Ensemble Extreme Learning Machine Method for Hemoglobin Estimation Based on PhotoPlethysmoGraphic Signals.

Non-invasive detection of hemoglobin (Hb) concentration is of great clinical value for health screening and intraoperative blood transfusion. However, the accuracy and stability of non-invasive detection still need to be improved to meet clinical requirement. This paper proposes a non-invasive Hb detection method using ensemble extreme learning machine (EELM) regression based on eight-wavelength PhotoPlethysmoGraphic (PPG) signals. Firstly, a mathematical model for non-invasive Hb detection based on the Beer-Lambert law is established. Secondly, the captured eight-channel PPG signals are denoised and fifty-six feature values are extracted according to the derived mathematical model. Thirdly, a recursive feature elimination (RFE) algorithm is used to select the features that contribute most to the Hb prediction. Finally, a regression model is built by integrating several independent ELM models to improve prediction stability and accuracy. Experiments conducted on 249 clinical data points (199 cases as the training dataset and 50 cases as the test dataset) evaluate the proposed method, achieving a root mean square error (RMSE) of 1.72 g/dL and a Pearson correlation coefficient (PCC) of 0.76 (p < 0.01) between predicted and reference values. The results demonstrate that the proposed non-invasive Hb detection method exhibits a strong correlation with traditional invasive methods, suggesting its potential for non-invasive detection of Hb concentration.

Asthma prevalence based on the Baidu index and China's Health Statistical Yearbook from 2011 to 2020 in China.

Background: Due to environmental pollution, changes in lifestyle, and advancements in diagnostic technology, the prevalence of asthma has been increasing over the years. Although China has made early efforts in asthma epidemiology and prevention, there is still a lack of unified and comprehensive epidemiological research within the country. The objective of the study is to determine the nationwide prevalence distribution of asthma using the Baidu Index and China's Health Statistical Yearbook. Methods: Based on China's Health Statistical Yearbook, we analyzed the gender and age distribution of asthma in China from 2011 to 2020, as well as the length of hospitalization and associated costs. By utilizing the Baidu Index and setting the covering all 31 provinces and autonomous regions in China, we obtained the Baidu Index for the keyword 'asthma'. Heatmaps and growth ratios described the prevalence and growth of asthma in mainland China. Results: The average expenditure for discharged asthma (standard deviation) patients was ¥5,870 (808). The average length of stay (standard deviation) was 7.9 (0.38) days. During the period of 2011 to 2020, hospitalization expenses for asthma increased while the length of hospital stay decreased. The proportion of discharged patients who were children under the age of 5 were 25.3% (2011), 19.4% (2012), 16% (2013), 17.9% (2014), 13.9% (2015), 11.3% (2016), 10.2% (2017), 9.4% (2018), 8.1% (2019), and 7.2% (2020), respectively. The prevalence of asthma among boys was higher than girls before the age of 14. In contrast, the proportion of women with asthma was larger than men after the age of 14. During the period from 2011 to 2020, the median [The first quartile (Q1)-the third quartile (Q3)] daily asthma Baidu index in Guangdong, Beijing, Jiangsu, Sichuan, and Zhejiang were 419 (279-476), 328 (258-376), 315 (227-365), 272 (166-313), and 312 (233-362) respectively. Coastal regions showed higher levels of attention toward asthma, indicating a higher incidence rate. Since 2014, there has been a rapid increase in the level of attention toward asthma, with the provinces of Qinghai, Sichuan, and Guangdong experiencing the fastest growth. Conclusion: There are regional variations in the prevalence of asthma among different provinces in China, and the overall prevalence of asthma is increasing.

Does seasonality affect snoring? A study based on international data from the past decade.

BACKGROUND: Though snoring is often regarded as a harmless condition that coincides with sound sleep, it is a sleep disorder that can be a potential indicator of more severe conditions such as sleep apnea syndrome. In the present study, we investigated the association between seasonal variations and snoring. METHOD: Search index for snoring (SIS) data were obtained from Google Trends and Baidu Index. SIS data were collected for the USA, India, Germany, Russia, Japan, Australia, China, and Brazil from 2011 to 2020, with the periodicity of the relationship between seasonal time series data and snoring evaluated using a time series decomposition model. RESULT: The highest average SIS growth rates from 2011 to 2020 were observed for Brazil, Japan, and Germany, with average SIS values of 94%, 68%, and 49%, respectively. The SIS of the USA, Russia, Japan, Brazil, Australia, Germany, and India increased by 22.3%, 12.4%, 11.9%, 35.4%, 12.3%, 28.0%, and 55.8%, respectively, in comparison with their SIS values in 2019, whereas for China, it decreased by 13.7%. Relative to countries in the southern hemisphere, those in the northern hemisphere showed comparable SIS trends, increasing from September to February and decreasing from March to August. CONCLUSION: The SIS data showed cyclical changes over the study period. The search index for snoring increased during the cold season or the heating season, suggesting that snoring is associated with seasonal changes.

