LASSO algorithm were validated using the GPL6947 and GPL13667 subset datasets, respectively. The IRG model outperformed the DEG model. Additionally, statistical analysis suggested that risk scores may be related to gestational age and birth weight, regardless of sex.

CONCLUSIONS

We identified six IRGs as potential diagnostic biomarkers for NS and developed diagnostic models for NS. Our findings provide a new perspective for future research on NS pathogenesis.

Single-cell and microarray chip analysis revealed the underlying pathogenesis of ulcerative colitis and validated model genes in diagnosis and drug response

The morbidity rate of ulcerative colitis (UC) in the world is increasing year by year, recurrent episodes of diarrhea, mucopurulent and bloody stools, and abdominal pain are the main symptoms, reducing the quality of life of the patient and affecting the productivity of the society. In this study, we sought to develop robust diagnostic biomarkers for UC, to uncover potential targets for anti-TNF-ɑ drugs, and to investigate their associated pathway mechanisms. We collected single-cell expression profile data from 9 UC or healthy samples and performed cell annotation and cell communication analysis. Revealing the possible pathogenesis of ulcerative colitis by Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA) analysis. Based on the disease-related modules obtained from weighted correlation network analysis (WGCNA) analysis, we used Lasso regression analysis and random forest algorithm to identify the genes with the greatest impact on disease (EPB41L3, HSD17B3, NDRG1, PDIA5, TRPV3) and further validated the diagnostic value of the model genes by various means. To further explore the relationship and mechanism between model genes and drug sensitivity, we collected gene expression profiles of 185 UC patients before receiving anti-tumor necrosis factor drugs, and we performed functional analysis based on the results of differential analysis between NR tissues and R tissues, and used single-sample GSEA (ssGSEA) and CIBERSORT algorithms to explore the important role of immune microenvironment on drug sensitivity. The results suggest that our model is not only helpful in aiding diagnosis, but also has implications for predicting drug efficacy; in addition, model genes may influence drug sensitivity by affecting immune cells. We suggest that this study has developed a diagnostic model with higher specificity and sensitivity, and also provides suggestions for clinical administration and drug efficacy prediction.

A novel promising diagnosis model for colorectal advanced adenoma and carcinoma based on the progressive gut microbiota gene biomarkers

BACKGROUND

Colorectal cancer (CRC), a commonly diagnosed cancer often develops slowly from benign polyps called adenoma to carcinoma. Altered gut microbiota is implicated in colorectal carcinogenesis. It is warranted to find non-invasive progressive microbiota biomarkers that can reflect the dynamic changes of the disease. This study aimed to identify and evaluate potential progressive fecal microbiota gene markers for diagnosing advanced adenoma (AA) and CRC.

RESULTS

Metagenome-wide association was performed on fecal samples from different cohorts of 871 subjects (247 CRC, 234 AA, and 390 controls). We characterized the gut microbiome, identified microbiota markers, and further constructed a colorectal neoplasms classifier in 99 CRC, 94 AA, and 62 controls, and validated the results in 185 CRC, 140 AA, and 291 controls from 3 independent cohorts. 21 species and 277 gene markers were identified whose abundance was significantly increased or decreased from normal to AA and CRC. The progressive gene markers were distributed in metabolic pathways including amino acid and sulfur metabolism. A diagnosis model consisting of four effect indexes was constructed based on the markers, the sensitivities of the Adenoma Effect Index 1 for AA, Adenoma Effect Index 2 for high-grade dysplasia (HGD) adenoma were 71.3% and 76.5%, the specificities were 90.5% and 90.3%, respectively. CRC Effect Index 1 for all stages of CRC and CRC Effect Index 2 for stage III-IV CRC to predict CRC yielded an area under the curve (AUC) of 0.839 (95% CI 0.804-0.873) and 0.857 (95% CI 0.793-0.921), respectively. Combining with fecal immunochemical test (FIT) significantly improved the sensitivity of CRC Effect Index 1 and CRC Effect Index 2 to 96.7% and 100%.

CONCLUSIONS

This study reports the successful diagnosis model establishment and cross-region validation for colorectal advanced adenoma and carcinoma based on the progressive gut microbiota gene markers. The results suggested that the novel diagnosis model can significantly improve the diagnostic performance for advanced adenoma.

[Application of TG/HDL-C Combined with Liver Function Indexes to Predict Metabolic-Associated Fatty Liver Disease]

Objective

To study the application of triglycerides to high-density lipoprotein cholesterol ratio (TG/HDL-C) combined with liver function indexes to predict metabolic-associated fatty liver disease (MAFLD).

