Our AI technology not only diagnoses cardiac arrhythmias from ECG, but also predicts the potential of cardiac arrhythmias within the next six months. Checkup my health status through ECG data any where in the world.
Being able to predict an episode of cardiac arrhythmias early enough would allow patients to take
preventive measures to keep their cardiac rhythm stable. However, current methods based on the analysis of
heart rate and electrocardiogram (ECG) data are only able to detect atrial fibrillation right before its
onset and do not provide an early warning.
CLholmes® not only diagnoses cardiac arrhythmias in real-time with precision but also supports the
development of appropriate medical actions and preventative plans through early prediction of arrhythmia
onset.
Notably, our technology overcome the lack of medical reliability of ECG analysis result obtained from
widely spread SmartWatch such as Galaxy Watch, Apple Watch, etc.
Bringing the power of AI to cardiac care CLholmes has developed the world’s first cardiology-as-a-service platform powered by artificial intelligence that augments a clinian’s ability to identify, triage and diagnose patients at scale.
Cutting-edge technology
Our algorithm (CAT) automatically identifies 21 cardiac patterns with cardiologist-level accuracy in a multi-class scenario, setting a new benchmark in cardiac diagnosticsDeviceneutral
The versatility of CAT allows for the analysis of ECG data from any type of lead, be it wearable, mobile, or implantable devices, ensuring broad applicability and ease of integration into various healthcare settings.Deep science and clinical expertise
With over two decades of experience in building data pipelines within cardiology, CLholmes® has contributed to cutting-edge scientific publications, demonstrating our profound expertise and commitment to advancing the field.Prediction capabilities
CLholmes® is at the forefront of developing predictive ECG-based AI models for use as prognostic tools. The model is designed to identify patients at an increased risk of developing persistent atrial fibrillation within the next six months. This model has been featured in the "Computer in Biology and Medicine" journal, which is among the highest-impact factor journals in the digital heath research domain.
We have developed the Component-Aware Transformer (CAT) technology, a proprietary innovation rooted in deep knowledge and research of cardiac clinical practices. Additionally, our diagnostic and prediction transformer models are based on an in-depth understanding and fundamental research of the human heart, emphasizing our commitment to advancing heart health through our own technological developments.
Our AI models were trained on the most diverse and longest ECG databases, from both healthy individuals and diagnosed patients. More than 10,000 patients (1 million ECG hours) were monitored in clinical trials.
Over 3 years of intensive research and development, we have succeeded in developing CLholmes® which accurately identifies 21 types of cardiac arrhythmia-matching the precision of cardiologists-level. Our novel technology classifies ECG into sub-waveforms (P, QRS, and T waves) and analyzes these patterns significantly to reduce the misdiagnosis rate to less than one-fifth compared to current commercialized AI technologies. Remarkably, CLholmes® achieves a diagnostic accuracy akin to a 12-lead ECG even with single ECG signals, boasting an AUROC: 0.9823, Sensitivity: 94.10%, Specificity: 94.33%.
CLholmes® is a flexible and versatile solution for performing clinical Holter ECG analysis. Its user-friendly and informative data presentation and intuitive analysis tools make the software efficient and easy to use when analyzing multiple days of ECG recordings. The software allows for scanning long measurements efficiently in a shorter time, thus speeding up the final diagnosis.
• At-a-glance dashboards allow a quick overview of the recording
• Intuitive user interface for faster editing and analyzing
• Updated Templates model allows even quicker reclassification of beats
• Arrhythmia analysis with overview and event strips
• Versatile P-wave and AV block detection
• Precise and detailed atrial analysis
• A patient event analysis tool
• QRS templates
• QT analysis
• ST analysis
• Heart rate variability (HRV) analysis
• Informative and graphical reporting
• Automated narrative in reporting
• Custom reporting and customized report templates
Simple user interface and intuitive ECG navigations make the software fast to learn and support the easy and efficient use of the software. CLholmes® provides views and dashboards that present all the important ECG data and analytics to the user at a glance. Summary views, categorized strip's view, and arrhythmia tools support easier data interpretation and quicker data editing, along with a Full Disclosure view that maintains the overall picture of the measurement being analyzed.
One of the main design principles of CLholmes® is arrhythmia centric analyzing where the user focus is on analyzing and reannotation rhythms instead of beats. The software has a template function, which groups different beats based on the morphologies and other values. Reclassifying and deleting beats is made easy and quick. If needed the application allows template editing either with a purpose-designed templates tool or from any view with ECG data.
The versatile P-wave and AV block detection, PQ and QTc time distribution by severity, novel De-/Repolarization surface map visualization as well as information on the effects of the changing heart rate to PQ and QT times provide valuable information on all of the critical atrial and QT events.
Available ready-made narrative phrases with editing possibilities and the visual presentation of data provide everything that is needed for comprehensive and informative reporting. You can choose which pages and analyses to include in the final report.
CLholmes® integrates a GPT4-based human-computer interaction service (Chatbot) that provides easy-to-understand and reliable analysis results. The GPT model, further trained on international arrhythmia diagnostic guidelines, analyzes ECG features extracted through CAT Algorithm to provide users with detailed information about their cardiac health. This Chatbot not only offers advice on lifestyles to improve heart health but also provides personalized answers to questions related to arrhythmias in daily life. These features are especially useful in regions with limited medical services, helping individuals manage their health more effectively.