Go to contents

THE DONG-A ILBO Logo

Open Menu Close Menu Open Search Bar
검색창 닫기

Medical / Digital Healthcare

Aidot Cure: AI Heart-to-Voice Detects Heart Disease

Dong-A Ilbo | Updated 2025.09.23
[SeoulTech x Dong-A.com Joint Planning] Seoul National University of Science and Technology (hereinafter referred to as SeoulTech) operates various support programs to aid the growth of startups, including preliminary and early startup packages, maker spaces, and global collaboration. Furthermore, in collaboration with Dong-A.com, it provides global news to assist startups in international promotion and expansion, introducing promising deep tech startups domestically and internationally.

Heart failure is a condition where the heart, which acts as the body's pump, fails to function properly. Individuals with chronic diseases such as hypertension, diabetes, and obesity are at potential risk for heart failure. The issue is that early symptoms are vague or mild, making early detection difficult.

Jung Kyung-ho, CEO of A.Cure / Source=IT Dong-A

According to the latest '2025 Heart Failure Fact Sheet' released by the Korean Society of Heart Failure, approximately 1.75 million people in Korea are estimated to have heart failure. Data from the Korean Society of Heart Failure and the Health Insurance Review & Assessment Service (2020 Fact Sheet) indicate that the one-year recurrence rate for heart failure patients in Korea is 50%, and the five-year mortality rate is also 50%, highlighting the urgent need for management. However, existing diagnostic methods are costly due to the comprehensive nature of the tests, and early diagnosis is challenging, making effective management difficult. The annual medical cost per person reaches KRW 8.5 million. A company has emerged to address this issue using artificial intelligence (AI) technology and voice. The digital healthcare startup A.Cure Co., Ltd.

A.Cure offers an AI voice analysis-based heart failure detection and severity prediction solution called 'Heart to Voice.' This solution non-invasively determines the presence and risk of heart failure by analyzing the patient's voice. It simplifies complex examination procedures to assist in patient health management and captures abnormal signs early for proactive response. This year, the company aims to enhance the severity analysis function to provide personalized management, such as medication guidance and exercise timing.

Based on Experience with the First Digital Therapeutic Device

Jung Kyung-ho, CEO of A.Cure, is an expert who has overseen the entire process from development to commercialization of digital therapeutic devices. After his co-founder, Professor Kim Eung-joo of Korea University, received an award for AI model research on heart failure status at the American College of Cardiology in 2023, he was invited to start a business together. Recognizing the global value, he was convinced that commercialization could not be delayed and officially launched the business in 2024.

CEO Jung stated, “The early symptoms of heart failure, such as shortness of breath or fatigue, are similar to non-specific symptoms, making them difficult to discern. Due to the incurable nature of the condition, if not consistently managed, the one-year recurrence rate exceeds 50%,” adding, “Heart to Voice helps to detect and manage abnormal signs early through subtle voice changes before the patient reaches an emergency situation, supplementing the limitations of the existing medical system. It aims to prevent emergency hospitalizations and surgeries that incur significant medical expenses and ultimately contribute to increasing survival rates.”

Voice-Based Heart Failure Detection Heart to Voice

AI voice analysis-based heart failure detection and severity prediction solution ‘Heart to Voice’ / Source=A.Cure

The core of Heart to Voice is heart failure diagnosis through voice analysis. When heart failure occurs, pulmonary congestion causes subtle changes in the voice, which AI detects to determine the presence and severity of heart failure. CEO Jung explained, “By having patients pronounce specific sentences like ‘Long live Korea,’ heart failure can be determined. The 20-second voice is divided into one-second segments and further split into 100 parts to extract features,” adding, “This year, we are enhancing the system to allow detection through natural conversation.”

Additionally, A.Cure is refining the assessment of heart failure severity to provide patient-tailored solutions. Recognizing the limitations of voice alone, they have combined nine clinical indicators of heart failure, such as blood tests (NT-proBNP), ultrasound, and electrocardiograms, with voice data. CEO Jung explained, “Heart failure is not a condition that can be clearly judged by numbers like blood pressure; it requires comprehensive tests for doctors to determine, making AI development challenging.”

