Pathway ranking by different methods on diabetes datasets.
The proposed algorithm identified true disease genes at 80%, 71 %, 69 %, 66 57%, and 57% for breand cancer, prostate cancer, Alzheimer's disease, colorectal cancer, diabetes mellitus, and lung cancer, respectively, significantly higher than all the rest; however, it was comparable to other methods on diabetes.
The literature review revealed that both national and international publications focused on the risk factors of diabetes in certain sub-populations, treatment methods of diabetes with special emphasis on complications, prevalence of disease in small geographical areas and few studies on quality of life measurements of diabetes patients.
To evaluate the simultaneous effect of different risk factors or risk indicators including age, gender, duration of diabetes, hypertension, HbA1C, nephropathy, hyperlipidemia and method of diabetes control on the presence of DR (the response variable), a multiple logistic regression model was used.
To judge quality, we abstracted information on population source; method of diabetes and outcome ascertainment; whether diabetes was the primary exposure variable or one of a group of prognostic variables; and adjustment for confounders.
Treatment methods for diabetes focus on improved glucose control and cardiovascular prevention.
The two methods had comparable performance, with the only statistically significant difference between the two being on diabetes, where the machine learning method was superior.
As an example, recent literature on diabetes costs is increasingly adopting non-traditional modeling methods [ 23– 27].
The following information was extracted from each study article: author names, year of publication, study design, country, period of data collection, description of participants, age, and data on diabetes assessment (type of diabetes, data collection, and method).
His research focused on diabetes.
Diabetes UK estimates that the NHS is already spending about £10bn a year on diabetes.
