Abstract:
In medicine, the best diagnostician are held in high esteem, in the case of pregnant women correct diagnosis of bleeding in late pregnancy (after age of viability) makes a difference between life and death for the mother and/or baby. Information gathering from patient by clinicians during diagnostic procedures requires some skills to adequately collect required relevant information that will be adequate for diagnosis. Diagnosis, decision making are all areas of application of artificial intelligence. Due to inadequate obstetricians, efforts to better reach underserved communities and manage obstetric emergencies like bleeding in late pregnancy in developing countries is increasingly based on task shifting to community health workers, nurses, midwives or doctors (WHO, 2008) Research has proven that the major challenge is to ensure health providers extract, in their precise and simple form, the information needed for the diagnostic task. This paper, therefore, provides a formalized input generating model that addresses this shortcoming through the creation of an inference process, rule set and natural language processing (NLP). And a text-based input graphical user interface (GUI) software to diagnose the cause of bleeding with the ability for storage, retrieval and editing if need be. The results show that the proposed model will add value to the accuracy of correct diagnosis of bleeding in late pregnancy among the targeted group.
Key words: Late bleeding, natural language process, input generation, inference process.