A.Cure's voice AI analysis technology image / Source=A.Cure

A.Cure plans to use unsupervised learning (a method of clustering data with similar characteristics without answer labels to predict outcomes for new data) to train complex clinical indicators and classify them into four stages: ‘normal,’ ‘mild,’ ‘moderate,’ and ‘severe.’ CEO Jung stated, “We are conducting clinical trials with 124 participants across four institutions, aiming to increase the existing 90.4% accuracy to 95% this year,” adding, “Ultimately, the goal is to achieve results similar to a detailed heart failure examination using only voice.”

Targeting Medical and Non-Medical Two-Track Market

A.Cure has adopted a two-track strategy, simultaneously pursuing medical and non-medical fields. The medical field is divided into three categories: ▲hospital examination equipment ▲patient personal monitoring devices ▲digital convergence pharmaceuticals linked with medication. Each will undergo clinical trials and regulatory approval from the Ministry of Food and Drug Safety, with plans to release examination equipment and personal monitoring apps for use in primary medical institutions and nursing facilities. In the future, they aim to develop digital convergence pharmaceuticals that guide medication dosage and timing based on severity to assist in patient management.

CEO Jung stated, “The medical field requires approval from the Ministry of Food and Drug Safety, which takes time, but it allows for the provision of reliable, evidence-based services. We plan to start clinical trials for approval from the Ministry of Food and Drug Safety next year, following a phased procedure,” emphasizing, “By guiding the administration and dosage of diuretics according to severity, we aim to prevent unnecessary administration and ensure optimal medication for the patient's condition.”

A.Cure app image / Source=A.Cure

The non-medical field focuses on wellness services that detect and manage abnormal signs. They plan to offer differentiated health management services by integrating AI speakers and wearable devices in senior residences, nursing hospitals, health centers, etc. CEO Jung explained, “When users naturally converse with an AI speaker or watch, voice data is collected and analyzed. It will be a premium service that prevents emergencies.” To this end, A.Cure plans to conduct technology verification at facilities operated by Lotte Hotel.

They are also preparing to enter the mobility sector. The technology will be used to manage driver safety in vehicles, with functions such as heart failure detection and alcohol measurement. Ultimately, it will be integrated into smartphones or smartwatches. CEO Jung stated, “We will continuously enhance the AI model by persistently acquiring voice data from high-risk heart failure groups in senior residences and health centers with user consent,” adding, “The high-quality data thus obtained will create a virtuous cycle that solidifies accurate grounds for disease detection.”

Expanding into Emergency Disease Field… Full-Scale Overseas Expansion

Jung Kyung-ho, CEO of A.Cure / Source=IT Dong-A

In 2024, A.Cure transferred technology from the Korea University Medical Center Industry-Academic Cooperation Foundation and established a faculty startup, achieving success in investment rounds while carrying out government projects. A.Cure plans to enhance its self-developed AI-based voice analysis technology within the year and continuously improve AI model performance by incorporating medical data and research results from Korea University Medical Center. The goal is to complete domestic confirmatory clinical trials and obtain regulatory approval from the Ministry of Food and Drug Safety by next year.

Simultaneously, they are preparing for global expansion, focusing on the United States, Japan, Singapore, and China. Being selected as a collaboration company for Oracle Cloud Infrastructure (OCI) has been beneficial in this regard. CEO Jung stated, “Through the Miracle Program, we established an environment for algorithm development and service operation. Additionally, we were able to meet overseas investors and establish networks for potential collaboration in clinical trials or distribution.”

The ultimate goal of A.Cure is to build a platform for predicting emergency diseases using voice, beyond just a heart failure solution. After establishing successful cases and securing trust in the technology, they plan to expand the solution to other conditions where emergency situations are critical, such as asthma. In particular, they are already participating in national projects with Seoul National University and Yonsei University in the field of developmental disabilities, researching early detection of autism spectrum disorders through changes in children's voices and behaviors.

CEO Jung acknowledged the challenges due to the lack of many precedents in the digital medical field but expressed the desire to present a successful model in the healthcare market. He stated, “A.Cure aims to positively influence latecomers and lead the overall growth of the digital medical industry.”

IT Dong-A Reporter Kim Ye-ji (yj@itdonga.com)
AI-translated with ChatGPT. Provided as is; original Korean text prevails.
Popular News