Methods

A total of 2971 outpatients diagnosed with MAFLD and 2794 healthy controls were enrolled, and their relevant data were collected. Two-sample Mann-Whitney U test and binary logistic regression analysis were conducted to study the relationship between TG/HDL-C and MAFLD and to construct combined diagnosis models of MAFLD. The area under the curve (AUC) of receiver operating characteristic (ROC) was used to pick out the optimal model.

Results

The TG/HDL-C of MAFLD patients was significantly higher than that of healthy controls. In multivariate analysis, after adjusting for body mass index, systolic blood pressure, diastolic blood pressure, fasting blood glucose, triglycerides, high-density lipoprotein cholesterol, uric acid and creatinine, the odds ratio of TG/HDL-C was 2.356 (95% confidence interval [CI]: 1.028-5.400). Therefore, TG/HDL-C was an independent risk factor for MAFLD. ROC curve analysis showed that the AUC of using TG/HDL-C to predict MAFLD was 0.795 (95% CI: 0.784-0.807), and when the cut-off value was 1.09, the sensitivity was 0.679 and the specificity was 0.755. The AUC of the diagnosis model established by a combined use of TG/HDL-C, alanine aminotransferase (ALT), aspartate aminotransferase (AST), and albumin (ALB) was 0.890 (95% CI: 0.882-0.898), and when the cut-off value was 0.47, the sensitivity and specificity were 0.792 and 0.839, respectively.

Conclusion

TG/HDL-C is an independent risk factor for MAFLD. TG/HDL-C can well predict MAFLD when it is used in combination with ALT, AST, and ALB.

Early differential diagnosis model for acute radiation pneumonitis based on multiple parameters

Objective: The present study aimed to construct a diagnosis model for the early differentiation of acute radiation pneumonitis (ARP) and infectious pneumonitis based on multiple parameters.

Methods: The present study included data of 152 patients admitted to the Department of Radiochemotherapy, Tangshan People’s Hospital, who developed ARP (91 patients) or infectious pneumonia (IP; 61 patients) after radiotherapy. The radiophysical parameters, imaging characteristics, serological indicators, and other data were collected as independent variables, and ARP was considered as a dependent variable. Logistics univariate analysis and Spearman correlation analysis were used for selecting independent variables. Logistics multivariate analysis was used to fit the variables into the regression model to predict ARP.

Results: The univariate analysis showed that the positional relation between lesions and V20 area (PRLV), procalcitonin (PCT), C-reactive protein (CRP), mean lung dose (MLD), and lung volume receiving ≥20 Gy (V20) correlated with ARP while the planning target volume (PTV) dose marginally correlated with ARP. The multivariate analysis showed that the PRLV, PCT, white blood cell (WBC), and MLD were independent diagnostic factors. The nomogram was drawn on the basis of the logistics regression model. The area under the curve (AUC) of the model was 0.849, which was significantly better than that of a single indicator and the sensitivity and specificity of the model were high (82.4 and 82.0%, respectively). These results predicted by the model were highly consistent with the actual diagnostic results. The decision curve analysis (DCA) demonstrated a satisfactory positive net benefit of the model.

Conclusion: The diagnosis model constructed in the present study is of certain value for the differential diagnosis of ARP and IP.

Modeling of diagnosis for metabolic syndrome by integrating symptoms into physiochemical indexes

BACKGROUND

Metabolic syndrome (MS) is a major global health concern comprising a cluster of co-occurring conditions that increase the risk of heart disease, stroke and type 2 diabetes. MS is usually diagnosed using a combination of physiochemical indexes (such as BMI, abdominal circumference and blood pressure) but largely ignores clinical symptoms when investigating prevention and treatment of the disease. Exploring predictors of MS using multiple diagnostic indicators may improve early diagnosis and treatment of MS. Traditional Chinese medicine (TCM) attaches importance to the etiology of disease symptoms and indications using four diagnostic methods, which have long been used to treat metabolic disease. Therefore, in this study, we aimed to develop predictive indicators for MS using both physiochemical indexes and TCM methods.

METHODS

Clinical information (including both physiochemical and TCM indexes) was obtained from a cohort of 586 individuals across 4 hospitals in China, comprising 136 healthy controls and 450 MS cases. Using this cohort, we compared three classic machine learning methods: decision tree (DT), support vector machine (SVM) and random forest (RF) towards MS diagnosis using physiochemical and TCM indexes, with the best model selected by comparing the accuracy, specificity and sensitivity of the three models. In parallel, the best proportional partition of the training data to the test data was confirmed by observing the changes in evaluation indexes using each model. Next, three subsets containing different categories of variables (including both TCM and physicochemical indexes combined - termed the "fused indexes", only physicochemical indexes, and TCM indexes only) were compared and analyzed using the best performing model and optimum training to test data proportion. Next, the best subset was selected through comprehensive comparative analysis, and then the important prediction variables were selected according to their weight.

