1. Methodology development, data analysis and application for continuous monitoring of physical signals

Lead: Douglas
Members: Douglas, Chen HUANG, Teng ZHANG, Yu JIN, Dongliang LENG, Dandan WANG, Shixue SUN
Introduction: The continuous monitoring of physical signals (CMPS) plays a critical role in digital health/medicine. One key research direction in our lab is to develop/explore quantitative methods for analyzing CMPS studies and to apply them in various areas to improve human health. Three key CMPS research areas in our lab are continuous glucose monitoring with application to diabetes and nutrition, continuous monitoring of respiratory signals with application to COPD and asthma, continuous monitoring of cardiovascular activity such as ECG with application to cardiovascular diseases and sports.
Examples of articles :
(1). Zhang XD#*, Zhang Z#, Wang D. 2018. CGManalyzer: a R package for analyzing continuous glucose monitoring studies Bioinformatics (DOI: 10.1093/bioinformatics/btx826)
(2). Nin J#, Shi Y#*, Cai M, Cao Z, Wang D, Zhang Z, Zhang XHD*. 2017. Detection of sputum by interpreting the time-frequency distribution of respiratory sound signal using image processing techniques. Bioinformatics (DOI: 10.1093/bioinformatics/btx652)
(3). Zhang XD#, Pechter D#, Yang L, Ping X, Yao Z, Zhang R, Shen X, Li N, Connick J, Hill D, Nawrocki AR, Li C*. 2017. Decreased complexity of glucose dynamics preceding the onset of diabetes in preclinical species. PloS ONE 12(9):e0182810 .( doi:10.1371/journal.pone.0182810)

2. Deep investigation of long noncoding RNAs implicated in the occurrence and process of respiratory diseases using high-throughput RNA-seq technology

Lead: Chen HUANG
Members: Chen HUANG, Dongliang LENG, Douglas
Introduction:respiratory diseases, such as asthma, Allergic bronchopulmonary aspergillosis (ABPA) kill large crowd of people every year, which brings a great negative impact on families and societies. Currently, the detailed mechanism of many respiratory diseases is still poor understood, like ABPA. In our study, we apply a series of bioinformatics analysis based on the deep RNA sequencing to the patients with respiratory diseases, comparing the health group with disease group or pre- and post-treatment, in order that we could uncover the roles of long noncoding RNAs in the occurrence and process of respiratory diseases based on the deep RNA sequencing.

3. A COPD database for precision medicine

Lead: Dongliang LENG
Members: Dongliang LENG, Chen HUANG, Dandan WANG, Chang CHEN, Yu JIN, Shixue SUN, Teng ZHANG, Douglas
Introduction: With the rapid development of digital medical in recent years, a huge volume of medical data which contains vast amount of information has been collected without fully exploration. So it is meaningful to manage these medical data for digging its potential value. This database is built to collect and share the real-world data of COPD patients on user-friendly web pages. This database stores continuously monitored biological data of COPD patients, such as PAW, flow and SpO2. The data were collected at a frequency of 5 Hz, and the data volume reaches 0.4 million for a full day continuously monitoring recording. Currently this database has more than 100 million data points from 13 patients and more data is being collected. The data can be filtered and downloaded by their different attributes. For the further development of the database, users are encouraged to share their own data via the upload function. Besides, some mainstream algorithms and software are provided with in the website with a brief introduction for the uses to deep mining the big data.

4. CRISPR/CAS9 high-throughput screening studies and data analysis

Lead: Shixue SUN
Members:: Shixue SUN, Chen HUANG, Dongliang LENG, Douglas
Introduction:CRISPR/cas9 technology is now being widely used in a variety of biological researches for its high efficiency in targeted genome editing. CRISPR/cas9 based high-throughput screens provide a new powerful tool to identify target cancer genes which then lead to new drug design and development. The key to a successful CRISPR/cas9 high-throughput screen is the combination between experimental big data and mathematical methods. The aims of this project are to optimize chemical synthesis method of CRISPR/Cas9 library, to prepare large-scale CRISPR/Cas9 library with optimized quality control, and to further develop a HTS analysis software based on the real world data derived from CRISPR/Cas9 based screen.
This project is supported by 018 Guangzhou science and technology innovation funding (广州市2018 年科技创新发展专项资金) and will be conducted with the collaboration of Guangzhou RiboBio Co., Ltd.

