# Clinical evidence Below are the studies on Cardisiography, each with its own population and its own results; supplementary literature on the technology follows at the end. ## How we report the values Each value is given as the manufacturer's site reports it for that study; values are never averaged or rounded. Fields with no citation or link in the source are left blank; copies on file-sharing sites are not linked. ## Source Study summaries and values are taken from the Cardisio GmbH studies page and the publications it links to; the four summary values of the multicenter study come from the manufacturer's home page. ### NHS, SBRI Healthcare: community-based heart testing in primary care (prospective cohort study) 628 participants; three primary care settings (general practice, pharmacy and a community health centre); asymptomatic, elevated-risk adults aged ≥18 years - Sensitivity: 73.8% - Specificity: 94.4% - Positive Predictive Value (PPV): 80% - Negative Predictive Value (NPV): 90.4% - Net Promoter Score: 88% - Recommendation rate among participants (492 of 560): 87.5% [British Journal of General Practice, 2025](https://bjgpopen.org/content/early/2025/04/25/BJGPO.2024.0183) ### AI-CCC study: comparison with coronary calcium scoring and CT angiography 118 consecutive outpatients evaluated for suspected CAD - Calcium score alone, AUC: 0.971 - CSG-index, AUC: 0.817 - p-factor, AUC: 0.811 - Calcium score, CSG-index and p-factor combined, AUC: 0.977 - Combined model, sensitivity: 95.7% - Combined model, specificity: 87.1% [](https://cardis.io/wp-content/uploads/2026/05/Abstract-E-1.pdf) ### New vectorcardiography algorithms for left ventricular hypertrophy and impaired systolic function Heidelberg University Hospital; 245 participants in total undergoing both CMR and VCG. Groups: reduced LVEF (n=40), left ventricular hypertrophy (n=208), reduced LVEF and left ventricular hypertrophy combined (n=34), controls with structurally normal hearts on CMR (n=31) - Reduced LVEF, AUC: 0.843 - Left ventricular hypertrophy, AUC range: 0.739–0.791 - Reduced LVEF and LVH combined, AUC: 0.866 - Reduced LV-EF (<40%), sensitivity: 80.0% - Reduced LV-EF (<40%), specificity: 83.9% - Reduced LV-EF (<40%), positive predictive value (PPV): 86.5% - Reduced LV-EF (<40%), accuracy: 81.7% - LV hypertrophy (LVH), sensitivity: 77.9% - LV hypertrophy (LVH), specificity: 80.6% - LV hypertrophy (LVH), positive predictive value (PPV): 96.4% - LV hypertrophy (LVH), accuracy: 78.2% - Reduced LV-EF and LVH combined, sensitivity: 82.4% - Reduced LV-EF and LVH combined, specificity: 83.9% - Reduced LV-EF and LVH combined, positive predictive value (PPV): 84.8% - Reduced LV-EF and LVH combined, accuracy: 83.1% [Janek Salatzki, Arne Kristian Schwarz, Sarah Wolfsteller, Nicolas Hein, Erika Poyo Medina, Norbert Frey, Henning Steen, Florian André, Alexander Passow. Universitätsklinikum Heidelberg, Journal of Electrocardiology, Volume 97, July–August 2026, 154272](https://doi.org/10.1016/j.jelectrocard.2026.154272) ### A mathematical approach to demonstrate R to T wave concordance of the human ECG No patient population; a three-dimensional, time-dependent setup simulating the transmural gradient [](https://www.nature.com/articles/s41598-025-20754-8) ### New algorithms for reduced left ventricular ejection fraction and left ventricular hypertrophy (DGK 2025) 280 patients in total. Groups: reduced LV-EF (n=40), LVH (n=209), reduced LV-EF and LVH combined (n=64), controls without cardiac pathology (n=31). Cardiac magnetic resonance imaging as reference standard - Reduced LV-EF, sensitivity: 80.0% - Reduced LV-EF, specificity: 85.7% - Reduced LV-EF, accuracy: 82.9% - LVH, sensitivity: 74.5% - LVH, specificity: 68.6% - LVH, accuracy: 73.6% - Reduced LV-EF and LVH combined, sensitivity: 79.3% - Reduced LV-EF and LVH combined, specificity: 85.7% - Reduced LV-EF and LVH combined, accuracy: 82.8% [German Society for Cardiology - Cardiovascular Research e.V – 91st Annual German Cardiac Society Meeting 2025](https://cardis.io/wp-content/uploads/2025/05/Evaluation-of-new-vector-electrocardiography-algorithms-for-identifying-patients-with-reduced-left-ventricular-ejection-fraction-and-left-ventricular-hypertrophy.pdf) ### AI-based 