Main Highlights
- Cardiomatics has announced a $3.2 million seed round to broaden the usage of its electrocardiogram (ECG) reading automation technology.
- A $1 million non-equity grant from the Polish National Centre of Research and Development is also included in the seed round.
- Cardiomatics advertises their technology as aiding in the democratization of healthcare access, claiming that it allows cardiologists to optimize their workflow.
- According to the firm, the AI tool has analyzed more than 3 million hours of ECG data commercially so far.
Cardiomatics, a Poland-based healthtech AI firm, has announced a $3.2 million seed round to broaden the usage of its electrocardiogram (ECG) reading automation technology. Kaya, a Central and Eastern European venture capital firm, led the round, with Nina Capital, Nova Capital, and Innovation Nest also participating. A $1 million non-equity grant from the Polish National Centre of Research and Development is also included in the seed round.
The 2017-founded startup sells a cloud tool to help cardiologists, clinicians, and other healthcare professionals interpret ECGs by automating the detection and analysis of some 20 heart abnormalities and disorders, with the software generating reports on scans in minutes, faster than a trained human specialist could.
Cardiomatics advertises their technology as aiding in the democratization of healthcare access, claiming that it allows cardiologists to optimize their workflow so that they may see and treat more patients. It also claims that it enables GPs and smaller practices to provide ECG analysis to patients without referring them to specialized facilities.
According to the firm, the AI tool has analyzed more than 3 million hours of ECG data commercially so far. Its product is utilized by over 700 clients in 10+ countries, including Switzerland, Denmark, Germany, and Poland.
What does Cardiomatics promote?
At this point, the program can interact with more than 25 ECG monitoring devices, and it promotes providing a contemporary cloud computing interface as a difference from older medical software. Cardiomatics is a cloud AI tool for ECG analysis. It simply gives Them a raw ECG signal and their algorithms will turn it into the value analysis.
Cardiomatics uses ninety percent of the data as a training set, while the remaining ten percent is used for algorithm validation and testing. According to data-centric AI, they place a high value on test sets in order to ensure that they include the greatest possible representation of signals from their clients.
They assess the correctness of the algorithms in experimental work once a month throughout the ongoing development of both algorithms and data. In clinical practice, their customers verify it on a daily basis. This is how Cardiomatics ensures their quality.
Cardiomatics stated that the seed financing would be used to invest in product development, grow its commercial activities in existing areas, and prepare to enter new markets. The proceeds of the investment will be utilized to fund their rapid development ambitions across Europe, including scaling up their market-leading AI technology and ensuring physicians enjoy the greatest experience possible.
They are also preparing the product for expansion into other markets. Their future goals also include gaining FDA approval and accessing the US market. The AI tool received European medical device certification in 2018 — but it’s worth noting that the European Union’s regulatory regime for medical devices and AI is still evolving, with an update to the bloc’s Medical Devices Directive (now known as the EU Medical Device Regulation) going into effect earlier this year (May).
A new risk-based framework for AI applications, known as the Artificial Intelligence Act, is also on the way, and it will most certainly extend compliance standards on AI health tech products like Cardiomatics, requiring them to demonstrate safety, dependability, and a lack of bias in automated findings.
Regulatory Environment
Cardiomatics closely monitored the situation in Europe and the process of enacting a risk-based framework for regulating AI applications. They also keep an eye out for proposed regulations and requirements that could be implemented shortly. If new artificial intelligence standards and regulations are introduced, Cardiomatics will quickly incorporate them in the company’s and product operations, as well as extend the documentation and algorithm validation with the necessary proof for the dependability and safety of their product. It did, however, admit that objectively evaluating the efficacy of ECG reading algorithms is difficult.
“An impartial assessment of the effectiveness of algorithms can be quite difficult,” the report stated. Most of the time, it is conducted on a limited collection of data from a restricted population of patients who have been enrolled with only one device. They get signals from diverse groups of patients, which come from various recorders. They are developing this way of evaluating efficacy. The algorithms, which would allow them to evaluate their performance reliably independent of the study’s different circumstances, such as the recording equipment or the social group on which it would be tested.
ECG interpretation is based on experience, regulations, and art. When a person interprets an ECG, he or she notices a curve. It operates on the visual layer. Because an algorithm perceives a stream of numbers rather than an image, the work is transformed into a mathematical problem. But, in the end, you can’t design successful algorithms unless you know the domain, Cardiomatics said. “Our medical team’s expertise and experience are works of art in Cardiomatics. It is important to remember that algorithms are also trained on data generated by cardiologists. There is a substantial link between medical practitioners’ expertise and machine learning.”