Volume 20, Issue 12 (3-2018)
J Arak Uni Med Sci 2018, 20(12): 44-50 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Shariatpanahi S P, Habibi D, Rafiei M, Ghandi Y, Anvari M. Determination of Relations between Systolic Blood Pressure and Heart Attack in Patients with Type 2 Diabetes with Association Rules. J Arak Uni Med Sci 2018; 20 (12) :44-50
URL: http://jams.arakmu.ac.ir/article-1-5379-en.html
URL: http://jams.arakmu.ac.ir/article-1-5379-en.html
1- , d.habibi@arakmu.ac.ir
Abstract: (3639 Views)
Abstract
Background: Today, the high prevalence of diabetes and its complications are one of the most important public health issues worldwide. For this reason, finding relations between diabetes risk factors is very effective in preventing and reducing complications. For discovering these relations, the data mining methods can be used. By extracting association rules, which is one of the data mining techniques, we can discover the relations between a large numbers of variables in a disease.
Materials and Methods: The population of this study was 1046 patients with type 2 diabetes, whose data had recorded between 2011 and 2014 at the Special Clinic for Diabetes in Tehran's Imam Khomeini Hospital. After pre-processing step with SPSS19 software, 573 people entered the analysis phase. The FP-Growth algorithm was applied to the data set to discover the relations between heart attack and other risk factors using Rapid miner5 software. Relations, after extraction, were given to the doctor to confirm clinical validation.
Results: The obtained results of studying these 573 people (Including 292 (51%) women and 281 (49%) men, with age range 27 to 82 years) showed that the lack of blood pressure, creatinine and diastolic blood pressure at its normal level, despite higher systolic blood pressure level than normal, doesn't increase the probability of heart attack.
Conclusion: Using association rules is a good way of identifying relations between the risk factors of a disease. Also, it can provide new hypotheses to do epidemiological studies for researchers.
Background: Today, the high prevalence of diabetes and its complications are one of the most important public health issues worldwide. For this reason, finding relations between diabetes risk factors is very effective in preventing and reducing complications. For discovering these relations, the data mining methods can be used. By extracting association rules, which is one of the data mining techniques, we can discover the relations between a large numbers of variables in a disease.
Materials and Methods: The population of this study was 1046 patients with type 2 diabetes, whose data had recorded between 2011 and 2014 at the Special Clinic for Diabetes in Tehran's Imam Khomeini Hospital. After pre-processing step with SPSS19 software, 573 people entered the analysis phase. The FP-Growth algorithm was applied to the data set to discover the relations between heart attack and other risk factors using Rapid miner5 software. Relations, after extraction, were given to the doctor to confirm clinical validation.
Results: The obtained results of studying these 573 people (Including 292 (51%) women and 281 (49%) men, with age range 27 to 82 years) showed that the lack of blood pressure, creatinine and diastolic blood pressure at its normal level, despite higher systolic blood pressure level than normal, doesn't increase the probability of heart attack.
Conclusion: Using association rules is a good way of identifying relations between the risk factors of a disease. Also, it can provide new hypotheses to do epidemiological studies for researchers.
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
