Kreatik talks – Wojciech Radomski about artificial intelligence in medicine

Maciej, UX Designer at Kreatik talks to Wojciech Radomski – CEO and Founder of StethoMe, about the impact of technology on medicine, changes in diagnosis, and the future of the MedTech industry.

I would like to thank you for finding a moment for us. It is a great pleasure for me. Please, tell us who you are, what you do, and what StethoMe is.

– My name is Wojciech Radomski, I have been the president and co-founder of StethoMe for seven years. Previously, for 10 years, my partners and I ran a software house. Then we sold it and started our adventure with StethoMe. Honorata and Jędrzej – PhDs from the Adam Mickiewicz University joined us in this adventure. They came to us with the idea of ​​the electronic stethoscope.

StethoMe is an electronic stethoscope for home use with artificial intelligence algorithms to detect abnormal sounds in the respiratory system. While more and more devices such as electronic stethoscopes appear on the market, we are still the only company in the world that has medical certification for artificial intelligence algorithms that can analyze these sounds. There is a big difference between what is available on the market and StethoMe. 

With a typical electronic stethoscope, the sound is recorded at home by the user and must be sent to the doctor to evaluate the sound. They work like ordinary microphones – we turn on the recording, turn off the recording, and send the sound to the doctor. At the same time, these sounds are very subtle (because we try to record something from the inside of the body, something that we cannot hear with the proverbial naked ear). The same recording process turns out to be very complicated in terms of the correctness of the sound capturing those specific sounds, and not, for example, our dog’s barking or the sound of a humming refrigerator in the kitchen. With solutions that only have the function of recording, these sounds are simply of poor quality and with a large amount of distortion, so when it comes to the doctor, he is not able to assess it correctly. 

Algorithms ensure the sound quality of the recording, which is then analyzed for abnormalities in the lungs – detecting and classifying them. Later the patient can send the recording to the doctor. Algorithms themselves are certified medical devices. We have proved in many of our scientific publications that this solution is more effective than such a statistical family doctor. Therefore, it is a tool with great potential for all of us. As a small startup with a limited budget, we had to focus on one thing, so we chose the target group of children with asthma. Every tenth child in the world suffers from asthma and half of these children have so-called uncontrolled asthma, which means that they have asthma attacks several times a week. Thanks to StethoMe, we can monitor the disease much better, because the parent, based on the results, can determine what steps to take next. Should I give the child medicine, should I see a doctor, or should I give inhalations?

It is the answer to all these questions that StethoMe helps, giving precise information on what is currently happening in the child’s lungs. Here it should be noted that this variability of the disease is very large and very often we meet with stories that someone has visited a doctor and he said everything was fine with the child and the next day he went to another doctor who said it was pneumonia. The most common reason for different diagnoses by these doctors may be the very dynamic variability of the disease in the lungs. In fact, within an hour, the lungs can change from normal to very bad. Therefore, it is important to perform an auscultation examination frequently and monitor changing conditions.

And here is another advantage of StethoMe – tests can be carried out more often. We do not, for example, test a child with a thermometer only once a day, as if a doctor did it with a traditional stethoscope, but usually, we do it several times a day to see how this temperature changes, and here it is very similar. We can measure very quickly and precisely and thus we can see how the disease is changing. We can react appropriately.

Want to know how to design healthcare applications? Check out our tips!

Please tell us something more about the very beginnings. Where did the idea for this product come from, and how long did it take to develop? What was the biggest challenge you faced during this whole process?

– Where did the idea come from? After 10 years of running an IT company, one day I came to the office and suddenly found that I no longer wanted to do these boring IT systems for another company where each of them is very similar to the other. And it would be nice to use these resources, vast knowledge, and experience to do something that will change something in the world. I returned to the subject of my studies, “Computer Science in Medicine”. We had a lovely teacher there. The final task was to find a doctor with a problem and solve it using technology. I remember today the moment of passing this course when all the teams presented their projects in the laboratory room. I was captivated by the enormity of these interesting solutions, which at that time were such amazing innovations and had great potential to support medicine. The reality was that after passing, they were hidden in a drawer and nothing happened to them afterward. It stuck in my mind that there is enormous potential for combining computer science with medicine. This is how the idea came to my mind: maybe this is the direction that will help us change something in the world. I wrote a short post on Facebook that we are an IT company, we have experience, we have capital, and connections and we are looking for a doctor who has a problem that we will be able to solve using new technology. Well, this post was answered by Honorata and Jędrzej. Indeed, they were not doctors, but PhDs of acoustics, and also they are both parents of young children who face lung problems.

As PhDs of acoustics, they have repeatedly wondered what the doctor hears in these headphones. When they researched the topic, it turned out that the stethoscope as a medical device since its production 200 years ago wasn’t improved. It turned out that stethoscope study is very subjective and there are many mistakes.

