EVALUATION OF AI MEDICAL INTERPRETATION

AI has been the core of many studies from pure human studies to scientific experiments. The use of AI in translating medical documents has been going on for over a decade with solid results; however, the effectiveness of AI in interpreting in clinical settings is controversial due to several reasons including the language pairs tested, the expected accuracy, the dialect issues, and the satisfaction of both clinicians and patients. This study aims at studying the effectiveness and accuracy of three AI platforms interpreting between an English doctor and a Yemeni patient from Sana’a. The same dialogue was interpreted by a human person. The comparison shows that Arabic is a low resource language, let alone the Yemeni dialect. The intervention of a human interpreter is vital to maintain a high level of accuracy. Further studies should evaluate other Yemeni dialects and more complicated medical situations.