RESULTS

When comparing the three models, we found that the RF model had the highest average accuracy (average 0.942, 95%CI [0.925, 0.958]) and sensitivity (average 0.993, 95%CI [0.990, 0.996]). Besides, when the training set accounted for 80% of the cohort data, the specificity got the best value and the accuracy and sensitivity were also very high in RF model. In view of the performance of the three different subsets, the prediction accuracy and sensitivity of models analyzing the fused indexes and only physicochemical indexes remained at a high level. Further, the mean value of specificity of the model using fused indexes was 0.916, which was significantly higher than the model with only physicochemical indexes (average 0.822) and the model with only TCM indexes (average 0.403). Based on the RF model and data allocation ratio (8:2), we further extracted the top 20 most significant variables from the fused indexes, which included 14 physicochemical indexes and 6 TCM indexes including wiry pulse, chest tightness, spontaneous perspiration, greasy tongue coating etc. CONCLUSION: Compared with SVM and DT models, the RF model showed the best performance, especially when the ratio of the training set to test set is 8:2. Compared with single predictive indexes, the model constructed by combining physiochemical indexes with TCM indexes (i.e. the fused indexes) exhibited better predictive ability. In addition to common physicochemical indexes, some TCM indexes, such as wiry pulse, chest tightness, spontaneous perspiration, greasy tongue coating, can also improve diagnosis of MS.

Classification of COVID-19 by Compressed Chest CT Image through Deep Learning on a Large Patients Cohort

Abstract

Corona Virus Disease (COVID-19) has spread globally quickly, and has resulted in a large number of causalities and medical resources insufficiency in many countries. Reverse-transcriptase polymerase chain reaction (RT-PCR) testing is adopted as biopsy tool for confirmation of virus infection. However, its accuracy is as low as 60-70%, which is inefficient to uncover the infected. In comparison, the chest CT has been considered as the prior choice in diagnosis and monitoring progress of COVID-19 infection. Although the COVID-19 diagnostic systems based on artificial intelligence have been developed for assisting doctors in diagnosis, the small sample size and the excessive time consumption limit their applications. To this end, this paper proposed a diagnosis prototype system for COVID-19 infection testing. The proposed deep learning model is trained and is tested on 2267 CT sequences from 1357 patients clinically confirmed with COVID-19 and 1235 CT sequences from non-infected people. The main highlights of the prototype system are: (1) no data augmentation is needed to accurately discriminate the COVID-19 from normal controls with the specificity of 0.92 and sensitivity of 0.93; (2) the raw DICOM image is not necessary in testing. Highly compressed image like Jpeg can be used to allow a quick diagnosis; and (3) it discriminates the virus infection within 6 seconds and thus allows an online test with light cost. We also applied our model on 48 asymptomatic patients diagnosed with COVID-19. We found that: (1) the positive rate of RT-PCR assay is 63.5% (687/1082). (2) 45.8% (22/48) of the RT-PCR assay is negative for asymptomatic patients, yet the accuracy of CT scans is 95.8%. The online detection system is available: http://212.64.70.65/covid.

Graphic Abstract

A Prognosis Tool Based on Fuzzy Anthropometric and Questionnaire Data for Obstructive Sleep Apnea Severity

Obstructive sleep apnea (OSA) are linked to the augmented risk of morbidity and mortality. Although polysomnography is considered a well-established method for diagnosing OSA, it suffers the weakness of time consuming and labor intensive, and requires doctors and attending personnel to conduct an overnight evaluation in sleep laboratories with dedicated systems. This study aims at proposing an efficient diagnosis approach for OSA on the basis of anthropometric and questionnaire data. The proposed approach integrates fuzzy set theory and decision tree to predict OSA patterns. A total of 3343 subjects who were referred for clinical suspicion of OSA (eventually 2869 confirmed with OSA and 474 otherwise) were collected, and then classified by the degree of severity. According to an assessment of experiment results on g-means, our proposed method outperforms other methods such as linear regression, decision tree, back propagation neural network, support vector machine, and learning vector quantization. The proposed method is highly viable and capable of detecting the severity of OSA. It can assist doctors in pre-diagnosis of OSA before running the formal PSG test, thereby enabling the more effective use of medical resources.