Home.RESEARCH.Matem.ly – A Customized Electronic Medical Record System specializing in maternity care for a private hospital in Sri Lanka
Matem.ly – A Customized Electronic Medical Record System specializing in maternity care for a private hospital in Sri Lanka
Sarah Ratnayake and Kumudini Sarathchandra
Paper based Medical Records, Electronic Medical Records, Maternity Care, Dietary Recommendation
18th February 2022
Over the decades, the need for recording and examining pregnant women ‘s medical records has been emphasized since they are crucial in calculating the country’s maternal mortality rate. Among the several perceived challenges confronting private healthcare institutions in Sri Lanka is the fact that many of them continue to maintain medical records on paper. Numerous issues develop when medical records are maintained on paper, including inaccuracies and a significant lack of information, such as misplaced medical records, excessive use of paper and other stationary materials that are not ecologically friendly, and a lack of updated medical files. The project was undertaken to ascertain the difficulties associated with the patient record management system at a Sri Lankan maternity care center and to then develop, build, and assess an information management system that would benefit both medical workers and pregnant patients. The project was created in stages, each step using an agile structure. The web-based prototype was created to minimize the problems involved with maintaining paper medical records for hospital personnel and patients receiving maternity care. An automated diet recommendation algorithm was developed to provide nutritional recommendations based on a patient’s calcium, iron, and folate levels, three of the most frequently examined micronutrients during pregnancy to guarantee the mother’s and infant’s health and safety. It was included in the programme as a feature that sets it apart from similar solutions. A machine learning algorithm will be implemented in the future with the data gathered from the rule-based model to train the algorithm.