Proposed CAD system using image processing and deep learning techniques for detecting malaria
Malaria is a disease transmitted by female Anopheles mosquitoes that can be deadly, especially in regions with limited health facilities like rural areas in Eastern Indonesia.
Rapid and accurate diagnosis is important for reducing mortality rates, but there is a shortage of parasitologists to evaluate blood smear slides.
Read this open access paper on the FACETS website.
To address this issue, researchers proposed a computer-aided detection (CAD) system for malaria using image processing and deep learning techniques.
They developed a method that uses optimized double-Otsu to detect malaria parasites and deep learning to recognize and segment them in microscopic images.
The method was tested on a dataset of 468 infected-malaria images from Indonesia and achieved a high F1-score of 0.91 in parasite detection, indicating its potential for use in CAD malaria detection.
The proposed method also outperformed original semantic segmentation methods in terms of sensitivity, specificity, and F1-score for parasite segmentation.
Read the paper — A combination of optimized threshold and deep learning-based approach to improve malaria detection and segmentation on PlasmoID dataset by Hanung Adi Nugroho and Rizki Nurfauzi.
