Data overload in cardiology
Data overload in cardiology
Get in touch

The pandemic has highlighted the challenge of data overload in cardiology: after just ten months, 87,000 scientific publications have addressed Corona and COVID-19. Monitors from “Retraction Watch” identified hundreds of articles that were retracted, including some from leading journals. In addition, beyond the pandemic, physicians from all disciplines find it challenging to keep up to date with the enormous volume of findings from research, to verify their validity, and to align them with their individual patient cases. Technology, including AI, is there to help.

Like their colleagues, cardiologists find it hard to cope with the steady increase in scientific literature, particularly in times of increasing workloads. “Already today, data is generated in the range of several gigabytes per each inpatient case in cardiology,” underlines Prof. Dr. med. Benjamin Meder, FESC. Meder is Deputy Medical Director of the Department for Cardiology, Angiology, and Pneumology as well as Director of the Institute for Cardiomyopathies at Heidelberg University Hospital in Germany. “Nowadays, this includes not only imaging data (3D and 4D), but also molecular high-throughput data from genomics, and gene expression analyses from molecular pathology.

We have been noticing, in our daily routine, a significant increase in the volume of data in recent years,” comments PD Dr. med. David Duncker. He is Deputy Director of the Hannover Heart Rhythm Center (HHC), Department of Cardiology and Angiology, Medical University Hannover (MHH). “Last year, contact restrictions served as a catalyst for these developments – in many cases, we could not see our patients and had to rely on phone and video consultations. In the heart rhythm outpatient department, this virtual setting led to monitoring gaps.” Dr. Duncker’s team switched to an app which records heart rhythm and rate. “We learned a lot for our clinical routine from this project,” outlines Dr. Duncker, “including the fact that wearables can easily be integrated into clinical pathways and that they can support clinical decisions.” Seamless integration, adds Prof. Meder, is a success factor for IT solutions. With the daily routine of physicians characterized by stress and high throughput, software needs to be easily accessible and deliver results fast. Reimbursement for the use of technology, observes Dr. Duncker, remains to be settled.


AI reduces complexity, identifies risks

Software can support physicians by pre-sorting and analyzing these large volumes of data,” says Dr. Duncker. Physicians need to decide which information is relevant for diagnosis and definition of therapy for the right patient at the right time. Patient history, exam results, and lab and imaging reports need to be taken into consideration, while the availability of more data adds to the complexity. 

All this data, adds Prof. Meder, allows new knowledge to be garnered and applied in cardiology. Technologies such as machine learning provide essential tools in an environment defined by a multiplicity of data types and streams. 

According to Dr. Duncker, available AI and deep learning algorithms based on simple ECGs are capable of predicting whether a patient is currently suffering or will suffer in the near future from heart insufficiency, whether he/she has myocardial stenosis, or will develop arrhythmia. “These developments are extremely promising because they demonstrate the potential.” Dr. Duncker thinks that AI will not eliminate physicians but instead will “enable us to make better use of information available than we as humans are capable of.” 

Until now, however, the potential of this data has gone way beyond what is actually being achieved in tapping it. “Variables are being selected in order to fit the capabilities of the human brain,” explains Prof. Meder. “The challenge is not only in devising new AI methods but also in evaluating them in a clinical setting, and to integrate them into a daily routine – provided their operating principles are made transparent.


Decision support – a major trend

In Germany and beyond, clinical decision support involving vetted data and innovative technologies has developed into a major trend. It is a key item in the German Hospital Future Act (KHZG) which finances hospital investment in technology. “Large and established, as well as startup companies digitize medical knowledge and create algorithms designed to support physicians,” says Prof. Meder. He is active in the eCardiology section of the German Society of Cardiology (DGK), which works on multiple projects. The DGK has engaged in this field at an early stage, creating a dedicated task force and a subsequent project group. Prof. Meder summarizes: “If we do this right, these new solutions will serve to improve the quality of medical care significantly.” 

Dr. Duncker, also active in DGK eCardiology, adds: “education needs to provide the foundation for physicians to learn to manage voluminous data and technological tools.” 



In this world of AI systems, Cardiomatics provides automated analysis of ECG vetted by the analysis of millions of clinical cases. As a result, it saves clinicians valuable time and enables precision in diagnosis and therapy. Find out more about Cardiomatics technology.

Get in touch