TEMPO.CO, Jakarta - The utilization of artificial intelligence (AI) in the health sector continues to evolve, including to aid in diagnosing tropical diseases such as malaria, which still poses a serious threat in several regions of Indonesia.
Researcher from the Artificial Intelligence and Cyber Security Research Center of Indonesia's National Research and Innovation Agency (BRIN), Anto Satriyo Nugroho, explained the potential of AI technology in accelerating and improving the accuracy of malaria diagnosis through the analysis of microscopic images of blood smears.
According to Anto, the diagnosis of malaria has generally been done manually through microscopic examination. This process not only consumes time but also heavily relies on the skill and precision of healthcare workers. Challenges such as fatigue, limited numbers of experts, and the morphological variations of parasites in different stages of their life cycle often hinder ensuring accurate diagnoses.
In response to these challenges, BRIN has developed an AI-based diagnosis system to assist healthcare workers in identifying the presence of malaria parasites. The system works by analyzing thin and thick blood smear microscopic photos to recognize signs of infection.
"Our main goal is to create a Computer-Aided Diagnosis system that can automatically recognize the status of malaria from blood smear images," Anto said in a written statement on Tuesday, May 13, 2025.
The researcher, with a background in image processing and biometrics research, mentioned that in developing this system, they used a dataset consisting of 1,388 blood smear microphotos collected from various endemic areas in Indonesia.
"The research data were obtained from various endemic areas in Indonesia, such as Kalimantan, Papua, and Sumba, in collaboration with the Eijkman Institute for Molecular Biology as part of BRIN Research Center," said Anto, who currently serves as the Head of the PR-KAKS BRIN.
According to him, these samples include various types of malaria parasites, such as Plasmodium falciparum, P. vivax, P. malariae, and P. ovale, including one case of mixed infection and one negative sample.
"Based on the analysis of 35 micrographs from real cases in endemic areas of Indonesia (covering 3,362 cells), this system has shown significant capability in identifying malaria parasites," explained Anto.
In developing this AI-based malaria diagnosis system, initial test results have shown promising performance.
"The system's sensitivity reaches 84.37 percent in distinguishing healthy and infected cells, with an accuracy value (F1-score) of 80.60 percent and a positive predictive value (PPV) of 77.14 percent in identifying parasite species and stages," said Anto. This means, according to Anto, the system is quite reliable in distinguishing healthy blood cells from infected ones.
Opening Opportunities for Remote Diagnosis
This diagnosis system is also designed to support mass blood surveys in the field. "In endemic conditions, one blood smear may require observation of 500 to 1,000 erythrocytes or 200 leukocytes. AI can expedite this process without sacrificing accuracy," said Anto.
Aside from efficiency, the system also opens up possibilities for remote diagnostics, making it highly relevant for use in remote areas. Microscopic knowledge and experience are also stored in the AI system, thus assisting healthcare workers with limited training.
Anto also emphasized that the development of AI in the biomedical field requires special attention to the characteristics of datasets, data quality, model selection, and appropriate performance evaluation methods. In other words, AI cannot work alone. Collaboration between computational experts and biomedical researchers is essential for this technology to be reliable.
"We cannot solely rely on computational capabilities. Understanding the medical context is key in ensuring that diagnosis results are truly beneficial for patients," he said.
Editor's Choice: BRIN: Kosmos 482 Spacecraft Module Crashes into Pacific Ocean at Midnight
Click here to get the latest news updates from Tempo on Google News
New Species of Blind Lizard Discovered in Indonesia
1 jam lalu
A new species of blind lizard has been discovered in Buton Island, Southeast Sulawesi, Indonesia
Indonesia Sanctions 343 Landfills for Open Dumping
11 jam lalu
Administrative penalties have been issued to 343 final waste processing sites (TPA) across Indonesia that are using open dumping systems.
Indonesian Govt Ramps Up Crab Cultivation to Meet Export Surge
14 jam lalu
The Ministry of Indonesian Maritime Affairs and Fisheries is pushing for the cultivation of crab to accommodate the rising export demand
Kadin Indonesia Warns Panasonic Layoffs May Affect Local Industry
16 jam lalu
Anindya Bakrie stated that layoffs not only occur due to economic factors but also due to technological disruption and policies
Unpad Gears Up with Experts for Phase 3 Trials of TB Vaccine
16 jam lalu
The Faculty of Medicine Universitas Padjajaran (Unpad) is participating in the phase 3 clinical trial of a tuberculosis (TB) vaccine candidate
Indonesia Pushes Green Industry to Boost Global Investment
18 jam lalu
Minister of Industry Agus Gumiwang stated that promoting green industry would attract investors to Indonesia
People's School to Operate in July 2025, Selection of Student Candidates Begins
20 jam lalu
Indonesia's Ministry of Social Affairs will operate people's schools in the new academic year 2025/2026, and the program will begin in July 2025.
ISDS Questions Basis for TNI Assignment at AGO without Actual Military Threat
1 hari lalu
The ISDS is questioning the Indonesian National Armed Forces (TNI) assignment policy in securing the Attorney General's Office (AGO).
Countries Beyond Indonesia Participating in Bill Gates' TB Vaccine Trial
2 hari lalu
Besides Indonesia, the clinical trial of Bill Gates' TB vaccine is also conducted in South Africa, Kenya, Zambia, and Malawi
Pope Leo XIV Begins Papacy with Cardinal Meeting, Raises AI Concerns
2 hari lalu
Pope Leo XIV holds his first meeting with cardinals from all over the world. He touches on the topic of AI