5. Medical studies for various diseases collaborated with hospitals

Lead: Dandan WANG
Members: Dandan WANG, Yu JIN, Teng ZHANG, Dongliang LENG, Chang CHEN, Shixue SUN, Douglas
Introduction: Our medical studies are aimed at using continuous monitoring of physiological signals, such as blood glucose level, air flow, blood pressure, heart rate, pulse rate, respiration rate, oxygen saturation, etc., for disease diagnosis and prognosis. Currently, we are conducting several studies for diabetes, allergic rhinitis and asthma collaborated with the first affiliated hospital of Guangzhou Medical University as well as for COPD, pressure injury and acute respiratory failure collaborated with Chaoyang Hospital, affiliated hospital of Capital Medical University. We are getting involved in experimental designs, sample size determination, coordination of experimental execution, data collection, data analysis and so on, for these studies.
Examples of articles :
(1). Xue M#, Wang D#, Cao Z*, Luo Z, Zheng Y, Lu J, Zhao Q, Zhang XD*. 2018. A study demonstrating the potential of using transcutaneous oxygen and transcutaneous carbon dioxide tensions to assess the risk of pressure injury. (Submitted)
(2). Chen X#, Wang D#, Lin J, Jin Y, Deng S, Huang L, Zhang T, Zheng J, Sun B, Zhang XD*. 2018. Clinical Practice of Integrating CSII, CGM and Personalized Data Analysis to Treat a Pregnant Woman with Type 2 Diabetes. (Submitted)

6. Dynamics of physiological signals for sports with focus on cardiovascular health

Lead: Chang CHEN
Members: Chang CHEN, Yu JIN, Dandan WANG, Dongliang LENG, Shixue SUN, Douglas
Introduction: Aerobic exercise and anaerobic exercise have different exercise effects for fitness. The traditional indicators can’t describe them comprehensively and effectively. Our research direction is how to analysis the heart function after exercise by using dynamic continuous physiological indicators such as heart rate, pulse, blood pressure, etc. Our main method is heart rate variability. We are also involved in building related R packages at the same time. In addition, we conduct a study of the agreement analysis of different electrocardiographic instruments and the comparison of heart rate variability between patients with autonomic neurosis and healthy people.
Examples of articles :
(1). Chen C#, Jin Y#, Lo IL, Zhao H, Sun B, Zhao Q, Zheng J, Zhang XD*. 2017. Complexity Change in Cardiovascular Diseases. International Journal of Biological Sciences (Accepted)

7. Dynamics of respiratory signals for COPD and other diseases

Lead: Yu JIN
Members:: Yu JIN, Chang CHEN, Dandan WANG, Dongliang leng, Douglas
Introduction: Chronic obstructive pulmonary disease (COPD) is a major threaten to public health. Its diagnosis, assessment, and treatment vary based mostly on the severity of airflow limitation. The largest component of related healthcare costs is attributed to admissions due to its acute exacerbation (AECOPD). The evidence that might support the effectiveness of telemonitoring interventions in COPD is limited partially due to the lack of useful predictors for early detection of AECOPD. Therefore, the aim of this study is to explore the application of continuous measurement of physiological parameters in the acute exacerbation of COPD. This is a parallel-group, open labeled, and observational study. Subjects will be interviewed via phone or lab visit to determine if they meet inclusion criteria (see below Inclusion Criteria and Exclusion criteria). The experiment will last for 12 months. We will use ventilators to collect continuous SpO2, Paw, Flow data. Once a patient is admitted to hospital due to an acute exacerbation, the event and time are recorded immediately. The goal of this study is to develop a robust, clinically based model to predict risk of frequent exacerbation.
Examples of articles :
(1). Jin Y#, Chen C#, Cao Z, Sun B, Lo IL, Liu T, Zheng J, Sun S, Shi Y*, Zhang XD*. Entropy change of biological dynamics in COPD. International Journal of Chronic Obstructive Pulmonary Disease. ( doi:10.2147/COPD.S140636)

8. Methodology development and numerical simulation

Lead: Teng Zhang
Members:: Teng Zhang, Douglas
Introduction:Our research aims at the development of optimal quantitative methods that improve reliability and stability of signal processing and nonlinear analysis. The traditional methods, such as wavelet analysis and multiscale entropy, still have limitations in clinical application despite their wide range of uses in the analysis of physiological signals. Therefore, in order to make non-linear methods more suitable for analysis of clinical data, these traditional methods need to be improved. We currently are working on the assessment of traditional algorithms and the development of new algorithms based on numerical simulation. The primary advantage of numerical simulation is that the influencing factors are controllable so that the influence of the factors on the final result can be investigated independently. Besides, numerical simulation is time- and cost-saving. The mathematical models we are numerically simulating take into consideration certain noises, sine functions, and Lorentz oscillator.