| Popis: |
For the past few years, health care data has reached a genomic scale in volume as well as in complexity. In the field of healthcare informatics, mining and exploiting massive health data help to analyze different disease outcomes, hidden patterns, and progression of chronic disease so that preventive clinical measures can be practiced. But, it is observed that due to the lack of structured and complete clinical information, often it becomes difficult to anticipate any long-term complications, comorbidity, and mortality associated with a disease. The health records we have collected so far for the comorbidity study in type 2 diabetes, lack accuracy, and consistency. Therefore, we presented a fully annotated and structured dataset with 5000 sample records, considering laboratory markers, complications, comorbidities, and risk factors. The dataset includes 20 relevant clinical parameters of type 2 diabetes patients which can be used in machine learning models for clinical outcome prediction and patient care. |
