|
|
| |
| A Bayesian Network Expert System For Diagnosing Hormone Imbalance |
Oyenike Mary Olanrewaju, Mukhtar Umar Shitu, Eli Jiya
|
Abstract:
In the overall health of women, hormones play significant roles. A fluctuation in hormone level leads to imbalance which at the end may have a negative impact on the health of the individual affected. Most diagnostic errors occur as a result of uncertainty, missing information and miscommunication. The lack of certainty about an outcome of diagnosis may lead to wrong medication, and this may lead to death of a patient or an irreversible medical condition. Providing correct information and making accurate diagnosis at the right time has the potential to prevent these unfavorable circumstances. This research proposes a web-based interactive expert system for the diagnosis and prediction of hormone imbalance using Bayesian Networks. Based upon the accuracy measure, all the classifiers used in the modeling process performed well. The selected model used for hormone imbalance expert system (Bayesian Networks) has an accuracy value of 98.33% with 0.98 precision and 0.98 recall values. Furthermore, the expert system was evaluated through usability testing (efficiency, few errors, learnability, memorability and satisfaction). The overall experience value of 4.29 was obtained after the usability testing. The expert system in this study was designed in such a way that a user can make use of the application remotely by installing it on an android device. This research discovered that expert system could be a more efficient method in addressing hormone imbalance diagnosis. Also, adoption of experts system could help save time in the diagnostic procedure of patients.
|
|
|
| | |