Abstract:
Gynaecological diseases diagnosis is one of the important issues in the medical field globally because gynaecologists have to analyze and diagnose the disease according to various and similar symptoms. They may accidentally miss some symptoms that lead to a misdiagnosis. Hence, this paper aimed for developing a clinical decision support system (CDSS) to assist the gynaecologists in the diagnosis process for eleven types of gynaecological diseases that represent the most important diseases that are frequently diagnosed in gynaecologist's clinics which are: Polyps, Infections, Fibroids, Prolapse, Cancer, Endometrial hyperplasia, Migrants, Amenorrhea, Abortions, Dysmenorrhea and Infertility. In the proposed system, a multilayer perceptron (MLP) feed-forward neural network was used. The input layer of the proposed system included 54 input variables. An iterative process was used to determine the number of neurons and hidden layers. Furthermore, a resilient backpropagation algorithm (Rprop) was used to train the system. In particular, a 10-fold cross-validation scheme was used to access the generalization of the proposed system. We obtained 94.5% classification accuracy from the experiments made on the data that were taken from 550 patients' medical records suffering from eleven gynaecological diseases managed at the gynaecological clinics at Jordan University Hospital (JUH).
Page(s):
3030-3046
DOI:
DOI not available
Published:
Journal: Journal of Theoretical and Applied Information Technology, Volume: 98, Issue: 15, Year: 2020
Keywords:
Multilayer perceptron MLP
,
Clinical Decision Support Systems CDSSs
,
Resilient Backpropagation Algorithm Rprop