Case Study: Revolutionising Dialysis, how a NVIDIA AI Model can predict Heart Failure risk with 90% accuracy

Image credit: NVIDIA

Overview

Taipei Veterans General Hospital (TVGH) in Taiwan is deploying real-time AI risk prediction during kidney dialysis procedures to improve outcomes for the country’s nearly 85,000 dialysis patients, aiming to mitigate the risk of heart failure, the leading cause of death among dialysis patients.

Taipei Veterans General Hospital is analysing streaming data during dialysis procedures with the NVIDIA Jetson edge AI platform.

The Challenge

Kidney dialysis patients face the risk of heart failure during the procedure due to complications such as inaccurate dry weight estimation, which can lead to hypertension or dehydration. Monitoring multiple data points during dialysis is crucial to prevent these complications.

The Solution

TVGH developed an AI risk assessment tool that processes various physiological and dialysis machine values, medical records, blood test results, and medication information to predict heart failure risk in real-time during dialysis. The tool utilises AI algorithms and was trained on a dataset of 3 million health records. It leverages NVIDIA Jetson Xavier NX for real-time AI inference and edge computing power, allowing for immediate data analysis during dialysis sessions.

The TVGH team. Image credit: NVIDIA.

Benefits & Results

TVGH’s AI model has achieved 90% accuracy in predicting heart failure risk and has reduced the deviation rate in clinicians’ assessment of dry weight by 80%, minimising the risk of complications for dialysis patients. The adoption of real-time AI inference using NVIDIA Jetson Xavier NX ensures timely detection of abnormal patterns in patient data, potentially saving lives and improving patient outcomes. To read more about the TVGH’s AI model please visit the NVIDIA blog.

 

Testimonial

“A half-hour window between data analysis still left gaps where a patient may begin experiencing complications that can lead to heart failure, so our team worked to find a real-time solution that could receive and compute data generated by dialysis machines within milliseconds.” 

Shou-Ming Ou, visiting staff in the nephrology division at TVGH

This case study first appeared on nvidia.com

#LetsOKdo

About NVIDIA

The AI revolution is transforming nearly every industry. Intelligent autonomous machines powered by NVIDIA are creating new business models, services, and solutions to change the world.

 

ROCK chevron divider grey

Let’s invent the future together

Interested to learn more about the power of IoT and single board computers across a wide range of industries, as well as our OKdo Services and we can help you in your design journey? Talk to us today! Together, we can design a smarter world.

Get in touch today

Like what you read? Why not show your appreciation by giving some love.

From a quick tap to smashing that love button and show how much you enjoyed this project.

Privacy

Our website uses cookies and similar technologies to provide you with a better service while searching or placing an order, for analytical purposes and to personalise our advertising. You can change your cookie settings by reading our cookie policy. Otherwise, we’ll assume you’re OK with our use of cookies.

Location

Please select an option to see content specific to your location and shop online.

Browse the US site