Automatic epileptic seizure detection based on persistent homology.

Epilepsy is a prevalent brain disease, which is quite difficult-to-treat or cure. This study developed a novel automatic seizure detection method based on the persistent homology method. In this study, a Vietoris-Rips (VR) complex filtration model was constructed based on the EEG data. And the persistent homology method was applied to calculate the VR complex filtration barcodes to describe the topological changes of EEG recordings. Afterward, the barcodes as the topological characteristics of EEG signals were fed into the GoogLeNet for classification. The persistent homology is applicable for multi-channel EEG data analysis, where the global topological information is calculated and the features are extracted by considering the multi-channel EEG data as a whole, without the multiple calculations or the post-stitching. Three databases were used to evaluate the proposed approach and the results showed that the approach had high performances in the epilepsy detection. The results obtained from the CHB-MIT Database recordings revealed that the proposed approach can achieve a segment-based averaged accuracy, sensitivity and specificity values of 97.05%, 96.71% and 97.38%, and achieve an event-based averaged sensitivity value of 100% with 1.22 s average detection latency. In addition, on the Siena Scalp Database, the proposed method yields averaged accuracy, sensitivity and specificity values of 96.42%, 95.23% and 97.6%. Multiple tasks of the Bonn Database also showed achieved accuracy of 99.55%, 98.63%, 98.28% and 97.68%, respectively. The experimental results on these three EEG databases illustrate the efficiency and robustness of our approach for automatic detection of epileptic seizure.

Gesture Recognition by Ensemble Extreme Learning Machine Based on Surface Electromyography Signals.

In the recent years, gesture recognition based on the surface electromyography (sEMG) signals has been extensively studied. However, the accuracy and stability of gesture recognition through traditional machine learning algorithms are still insufficient to some actual application scenarios. To enhance this situation, this paper proposed a method combining feature selection and ensemble extreme learning machine (EELM) to improve the recognition performance based on sEMG signals. First, the input sEMG signals are preprocessed and 16 features are then extracted from each channel. Next, features that mostly contribute to the gesture recognition are selected from the extracted features using the recursive feature elimination (RFE) algorithm. Then, several independent ELM base classifiers are established using the selected features. Finally, the recognition results are determined by integrating the results obtained by ELM base classifiers using the majority voting method. The Ninapro DB5 dataset containing 52 different hand movements captured from 10 able-bodied subjects was used to evaluate the performance of the proposed method. The results showed that the proposed method could perform the best (overall average accuracy 77.9%) compared with decision tree (DT), ELM, and random forest (RF) methods.

[Research progress of epileptic seizure predictions based on electroencephalogram signals].

As a common disease in nervous system, epilepsy is possessed of characteristics of high incidence, suddenness and recurrent seizures. Timely prediction with corresponding rescues and treatments can be regarded as effective countermeasure to epilepsy emergencies, while most accidental injuries can thus be avoided. Currently, how to use electroencephalogram (EEG) signals to predict seizure is becoming a highlight topic in epilepsy researches. In spite of significant progress that made, more efforts are still to be made before clinical applications. This paper reviews past epilepsy studies, including research records and critical technologies. Contributions of machine learning (ML) and deep learning (DL) on seizure predictions have been emphasized. Since feature selection and model generalization limit prediction ratings of conventional ML measures, DL based seizure predictions predominate future epilepsy studies. Consequently, more exploration may be vitally important for promoting clinical applications of epileptic seizure prediction.

Potential Association Between Asthma, Helicobacter pylori Infection, and Gastric Cancer.