5-lead 3D-vectorcardiography (5L3DVCG-AI) detecting cardiac pathologies at rest (DGK Herztage 2024) 287 patients, multicentre - Heart rate, correlation coefficient (r) with 12-lead ECG: 0.97 - Electric heart axis, correlation coefficient (r) with 12-lead ECG: 0.71 - Visual detection of ECG abnormalities from 12-lead ECG and VCG-ECG, sensitivity: 75% - Visual detection of ECG abnormalities from 12-lead ECG and VCG-ECG, specificity: 100% [DGK-Herztage 2024](https://doi.org/10.1007/s00392-024-02526-y) ### 5L3DVCG-AI for identification of cardiac pathology in a mixed population (90th DGK Annual Meeting 2024) Mixed population [German Society for Cardiology - Cardiovascular Research e.V - 90. DGK-Jahrestagung 2024](https://cardis.io/wp-content/uploads/2024/04/DGK-2024-AI-CP-mixed-population-20240326.pdf) ### Validation of the artificial intelligence based 5 lead 3D vectorcardiography in comparison to the 12 lead ECG in a mixed population (AHA Scientific Sessions 2023) Mixed population [American Heart Association Circulation. 2023;148:A16473](https://cardis.io/wp-content/uploads/2023/12/AHA-CSG-ECG-V1.0-20231107-Upload.pdf) ### 5-lead 3D vectorcardiography differentiates between high and low cardiovascular risk profiles in patients with suspected or known coronary heart disease (DGK Heart Days 2023) Patients with suspected or known coronary heart disease [German Society for Cardiology - Cardiovascular Research e.V. Heart Days 2023](https://cardis.io/wp-content/uploads/2023/12/DGK-Poster-2-V1.0-20231004.pdf) ### Sensitivity and specificity of the artificial intelligence-based 5-lead 3D vectorcardiography in patients with suspected or confirmed coronary heart disease (DGK Heart Days 2023) Patients with suspected or confirmed coronary heart disease [American Heart Association Circulation. 2023;148:A15181](https://cardis.io/wp-content/uploads/2023/12/DGK-Poster-1-V1.0-20231004.pdf) ### HDZ-NRW / Mediacc: Cardisiography compared with CVRF score for non-invasive assessment of CHD 407 patients: 225 from the Heart and Diabetes Center North Rhine-Westphalia (HDZ-NRW), Bad Oeynhausen; 182 from a GP practice in Berlin - Negative predictive value (NPV): 91% [Congress of the European Society of Cardiology 2023](https://cardis.io/wp-content/uploads/2023/10/ESC-CSG-V1.1_Summary65.pdf) ### Heart Center Bad Oeynhausen: Cardisiography compared with myocardial SPECT in suspected and known CHD Patients with suspected CHD and known CHD - Negative predictive value: 98% [](https://cardis.io/wp-content/uploads/2023/03/Poster88_Li_Kn_Korr_DGN_Cardisio_20220414_Li.pdf) ### Cardisio validation study, Sana Heart Center Cottbus 106 patients in whom coronary angiography was indicated and performed; the Cardisiography result was blindly correlated with coronary angiography - Sensitivity: 95.4% - Specificity: 90% - Positive predictive value: 97.6% [Temirlan Erkenov, Tomasz Stankowski, Oliver Grimmig, Sören Just, Prof. Oleg Remizov, Prof. Dirk Fritsche. Cardisiography as a novel non-invasive diagnostic tool for the detection of coronary artery disease at rest – a first prospective study of diagnostic accuracy](https://cardis.io/wp-content/uploads/2023/03/Abstract-ESVCS-Padua_Dr.-Erkenov.pdf) ### Multicenter study: Cardisiography compared with coronary angiography in suspected coronary artery disease 595 adult patients without acute myocardial infarction who had undergone coronary angiography at three centres in Germany (male n = 355, aged 39–84; female n = 240, aged 28–88). Patients with 1-, 2- or 3-vessel disease on angiography were classed as affected: 246/355 men (69.3%) and 123/240 women (51.3%). Coronary angiography was the reference standard; accuracy was assessed with five-fold nested cross-validation. Values are given as mean ± standard deviation. - Sensitivity, male (paper): 97.2 ± 3.1% - Sensitivity, female (paper): 90.2 ± 4.2% - Specificity, female (paper): 74.4 ± 9.8% - Specificity, male (paper): 76.1 ± 8.5% - Positive predictive value (PPV), male (paper): 90.2 ± 3.0% - Positive