Use of Machine Learning to Predict Medical Device Performance and Failure

Artificial Intelligence (AI) has already changed the way we deliver medicine, drive airplanes, and do our shopping. In this paper, we present our recent case study on the Application of AI to Predict Defibrillators Malfunctioning in Europe.

Limited (and sometimes scarce) budgets & regulations may hinder medical device maintenance, which can result in reduced performance, accuracy and safety, affecting the clinical accuracy and efficiency of diagnosis and treatments. With the recent advancements in medical devices and systems (which learned how to talk each-other), an abundance of real-world data is going to be generated.

This is opening unprecedented scenarios for improving medical device maintenance using AI.

Although there's an increase in technological sophistication of medical devices, incidents involving defibrillator malfunctions are unfortunately not rare. To address this, we have developed an automated system based on machine learning algorithms that can predict the performance of defibrillators and possible failures.

To develop this with high accuracy, an overall dataset containing safety and performance measurement data was acquired from periodical safety and performance inspections of 1,221 defibrillators.

These inspections were carried out in the period 2015–2017 in private and public health-care institutions in Bosnia and Herzegovina by an ISO 17,020 accredited laboratory. Out of the overall number of samples, 974 of them were used during the system development, and 247 samples were used for subsequent validation of the system's performance. During this development, 5 different machine learning algorithms were used and resulting systems were compared by obtained performance.

The results of this study demonstrate that Clinical Engineering and Health Technology Management benefit from the application of machine learning in terms of cost optimization and medical device management.

Automated systems based on machine learning algorithms can predict defibrillator performance with high accuracy. The adoption of these systems will help to overcome budget challenges, adapting maintenance and medical device supervision mechanism protocols to the rapid technological development of these devices.

Due to the increased complexity of the healthcare institution environment and increased technological complexity of medical devices, traditionally performing maintenance strategies are causing a lot of difficulties. For this same reason, we must evolve the same way that technology does; quickly and safely.

You can download the full article here:

Content provided by CED Members Leandro Pecchia & Almir Badnjevic.

Want to receive more CE & HTM news, events, and materials? Join our newsletter by clicking here. Help us promote our Worldwide CE & HTM Efforts by sharing this content in your social media. Together we can make it better!

Print   Email