Background: The prevalence of Helicobacter pylori infection (HPI) is still high around the world, which induces gastric diseases, such as gastric cancer (GC). The epidemiological investigation showed that there was an association between HPI and asthma (AST). Coptidis rhizoma (CR) has been reported as an herbal medicine with anti-inflammatory and anti-bacterial effects. Purpose: The present study was aimed to investigate the protective mechanism of HPI on AST and its adverse effects on the development of GC. Coptis chinensis was used to neutralize the damage of HPI in GC and to hopefully intensify certain protective pathways for AST. Method: The information about HPI was obtained from the public database Comparative Toxicogenomics Database (CTD). The related targets in AST and GC were obtained from the public database GeneCards. The ingredients of CR were obtained from the public database Traditional Chinese Medicine Systems Pharmacology (TCMSP). The network pharmacology including gene ontology (GO) enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis, and molecular docking were utilized. Protein-protein interaction was constructed to analyze the functional link of target genes. The molecular docking was employed to study the potential effects of active ingredients from CR on key target genes. Result: The top 10 key targets of HPI for AST were CXCL9, CX3CL1, CCL20, CCL4, PF4, CCL27, C5AR1, PPBP, KNG1, and ADORA1. The GO biological process involved mainly leukocyte migration, which responded to bacterium. The (R)-canadine and quercetin were selected from C. chinensis, which were employed to explore if they inhibited the HPI synchronously and protect against AST. The targets of (R)-canadine were SLC6A4 and OPRM1. For ingredient quercetin, the targets were AKR1B1 and VCAM1. Conclusion: CXCL9 and VCAM1 were the common targets of AST and HPI, which might be one of the imported targets of HPI for AST. Quercetin could be an effective ingredient to suppress HPI and help prevent AST.

A comb filter based signal processing method to effectively reduce motion artifacts from photoplethysmographic signals.

A photoplethysmographic (PPG) signal can provide very useful information about a subject's cardiovascular status. Motion artifacts (MAs), which usually deteriorate the waveform of a PPG signal, severely obstruct its applications in the clinical diagnosis and healthcare area. To reduce the MAs from a PPG signal, in the present study we present a comb filter based signal processing method. Firstly, wavelet de-noising was implemented to preliminarily suppress a part of the MAs. Then, the PPG signal in the time domain was transformed into the frequency domain by a fast Fourier transform (FFT). Thirdly, the PPG signal period was estimated from the frequency domain by tracking the fundamental frequency peak of the PPG signal. Lastly, the MAs were removed by the comb filter which was designed based on the obtained PPG signal period. Experiments with synthetic and real-world datasets were implemented to validate the performance of the method. Results show that the proposed method can effectively restore the PPG signals from the MA corrupted signals. Also, the accuracy of blood oxygen saturation (SpO2), calculated from red and infrared PPG signals, was significantly improved after the MA reduction by the proposed method. Our study demonstrates that the comb filter can effectively reduce the MAs from a PPG signal provided that the PPG signal period is obtained.

Motion artifact removal from photoplethysmographic signals by combining temporally constrained independent component analysis and adaptive filter.

BACKGROUND: The calculation of arterial oxygen saturation (SpO2) relies heavily on the amplitude information of the high-quality photoplethysmographic (PPG) signals, which could be contaminated by motion artifacts (MA) during monitoring. METHODS: A new method combining temporally constrained independent component analysis (cICA) and adaptive filters is presented here to extract the clean PPG signals from the MA corrupted PPG signals with the amplitude information reserved. The underlying PPG signal could be extracted from the MA contaminated PPG signals automatically by using cICA algorithm. Then the amplitude information of the PPG signals could be recovered by using adaptive filters. RESULTS: Compared with conventional ICA algorithms, the proposed approach is permutation and scale ambiguity-free. Numerical examples with both synthetic datasets and real-world MA corrupted PPG signals demonstrate that the proposed method could remove the MA from MA contaminated PPG signals more effectively than the two existing FFT-LMS and moving average filter (MAF) methods. CONCLUSIONS: This paper presents a new method which combines the cICA algorithm and adaptive filter to extract the underlying PPG signals from the MA contaminated PPG signals with the amplitude information reserved. The new method could be used in the situations where one wants to extract the interested source automatically from the mixed observed signals with the amplitude information reserved. The results of study demonstrated the efficacy of this proposed method.

[Study of gene data mining based on informatics theory].

By combining with informatics theory, ta system model consisting of feature selection which is based on redundancy and correlation is presented to develop disease classification research with five gene data set (NCI, Lymphoma, Lung, Leukemia, Colon). The result indicates that this modeling method can not only reduce data management computation amount, but also help confirming amount of features, further more improve classification accuracy, and the application of this model has a bright foreground in fields of disease analysis and individual treatment project establishment.