A fuzzy expert system to monitor pregnancy

Ambissa, Aschale

dc.contributor.author	Ambissa, Aschale
dc.date.accessioned	2020-06-04T06:24:47Z
dc.date.available	2020-06-04T06:24:47Z
dc.date.issued	2020-02-18
dc.identifier.uri	http://hdl.handle.net/123456789/10872
dc.description.abstract	Pregnancy is the most important natural phenomena in maternal life, which is a critical issue and stands to lead mothers and newborn chide to death. World Health Organization stated that each year, nearly 300,000 women die from complications of pregnancy and childbirth, and an estimated three million newborns die within the first month of life. According to Ethiopian’s context, the ratio of medical practitioner to the patient is imbalanced and difficult to treat, diagnosis as well as screen checkup of pregnancy for mothers due to lack of intelligent technologies. The main objective of this study is to develop a fuzzy expert system to monitor pregnancy for mother’s safety care and describe its risk possibility in terms of linguistic terms. To do this, we used a Mamdani fuzzy modeling system. In order to describe the fuzzy logic concepts, we used seven input variables with their linguistic terms to emphasis the degree of membership functions. By using theses input variables, the study achieves the possibility of the pregnancy risk status to be 50% by help of centroid defuzzification method type. This is a design science research type which used to articulate artifacts and discovering new innovations regarding to intelligent technologies. This study used both user acceptance and system testing to get its accurate performance. For the evaluation purpose, we used user acceptance evaluation which is evaluated by five evaluators selected from the hospital and they evaluated based on eight closed questions and open-ended feedbacks and its accuracy performance is 90.20%.the evaluators are selected based on qualification, experience and field of specialization department. The system testing evaluation method depends pregnancy sample dataset it predicts the risk possibility using learning algorithm classifier (Random Forest algorithm) and expressed as seven linguistic terms and the evaluation metrics are 94.23% ,95.00%, 93.00, 94.00%, accuracy, precision, recall and f-measure respectively. To get the system testing, we use pregnancy sample dataset collected and prepared from Bahir Dar Felege Hiwot Referral Hospital. The finding of this research can support and enables pregnant mothers to minimize pregnancy risk status happen during pregnancy and create awareness of using intelligent technologies. It also helpful to physicians and doctors in the healthcare system to give right way of decision-making process and facilitate the curing style of pregnant during pregnancy time.	en_US
dc.language.iso	en	en_US
dc.subject	Information Technology	en_US
dc.title	A fuzzy expert system to monitor pregnancy	en_US
dc.type	Thesis	en_US