We did short research, and saw that a device like StethoMe still doesn’t exist. It immediately grabbed our hearts, and this is how StethoMe was created. We established the company in May 2015 and started looking for financing. Back then, everyone was still in their “original” work, and after hours we started looking for funding and focused on the EU project. We have tried twice with the EU project and have not been able to get the money twice. The moment has come that we decided to look for an investor. Thanks to our contacts with previous projects, we managed to find an investor very quickly and very efficiently in the investment process, and thus we obtained the first 1.5 million PLN to start the project. On January 1, 2017, we started full-time work on this project. Honorata and Jędrzej, still working at the university, brought in their best students, whom we hired. And this is how the adventure with StethoMe began.

I am also interested in your application, which plays a communication role between the stethoscope and the human. Were you somehow involved in the process of creating this application? Are you going to expand any more with functionalities?

– Of course, we are involved in the application development process from day one until today. The application is alive all the time and a lot of changes in the application all the time. However, we must remember that we are a certified medical product and the application development process is different from traditional. Making an application is simple, but later certifying that application is very difficult. Sometimes even a change in the text or color requires certification. Depending on the shift certification can take up to a year, two, or three. Therefore, it requires a thoughtful internal process that will allow the application to develop, but also be safe for the user and legal. 

We devoted a lot of time to user experience because of a mix of software and hardware. The key task was that the user, who even has no clue what a stethoscope is, performs the test properly. There were such trivial problems that should not appear at all, e.g. the stethoscope has a screen at the top and a membrane at the bottom, so it seems obvious how we apply it to the body. There have been cases of parents applying the stethoscope with the screen to the child’s body instead of the membrane. In our case, the combination of user experience and hardware is crucial to use the device properly and thus maintain the high quality of the examination.

I would like to ask about the pandemic. How do you think the pandemic affected the medical industry? Was your stethoscope used during the pandemic times? What did it look like?

– It’s hard for me to answer this question because, on the one hand, it had a very positive impact on the development of MedTech, but above all in the acceptance of MedTech by doctors, service providers, medical units, and governments. The pandemic accelerated the development of MedTech for a good few years. We were forced to implement various types of IT solutions much faster. Thanks to the pandemic, many doctors who were not open to new solutions eventually had to open up to them.

There is also a negative side – everything has been closed. We had fairs scheduled, but they didn’t take place. Not a single meeting could happen. Everything went online. Communication with potential clients has been very difficult. Even if someone was already interested in it, it wasn’t possible to make a traditional demo, because everyone was afraid. It was a challenge.

The StethoMe helped the pandemic in many different ways. First of all, when the pandemic began, the value of StethoMe was noticed by doctors who worked in COVID hospitals. Due to the exceptional sound quality of our solution and technical parameters, it turned out to be very useful in monitoring patients in hospitals, when doctors had to be armed with all protective suits that also covered their ears.

There were situations where doctors shared tips on how to deal with a pandemic in a hospital using StethoMe, among others thanks to Bluetooth technology and the fact that this sound can be transmitted over distance using Bluetooth headphones. The StethoMe works with doctors, ambulances, and some patients until today. 

The pandemic also helped in our largest project so far – the Ministry of Health became interested in our product and bought a thousand stethoscopes for post-COVID patients.

The e-stethoscope is about to be reimbursed for primary care patients with chronic conditions. We believe that very soon e-stethoscope will become the crucial element of home care monitoring for the Polish population. Thanks to artificial intelligence algorithms, certified as medical devices and scientifically proven that they are more effective in detecting abnormalities in the respiratory system we don’t need to involve doctors in all examinations, and only refer patients who really require it to the doctor.

More talks with inspiring people: Kreatik talks – Rudradeb Mitra about why community-driven startups are the future of business.

Do you think that digitalization is a trend that can no longer be stopped or is not everyone open to it, though?

– This, of course, depends on the area of ​​MedTech. Even before the pandemic, digitization was progressing, only we often don’t see many of these changes for example, progress in MRI, which is a typical MedTech. What we have observed as a patient in the pandemic is the end of some services that affect us i.e. direct contact with the doctor. Will digitalization continue? I think that it cannot be stopped, because everyone saw the enormous benefits. At the same time the pandemic also showed a lot of risks related to the digitization of MedTech.

The pandemic caused us to implement many solutions without thinking about the consequences, or the specific steps. A pandemic in Poland caused a family doctor who previously did not perform any services remotely to have to diagnose patients over the phone. It turned out to be very dangerous because doctors had no tools whatsoever to assess this patient’s condition. It is impossible to diagnose a child over the phone without listening to what is happening in the lungs. This movement resulted in more cases in hospitals because the doctors did not diagnose well enough. It was this quick, mindless, risky digitization.

I would say that nowadays telemedicine is ideal for filtering out patients who need help from those who do not need it. Even 80% of the cases of patients in the medical office are unnecessary cases. These are the situations in which the patient has doubts, but in fact, he is fine. We can filter out most of them through technology, and then the doctor has 10 times more time for a single patient who needs help. Today the doctor statistically has 7 minutes for the entire visit, whereas they should spend at least 5 minutes auscultating the patient. Thanks to telemedicine, patients and doctors have the possibility of eliminating unnecessary medical visits and speeding up those that are necessary.

