CBGOUS/T2D_HTN_analysis

Analysis code for the Scientific Data paper "Hypertension and Diabetes Patients' Registry in Ethiopia: A Dataset from Selected Health Centres in Addis Ababa" # Dataset Analysis 1. Cleaning We started the cleaning process by identifying any duplicate entries in the dataset. We also corrected any errors in the dataset, such as missing or incorrect values. We identified and removed duplicate entries, we used the Z-Score methodology correct any deviations within the dataset. 2. Normality Assessment To assess the normality of our data distribution. This plot helped us visualize whether the data follows a normal distribution or not. 3. Statistical hypothesis test We employed the Student t-test to compare means and the Kolmogorov Smirnov two-sample test to assess the similarity of the distributions between the different columns of our dataset, with the aim of investigating their potential relationships. 4. Correlation Analysis We examined the correlation between T2D (Type 2 Diabetes) and HTN (Hypertension) conditions in the dataset. We used the Pearson correlation coefficient to measure the strength of the linear relationship between these two variables. We also plotted a scatter plot to visualize the correlation between these two variables. 5. Visualisation We created plots to visually represent the correlation and observations within our data. We aimed to gain insights into the relationships between different variables and any patterns or trends that might be present. 6. Data Mining: We extracted abstracts of publications from PubMed database that contain datasets with T2D and HTN conditions. We then applied a list of regular expressions (patterns) to facilitate the automatic extraction of the number of participants used in several studies from a collection of text abstracts. By including various synonyms for the term 'participants', such as 'subjects', 'patients', 'individuals', and others, the regular expressions are intended to match a wide range of possible strings within the text that represent the number of participants in each study. By automating this process, the code aimed to save time and effort compa …