| Abstrakt: |
Diabetes is a chronic (long-lasting) condition, and it is becoming more and more common in both developed and developing countries. Compared to people without diabetes, people with diabetes are more likely to have various illnesses. Comorbidity is the term used in clinical literature to describe this phenomenon in general. The majority of the recent studies have made considerable strides in extracting comorbidity patterns, but all the studies have some limitations in terms of: identifying interesting gender- and age-dependent diabetes-specific comorbidities patterns; little is known about differences in gender- and age-dependent diabetes-specific comorbidities patterns. The objective of this research paper is to identify and explore interesting gender- and age-dependent diabetes-specific comorbidities patterns from large Medical Information Mart for Intensive Care (MIMIC) datasets. An effective rule-based approach is proposed to identify interesting gender- and age-dependent diabetes-specific comorbidities patterns by considering all diagnosis information of a patient i.e., primary diagnosis, secondary diagnosis and so on of a patient. The proposed approach performs sorting to identify interesting gender- and age-dependent diabetes-specific comorbidities patterns. Finally, extracted interesting comorbidities patterns are validated using a network-based method in a non-traditional manner. The experimental results showed that the occurrence of comorbidity rises with age both in case of male and female patients and also showed that some conditions like hypertensive renal disease will co-occur with diabetes in almost all age groups both in case of male and female patients. In addition, it is also observed that the comorbid conditions depends on gender. Notably, our findings identify new, intriguing, and clinically significant gender- and age-dependent patterns of diabetes-specific comorbidities, assisting clinical practitioners in recommending the best course of treatment for diabetic patients. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink