Poland-based healthtech AI startup Cardiomatics has announced a $3.2M seed raise to expand use of its electrocardiogram (ECG) reading automation technology.
This round will be led by Central and Eastern European VC Kaya. Nina Capital Nova Capital and Innovation Nest are also involved.
A $1M grant non-equity from the Polish National Centre of Research and Development is also included in this seed fund.
This 2017 startup offers a cloud-based tool that speeds up the diagnosis of cardiologists and clinicians, and improves efficiency in interpreting ECGs. The software generates reports about scans within minutes and can detect and analyze 20 different heart disorders.
Cardiomatics claims its technology will help to make healthcare more accessible. The tool allows cardiologists to optimize their work flow so that they are able to see and treat more patients. The tool allows smaller practices and GPs to provide ECG analyses to patients, without the need to refer them directly to specialists.
According to the startup, more than 700 people in more than 10+ countries are using its AI software, with more than 3,000,000 hours of ECG signals analyzed commercially.
It can integrate with over 25 ECG monitor devices. The software also boasts a modern cloud-based interface that is different from legacy medical software.
When we asked how accurate its AI’s ECG readings were, the startup replied that it has validated their accuracy. “The data set we use to create algorithms contains over 10 billion heartbeats and approximately 100,000 patient records. It is continuously growing.” We have created the majority of these data sets ourselves. The rest are available in public databases.
Ninety percent is used for training, while 10% are used to validate and test algorithms. Data-centric AI places great emphasis on test sets in order to ensure that the set contains the most accurate representations of client signals. The accuracy of algorithms is checked in experiments during continuous data and algorithm development. We do this with an average of one per month. It is checked by our clients every day in their clinical practice.
Cardiomatics stated that it would use seed financing to fund product development and expand existing market business, as well as to prepare for launching into new markets.
The proceeds of the round will go to fast-paced European expansion. They will include scaling up AI technology that is market leading and making sure physicians enjoy the best possible experience. The product is being prepared for launch in new markets. It stated that future plans included FDA certification and entry into the US market.
In 2018, the European Medical Device Certification was granted to the AI Tool. However, it is worth noting that both the EU’s regulatory system for medical devices (and AI) continues to evolve. This year’s update of the EU’s Medial Devices Directive (now the EU Medical Device Regulation), came into effect in May.
A new risk-based framework for applications of AI — aka the Artificial Intelligence Act — is also incoming and will likely expand compliance demands on AI healthtech tools like Cardiomatics, introducing requirements such as demonstrating safety, reliability and a lack of bias in automated results.
When asked about regulatory issues, it stated that “when we launched in 2018, we were among the first AI-based medical devices approved in Europe. We monitor the European situation and work closely with the EU to create a framework that is risk-based for AI regulation. This allows us to keep up with the times. Draft regulations and any requirements may soon be implemented are also monitored. We will implement any new requirements or standards for artificial intelligence in our company and products immediately.
It also acknowledged that it is difficult to objectively measure the efficacy ECG reading algorithms.
TechCrunch said that an objective evaluation of algorithm effectiveness can prove difficult. It said that most often, it’s done on data only from one patient. Signals from different patients are received by us from different recorders. This is a method we are currently working on. They will be able to use our algorithms to accurately evaluate the performance of their participants regardless of the variables that are associated with the study such as the recording device and the group being tested.
ECG interpretation can be performed by doctors. A human can interpret an ECG by looking at a curve. The algorithm works with a visual layer. The algorithm can see a stream instead of seeing a photo, making the task a math problem. However, effective algorithms cannot be built without knowledge about the domain.” it said. This knowledge, as well as the expertise of our medical staff are an artifact of Cardiomatics. It is important to remember that algorithms can also be trained from data generated by cardiologists. Machine learning and medical experience have a strong connection.”
