Abstract:
Since its emergence in the early 20th century, Malaria has been confirmed as a deadly disease that has spread throughout the world with very high mortality and morbidity. This is in accordance with the WHO report in 2018 which stated that worldwide there have been more than 220 million cases of malaria with a death rate of nearly 500 thousand cases. However, Malaria is actually a disease that can be cured and prevented if treatment initiatives are implemented early and effectively. Unfortunately, this disease is often ignored because it is considered the common cold and is only diagnosed when it has reached a critical phase. This research is expected to be an alternative for early diagnosis of malaria. Hence, confirming the presence of the malaria parasite earlier will make the treatment of this disease more effective in reducing mortality. This research is expected to produce website-based computer-assisted disease diagnosis (CAD) software enriched with deep learning algorithms to become an alternative for early diagnosis of malaria. This CAD system has the potential to provide fast and reliable malaria diagnosis and avoid detection errors by experts due to human error. This research uses various pre-trained CNN architectures that have been proven to have the best performance in extracting features and recognizing image patterns such as MobileNetV2, EfficientNetBO, RestNet50, InceptionV3 and Xception. This architecture was then modified by adding several additional layers to improve its performance. To be concluded, this research succeeded in exceeding previous studies by obtaining an accuracy value above 97%. Moreover, this developed CAD software is also equipped with various features to make it easier to use.
