Artificial intelligence applications in allergic rhinitis diagnosis: Focus on ensemble learning.

Background: The diagnosis of allergic rhinitis (AR) primarily relies on symptoms and laboratory examinations. Due to limitations in outpatient settings, certain tests such as nasal provocation tests and nasal secretion smear examinations are not routinely conducted. Although there are clear diagnostic criteria, an accurate diagnosis still requires the expertise of an experienced doctor, considering the patient's medical history and conducting examinations. However, differences in physician knowledge and limitations of examination methods can result in variations in diagnosis. Objective: Artificial intelligence is a significant outcome of the rapid advancement in computer technology today. This study aims to present an intelligent diagnosis and detection method based on ensemble learning for AR. Method: We conducted a study on AR cases and 7 other diseases exhibiting similar symptoms, including rhinosinusitis, chronic rhinitis, upper respiratory tract infection, etc. Clinical data, encompassing medical history, clinical symptoms, allergen detection, and imaging, was collected. To develop an effective classifier, multiple models were employed to train on the same batch of data. By utilizing ensemble learning algorithms, we obtained the final ensemble classifier known as adaptive random forest-out of bag-easy ensemble (ARF-OOBEE). In order to perform comparative experiments, we selected 5 commonly used machine learning classification algorithms: Naive Bayes, support vector machine, logistic regression, multilayer perceptron, deep forest (GC Forest), and extreme gradient boosting (XGBoost).To evaluate the prediction performance of AR samples, various parameters such as precision, sensitivity, specificity, G-mean, F1-score, and area under the curve (AUC) of the receiver operating characteristic curve were jointly employed as evaluation indicators. Results: We compared 7 classification models, including probability models, tree models, linear models, ensemble models, and neural network models. The ensemble classification algorithms, namely ARF-OOBEE and GC Forest, outperformed the other algorithms in terms of the comprehensive classification evaluation index. The accuracy of G-mean and AUC parameters improved by nearly 2% when compared to the other algorithms. Moreover, these ensemble classifiers exhibited excellent performance in handling large-scale data and unbalanced samples. Conclusion: The ARF-OOBEE ensemble learning model demonstrates strong generalization performance and comprehensive classification abilities, making it suitable for effective application in auxiliary AR diagnosis.

Microarray analysis of differentially expressed microRNAs in allergic rhinitis.

BACKGROUND: Allergic rhinitis (AR) is a common disease characterized by chronic inflammation of the nasal mucosa, but we have not fully understood the mechanism responsible for the development of AR. MicroRNAs (miRNAs) are short endogenous noncoding RNAs regulating protein translation through a mechanism known as RNA interference. To understand the molecular mechanisms of miRNA involved in the pathogenesis of AR, expressed miRNAs in AR were investigated through genomewide microarray analysis. METHODS: Mammalian miRNA microarrays containing whole human mature and precursor miRNA sequences were used for analyzing eight samples of nasal mucosa of AR and eight samples of nonallergic patients. Quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) of some different expressed miRNAs was used to confirm the array results. RESULTS: The miRNA microarray chip analysis identified 421 miRNAs differentially expressed in the nasal mucosa of AR, and a total of 9 miRNAs were identified in the AR group with twofold change compared with control samples (p < 0.05). These included up-regulated miRNAs, hsa-hsa-miR-7, and hsa-miRPlus-E1194, and down-regulated miRNAs, hsa-miR-498, hsa-miR-187, hsa-miR-874, hsa-miR-143, hsa-miR-886-3p, hsa-miR-224, and hsa-miR-767-5p. RT-PCR results also confirmed that part of differentially expressed miRNAs as hsa-miR-224, hsa-miR-187, and hsa-miR-143 were down-regulated in AR. CONCLUSION: The report indicated that many miRNA expressions were altered in AR and differentially expressed miRNAs appear to be involved in the development of AR. The study of miRNAs may lead to a better understanding about the roles of identified miRNAs in the pathogenesis of AR; this would be considered in future therapeutic strategies.

A meta-analysis of the association of exhaled carbon monoxide on asthma and allergic rhinitis.

The objective of this study is to evaluate the effect of exhaled CO (eCO) on the development of asthma and allergic rhinitis (AR) by means of reviewing published literature. The literatures published between January 1997 and December 2008 from the US National Library of Medicine (NLM) Database were obtained according to inclusion criteria. Meta-analysis of randomized controlled trials (RCTs) was performed. CO levels of asthma and AR patients were compared with that of normal controls. HO-1(heme oxygenase-1) expression and effect of corticosteroids on eCO levels were also analyzed. Fifteen studies concerning asthma and four studies concerning AR were included in this analysis. Heterogeneity from different studies was evident (P < 0.0001), so a random-effects model was preferred. The meta-analysis revealed that asthmatic patients had significantly higher levels of eCO compared to normal controls. There was significant difference between asthma and control groups in terms of eCO (combined weighted mean difference (WMD) 1.33 (95% confidence interval 0.72 to 1.95), P < 0.0001), and no significant difference between AR and control (combined WMD 0.93 (95% confidence interval -0.54 to 2.40), P = 0.22). HO-1 expression were also reviewed, asthma group produced greater expression of HO-1 than control group with significant difference (combined standardized mean difference (SMD) 2.98 (95% confidence interval 1.13 to 4.84), P = 0.002). After corticosteroid therapy, significantly different levels of eCO were produced after corticosteroid therapy than did asthma group (combined WMD -1.23 (95% confidence interval -2.43 to -0.03), P = 0.04). The analysis reveals that eCO levels were significantly raised in asthma and it may attribute to high expression of HO-1, but there were no significantly high eCO levels between AR and control groups. Due to sensitivity to corticosteroid inhibition, eCO may be used as a practical marker to detect and monitor exacerbation of asthma.

Possible contribution of endogenous carbon monoxide to the development of allergic rhinitis in guinea pigs.

BACKGROUND: The mechanisms responsible for the development of allergic rhinitis(AR) are not fully understood. The present study was designed to explore the possible roles of carbon monoxide(CO) on the pathogenesis of AR. METHODS: AR guinea pig model was established by nasal ovalbumin sensitization. Twenty-four AR guinea pigs were divided into four groups, 6 in each: Saline control group, AR sensitized group, Hemin treated group, and Zinc protoporphyrin (ZnPP) treated group. The frequency of sneezing and nose rubbing was recorded. Leukocyte infiltration in nasal lavage fluid, serum IgE level and plasma CO were measured. Expression of heme oxygenase-1 (HO-1) mRNA in nasal mucosa was determined by real time RT-PCR, and expression of HO-1 protein was detected by immunohistochemistry. RESULTS: The frequency of sneezing and nose rubbing, leukocyte infiltration, serum IgE, plasma CO, and HO-1 mRNA levels in sensitized guinea pigs were higher than those of control (P < 0.05). Except for serum IgE level, all above parameters were even higher (P < 0.05) when treated with Hemin, a heme oxygenase-1 inducer; but significantly decreased (P < 0.05) when treated with ZnPP, a heme oxygenase inhibitor. Immunohistochemical results showed that positive staining of HO-1 was present in the lamina of mucosa of sensitized guinea pigs, and there was an increase of HO-1 immunoreactivity with Hemin administration (P < 0.05) and a decrease with ZnPP treatment. CONCLUSION: The endogenous CO may take part in the inflammation process of AR and is positively correlated with expression of HO-1 in nasal mucosa. Endogenous CO plays a significant role in the pathogenesis of AR.