Artificial Intelligence Application For Malaria Diagnosis Based On Thin Blood Smears Microphotograph
Anto Satriyo Nugroho, Center for the Assessment and Application of Technology (TIK-BPPT)

Artificial Intelligent has been widely applied in a broad spectrum, from robotics, meteorology, and many others including medical diagnosis. The aim of AI research is to make the machine perform the human's job and obtain advantages in term of cost and performance. One of important AI application is an intelligent system to assist medical diagnosis. A significant benefit of this application is to increase the diagnostic accuracy, in particular when the diagnosis is carried out by less experience practitioners. Our study is focused on the AI application for Automatic Malaria computer aided diagnosis. Malaria is an infectious tropical disease caused by unicellular protozoan parasite from the genus Plasmodium that infects human by entering bloodstream.
In Indonesia, the highest prevalence of malaria is found in remote and forest related areas including Papua, Kalimantan, and Sulawesi where the number of medical experts and high quality medical facilities are not sufficient. The diagnosis is conducted by analyzing manually both thick and thin Giemsa stained blood smears under light microscope, to identify the presence of parasites in erythrocytes. Confirming a negative slide requires great effort and time, especially in a Mass Blood Survey.

Furthermore, the result of the diagnosis highly depends on the medical practitioner's expertise and experience. Computer Aided Diagnosis system is developed to overcome the limitation of the expert by transforming the knowledge into an intelligent system to perform the diagnosis automatically. The performance of the system is evaluated by comparing the diagnosis results with those obtained by medical practitioners.

Background Review Article:

J.S. Duncan, N. Ayache, "Medical Image Analysis: Progress over Two Decades and the Challenges Ahead," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No. 1, pp.85-106, January 2000