I think that we are not yet able to replace some activities with MedTech. What do you think about it? Are there any activities that will stay with us as they are now?

– The diagnosis process is something we will not be able to replace with any technological solutions soon. Today, artificial intelligence has an intelligence comparable to ant intelligence that is very far from human, but note that the ant does one thing and does that one thing well and extremely precisely. A man would not be able to listen to as many recordings throughout his life as our artificial intelligence. Consequently, in the narrow area of ​​analysis of the lungs sounds, AI will be better than the doctor. But we have to look at the patient holistically, which is why we say that StethoMe does not diagnose the disease, and does not say that the patient has pneumonia or bronchitis. This topic is much wider, and here the broad knowledge of the doctor, the doctor’s experience is essential. The human mind, at least today copes with these diagnoses much better than artificial intelligence. Of course, there are systems like Infermedica, which is an excellent diagnosis support, but still it is supporting the diagnosis, not inventing the whole treatment process. Currently, I would leave the treatment of already sick patients to doctors, and AI should support diagnosis.

Check out our the best MedTech mobile apps!

How do you feel about self-control of health? Isn’t it a significant shift of responsibility to the user’s health? Maybe we’ve gotten up to speed with these innovations to the point where more and more of these technologies make people bypass doctors altogether.

– You’re asking a question about a very wide area. I would not like to generalize. If we take what we talked about earlier, i.e. a consultation over the phone with no additional data – I am a huge opponent of such a solution. If we take a patient who looks for his symptoms in Google and wants to be treated based on it – we should be as far away from such solutions as possible. If we are talking about a whole bunch of medical applications that are not certified we should not go in that direction either. However, if we are talking about certified medical devices that have passed the appropriate certification, the relevant tests, and procedures, and have the appropriate instructions for use, and assuming that patients use these instructions, these are the solutions that support the whole process of treatment. So it depends on the specific case. We need to carry out this digitization of healthcare in a very sensible way, based on data, based on the fact that the solution will be better for the patient and generate less risk.

We talked about lung variability. Tell me what will be better for the patient. That he would go to a doctor once a day and that doctor would listen to this patient once or will the patient be able to check his lungs 10 times a day? Now the following steps are essential because the fact that he can check 10 times a day does not mean that he has to make the decision himself. These solutions must support us and not replace various types of medical procedures.

I would like to give you a quick example. Imagine a diabetic who needs to test his insulin levels. He has an app for that and the app tells him when to inject the insulin. This is an excellent example where we eliminate the doctor 100% so that the patient becomes fully self-sufficient. This a great case that shows full digitization as a benefit for the patient, but not all cases are as simple as this. Therefore, even in the case of lungs, when something happens, we send such a patient to a doctor. We are somewhere in the middle and do not throw all responsibility on the patient.

So there are solutions in which the patient is self-sufficient, there are solutions that support, and there are also medical procedures and solutions that will require a visit to the doctor each time.

Imagine waking up in ten years. What do you think the world will be like? You can also share such a vision about the future of healthcare.

– Contrary to appearances, 10 years is not a lot of time. StethoMe was established 7 years ago, and a lot has not changed in the Polish health system during this time. I also think that what will happen in 10 years are those solutions that today are considered innovations (for example StethoMe) but which are still very little popularized and will simply be more widely available and more often used by users. I wouldn’t expect it to be a big revolution in 10 years. Simply the saturation of the market with new technologies will be much greater, which in turn will make this medical care definitely improve.

Thank you very much for this interview. It is a pleasure to listen to you. Congratulations and thank you very much.

– Thanks for the nice words. All the best!

Subscribe to our newsletter and stay up to date with MedTech knowledge!

Wojciech Radomski

Strategy and business project leader with more than 15 years of business-IT experience. Innovator, visioner, successfully launched and managed Software House and currently, Wojciech is focused on developing StethoMe®, which is about to revolutionize healthcare.

He works great in a team, the idea which he is convinced about becomes his passion, which he is fully devoted to. When it comes to the quality of products, he does not like compromises. He is interested in things, which are crucial socially and advanced technologically.

Privately, he likes challenges that build his character - Vice-champion of Polish ATV Cup, marathoner, triathlete, Enduro rider. He sees potential in people, and he knows how to develop it.


Maciej Marczak

UX Designer

See also
Frequent e-commerce mistakes – in developing strategy

Do you manage an online shop? Or are you planning to start selling your services online? What are the challenges you may face as an online shop founder? What can go wrong, what are the biggest risks and what is […]

Understanding Digital Health – what’s included in this term?

The pandemic has caused technology and medical experts to take a curious look at the Med-tech market. Remote medical services or diagnostics have proven to be not only necessary but also a driving force for the further development of this […]

Kreatik talks – Wojciech Radomski about artificial intelligence in medicine

Maciej, UX Designer at Kreatik talks to Wojciech Radomski – CEO and Founder of StethoMe, about the impact of technology on medicine, changes in diagnosis, and the future of the MedTech industry. I would like to thank you for finding […]

Price your project
en pl