Q&A: How AI is used to identify potential lung or heart problems

Web content editor Ian Bolland caught up with Wojciech Radomski, CEO of StethoMe, a startup that creates solutions in telemedicine. Its main product is a wireless stethoscope which works with an app using AI algorithms, which is available to consumers later this year.

Where did the idea for StethoMe come from?

The idea comes from the co-founders of StethoMe – Honorata Hafke-Dysand Jędrzej Kociński. Both of them are scientists, but also parents who struggled with frequent respiratory tract infections in their children. After going through this experience, Honorata and Jędrzej discussed the idea of a home stethoscope that could potentially change the lives of many families around the world.

The implementation of this idea would not have been possible if it had not been for Paweł Elbanowski and Marcin Szajek, along with myself. We have been running a service providing IT company for over 10 years.

Three years ago all five of us met and decided to commercialise auscultation sound studies that were already being conducted at that time by Honorata Hafke-Dys and Jędrzej Kociński.

These studies found that there are two main issues facing primary care. Firstly, the high rate of misdiagnosis that occurs (cases of delayed, missed, and incorrect diagnosis are common – in the range of 10% to 20%) and, secondly, the very high number of unnecessary visits (up to 70% in the US). This is why we created StethoMe – to help address these challenges.

What does StethoMe consist of?

StethoMe is a wireless, electronic stethoscope which works with a dedicated app and StethoMe AI. StethoMe AI is the world’s first certified algorithms intended for analysis and detection of abnormal sounds in the respiratory system of children. We are also working on the AI version for adults and elderly people.

It is enough to apply the device at places indicated on the phone screen and StethoMe will do the rest. StethoMe will inform of any potentially abnormal sounds and allow you to send results to the doctor for remote consultation. Results are shared directly with a doctor who then advises as to whether that patient needs to visit a doctor or emergency room, or can remain at home under close supervision. The result: peace of mind for families and a reduced burden on healthcare professionals by reducing the amount of unnecessary GP or hospital visits.

Give us an idea of the technology that went into creating StethoMe, and subsequently operates it?

StethoMe gives a possibility of full remote consultation for adults and children – anytime, anywhere. You can use telemedicine service for diagnosing respiratory problems and for paediatric consultations – the most common reasons for Primary Care visits.

Our algorithms are based on AI and proprietary and growing data of over 38,000 medical descriptions of auscultation sounds, comprising of over 1,000,000 detailed sound tags. The algorithms were created by a group of AI best specialist cooperating with medical centres. The certification StethoMe has received is the world’s first, which is hugely significant. The consistency between doctors in the recognition of auscultation sounds in primary care is very low – ranging from 20% to 80%*. Examinations with a traditional stethoscope depend on the experience and parameters of the physician’s auditory system. Medically tested, StethoMe has shown to increase accuracy by up to 13%. An improved examination and the use of AI means parents can examine their children safe in the knowledge that doctors can make informed decisions based on far superior readings.

*Weisse AB. An underappreciated problem with auscultation. Proc (Bayl Univ Med Cent). 2013;26(2):191–192

The product seems to be aimed at children, is it also aimed for adults to keep track of their own health too?

The StethoMe stethoscope gives a possibility of examination adults and children – anytime, anywhere. It enables auscultation lungs and heart young and adult patients. It also provides information automatically about respiratory and heart rate.

Our AI algorithms were created especially for children. They detect pathological sounds in the respiratory system of children. We’re also collecting data for adults and seniors.

What’s next for StethoMe now it has received CE certification?

Currently, we are developing AI algorithms for the respiratory system for adults. We are also working on AI algorithms for the heart.

We have chosen the KCRI company to be our partner to this end. Within the framework of collaboration, KCRI will be responsible for diagnostic evaluation of our algorithms in three research centres, including the USA and Europe. The studies conducted will enable future certification of StethoMe at the American Food and Drug Administration.

We’re also in the process of FDA approvals.

Anything else to add…?

What is unique about StethoMe is that it’s not just a team of scientists or inventors who decided to create a product like this. Most of the team are parents of young children who have faced challenging times where we’ve needed to visit doctors regularly.

There’s no doubt that AI technology is becoming increasingly prominent within healthcare. It’s the future, and we are excited to be at the forefront of this shift. When you combine this with telemedicine and how its being adopted across the globe to ease the pressure faced by healthcare professionals, we believe they have created something very special.




'Q&A: How AI is used to identify potential lung or heart problems' has no comments

Be the first to comment this post!

Would you like to share your thoughts?

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

© 2019 Rapid Life Sciences Ltd, a Rapid News Group Company. All Rights Reserved.

Privacy policy

Terms and conditions