Students create affordable AI stethoscope to help monitor lung disease at home.

AI - Ảnh 1. — A group of students are experimenting with AI-integrated smart headsets for monitoring - Photo: CHAU SA

This AI-powered stethoscope assists doctors in diagnosing lung diseases accurately and quickly. Developed by a group of students from Da Nang University, the product has received positive reviews for its humanitarian aspect and practical applications.

This project was awarded second prize in the 5th Student Technology Startup Competition - InTE-UD 2025, organized by the Youth Union and Student Association of Da Nang University in collaboration with the Department of Science and Technology.

We wanted to create a compact and smart device so that patients could more easily monitor their lung health. At the same time, it would assist doctors in making accurate diagnoses, saving time on examinations and treatment.

Student VO HOANG

Diagnosis of respiratory diseases

The authors include students Vo Hoang, Thien Quoc, Quoc Hung, and Hoang Long from the University of Technology (Da Nang University) and Van Anh from the University of Economics , Da Nang University. The project, named RespirAI – a smart medical device integrating AI – was developed by the research team in September of this year with the ambition to "help doctors not only hear but also see and understand lung sounds more clearly through visual data analysis."

Team leader Vo Hoang said RespirAI stems from the reality that Vietnam has a high rate of respiratory diseases, especially chronic obstructive pulmonary disease (COPD) and asthma in children. Many people in rural and remote areas have difficulty accessing regular medical facilities, while lung diseases develop silently and can easily become severe if not monitored.

RespirAI has a compact design similar to a traditional stethoscope but is equipped with sensors that capture sound at locations such as the apex, mid-lung, and hypopulmonary regions.

Student Thien Quoc shared that his stethoscope stands out thanks to AI that digitizes captured lung sounds, automatically filters out noise, performs real-time analysis, and identifies abnormalities through a trained AI model.

The results after processing will appear in the form of a visual chart including respiratory rate graphs and percentage predictions of symptoms (wheeze, crackles, normal) or diagnoses of diseases such as COPD, pneumonia. This will serve as reference data to help doctors make diagnoses.

"The device has the capability to support remote monitoring of disease progression and diagnosis via IoT, helping to reduce costs and travel time for patients," Quoc said.

The arduous research journey

Compared to some existing products on the market, RespirAI has the advantage of automated analysis using AI and clear information display through software. In addition, it has a lower cost, making it suitable for primary healthcare settings. Furthermore, it opens up the possibility of creating electronic medical records and tracking diseases through each stage.

The team not only programmed and designed in the lab, but also directly surveyed hospital needs to ensure the product was more suitable for real-world treatment. However, the research and development of this device was far from simple, with the biggest challenge being the lack of real lung sound data to train the AI.

Vo Hoang stated that recording could not be done arbitrarily due to confidentiality regulations and the need for professional assistance from doctors. Not to mention, obtaining permission and collecting data takes a significant amount of time.

"Students doing research have limited experience and funding. Many experiments failed, requiring them to redo the hardware, and the software malfunctioned midway, forcing them to reprocess the data from scratch," Hoang recounted.

Nevertheless, the students were determined to work on the project, and with the enthusiastic support of their teachers, they gradually perfected the product. Currently, RespirAI has a working prototype with functions for listening, noise filtering, and AI-based lung sound analysis, achieving relatively high accuracy. The team proudly stated that they are continuing to improve the hardware and optimize the AI model to increase accuracy.

The team expects that by the end of 2026, the product will be ready for real-world testing in hospitals and clinics. If all goes well, the application for licensing will be completed by the end of 2027, and it will be widely commercialized with an expected selling price of over 9 million VND per device.

In line with the trend of digital healthcare and smart medical examination and treatment.

Master's degree holder Nguyen Quang Tan - a lecturer in the Electrical Engineering Department of the University of Technology (Da Nang University), and the project's direct supervisor - assessed RespirAI as having practical significance in the context of increasingly common respiratory diseases, especially post-COVID-19. He said the device, which uses smart sensors, can greatly assist doctors in remote examinations and treatments.

Beyond the AI stethoscope, the team could expand the features to include more information such as measuring respiratory parameters and airflow, to comprehensively assess the user's lung condition. Most importantly, they need to build a standardized database, creating a foundation for future remote disease management and monitoring.

"This is a direction that aligns with current trends in digital healthcare, and the team needs to continue collaborating with healthcare facilities and specialists to build a more accurate dataset for the AI algorithm," said Master Tan.

VERMILLION

Source: https://tuoitre.vn/sinh-vien-lam-ong-nghe-ai-gia-mem-giup-theo-doi-benh-phoi-tai-nha-20251212092620939.htm