Doctor Develops AI-Based Tool for Early Detection of Heart Failure Risk
Heart failure remains a serious issue in healthcare services in Indonesia. Data from the Asian-HF Registry shows Indonesia ranks second with the highest number of heart failure cases in Asia after China. The one-year mortality rate for patients reaches 34.1 per cent, while around 30 per cent of patients must return to hospital for treatment due to worsening conditions after being discharged. One factor triggering the high readmission rate is the presence of residual pulmonary congestion, or fluid accumulation in the lungs, which goes undetected before patients leave the hospital. This condition is often missed during examinations using a conventional stethoscope. Meanwhile, methods such as Lung Ultrasound and NT-proBNP blood biomarker tests require special equipment, higher costs, and medical personnel with specific competencies.
Addressing these challenges, Dr Rony M. Santoso, a Cardiovascular Interventional Consultant and Vascular Medicine Consultant at Primaya Hospital Tangerang, developed the Novel Auscultation Device of Artificial Intelligence for Heart Failure (NAVI-HF). This artificial intelligence-based innovation is part of his doctoral dissertation research at the Faculty of Medicine, University of Indonesia. NAVI-HF is designed as a device to assist doctors in detecting signs of pulmonary congestion through a more practical, rapid, and objective analysis of chest cavity sounds. Unlike a regular stethoscope, this device records chest sounds from five examination points for approximately one minute. The recording is then analysed using an AI algorithm to identify whether the patient still has signs of pulmonary congestion that could potentially cause worsening heart failure after hospital discharge.
Research results on 246 acute heart failure patients showed NAVI-HF has promising diagnostic performance. Compared to Lung Ultrasound as the reference standard, the device recorded an accuracy rate of 86 per cent, sensitivity of 91 per cent, and specificity of 82 per cent. Additionally, a follow-up study conducted over six months found that patients with a positive NAVI-HF result had a 1.6 times higher risk of being readmitted due to heart failure compared to patients with a negative result. According to Dr Rony, this innovation is not intended to replace the role of doctors but rather to serve as a supporting tool in identifying high-risk patients so that treatment can be administered more quickly.
"One of the biggest challenges in managing heart failure is ensuring the patient’s condition is truly stable before leaving the hospital. We developed NAVI-HF to help doctors identify patients who are still at risk of deterioration through a simple, portable, AI-supported tool. This way, patients who require closer monitoring can be recognised earlier so that therapy can be adjusted before complications occur," he explained. He added that the development of NAVI-HF also opens opportunities for implementing home-based monitoring and telemedicine services in the future. "We hope this innovation can support earlier detection, assist doctors in clinical decision-making, and simultaneously reduce the risk of readmission due to heart failure," he added. The development of NAVI-HF represents an example of the use of artificial intelligence technology in the medical world. The presence of this technology is expected to help healthcare workers produce faster, more objective, and more targeted diagnoses without replacing the role of doctors, while improving the quality of life for heart failure patients and reducing the burden on healthcare services in Indonesia.