Fuzzy partitioning of clinical data for DMT2 patients.
Miroslava NedyalkovaHaruna Luz Barazorda-CcahuanaC SârbuSergio MadurgaVasil SimeonovPublished in: Journal of environmental science and health. Part A, Toxic/hazardous substances & environmental engineering (2020)
The present study represents an original approach to data interpretation of clinical data for patients with diagnosis diabetes mellitus type 2 (DMT2) using fuzzy clustering as a tool for intelligent data analysis. Fuzzy clustering is often used in classification and interpretation of medical data (including in medical diagnosis studies) but in this study it is applied with a different goal: to separate a group of 100 patients with DMT2 from a control group of healthy volunteers and, further, to reveal three different patterns of similarity between the patients. Each pattern is described by specific descriptors (variables), which ensure pattern interpretation by appearance of underling disease to DMT2.
Keyphrases
- data analysis
- end stage renal disease
- electronic health record
- ejection fraction
- healthcare
- big data
- newly diagnosed
- chronic kidney disease
- peritoneal dialysis
- machine learning
- single cell
- prognostic factors
- type diabetes
- deep learning
- metabolic syndrome
- patient reported outcomes
- artificial intelligence
- patient reported