predictive value (PPV), female (paper): 78.7 ± 8.6% - Negative predictive value (NPV), male (paper): 92.2 ± 6.9% - Negative predictive value (NPV), female (paper): 87.9 ± 8.4% - Area under the curve (AUC), male (paper): 0.900 ± 0.031 - Area under the curve (AUC), female (paper): 0.850 ± 0.059 - Accuracy, male (paper): 90.7 ± 3.3% - Accuracy, female (paper): 82.5 ± 6.4% Summary values from the manufacturer's home page; not reported in the paper. - Sensitivity: 95.4% - Specificity: 75.3% - Negative Predictive Value (NPV): 90.4% - Accuracy: 80.1% [Till Braun, Sotirios Spiliopoulos, Charlotte Veltman, Vera Hergesell, Alexander Passow, Gero Tenderich, Martin Borggrefe, Michael M. Koerner. Detection of myocardial ischemia due to clinically asymptomatic coronary artery stenosis at rest using supervised artificial intelligence-enabled vectorcardiography - A five-fold cross validation of accuracy. Journal of Electrocardiology 59 (2020) 100-105](https://doi.org/10.1016/j.jelectrocard.2019.12.018) #### Study limitations - The authors note that the sample of 595 patients is small compared with similar studies. - Only patients who had already undergone coronary angiography were included. According to the authors, applying the results to the general population must be discussed critically. - The study is retrospective. Conclusions about clinical applicability and diagnostic performance in primary care should be treated with caution. - The authors note that it is not straightforward to interpret what drives the neural network's output. - The authors state that prospective studies are needed. - The study was funded by Cardisio GmbH. Four of the authors are consultants to Cardisio GmbH. ### Artificial intelligence-supervised vectorcardiography for the diagnosis of a young adult with abnormal origin of the right coronary artery from aorta One young adult [Özyüksel A, Salatzki J, and Steen H (2024). Artificial intelligence- supervised vectorcardiography for the diagnosis of a young adult with abnormal origin of the right coronary artery from aorta. Cardiology in the Young, page 1 of 3.](https://www.cambridge.org/core/journals/cardiology-in-the-young/article/artificial-intelligencesupervised-vectorcardiography-for-the-diagnosis-of-a-young-adult-with-abnormal-origin-of-the-right-coronary-artery-from-aorta/5BEF0744C41055AB3D8487CA8FCD1B1A) ### The Poynting vector: power and energy in electromagnetic fields Carpenter KH. The Poynting vector: power and energy in electromagnetic fields. Department of Electrical and Computer Engineering. Kansas State University (2004) ### Spatial vector electrocardiography: the clinical characteristics of S-T and T vectors [Grant RP, Estes EH Jr, Doyle JT. Spatial vector electrocardiography; the clinical characteristics of S-T and T vectors. Circulation. 1951 Feb;3(2):182-97.](https://pubmed.ncbi.nlm.nih.gov/14812646/) ### Cardiogoniometry: a new noninvasive method for detection of ischemic heart disease [Saner H, Baur HR, Sanz E, Gurtner HP. Cardiogoniometry: a new noninvasive method for detection of ischemic heart disease. Clin Cardiol. 1983 May;6(5):207-10.](https://pubmed.ncbi.nlm.nih.gov/6851279/) ### Cardiogoniometry: an electrocardiographic, non-invasive and stress-free method for detecting cardiac ischemia [Sanz, Ee. Schüpbach, M. Cardiogoniometry: an electrocardiographic, non-invasive and stress-free method for the detection of cardiac ischemia. GMS Medical Informatics, Biometry and Epidemiology. 5 (2009).](https://www.egms.de/static/de/journals/mibe/2009-5/mibe000099.shtml) ### Applicability of cardiogoniometry as a non-invasive screening tool for the detection of graft vasculopathy in heart transplant recipients Heart transplant recipients [Spiliopoulos S, Hergesell V, Fischer D, Dapunt O, Krueger U, Koerfer R, Tenderich G. Applicability of cardiogoniometry as a non-invasive screening tool for the detection of graft vasculopathy in heart transplant recipients. Interact Cardiovasc Thorac Surg. 2016 Dec;23(6):976-978.](https://pubmed.ncbi.nlm.nih.gov/27481685/) ### Computer analysis of the corrected, orthogonal cardiogram [von Mengden, H.J., Brodda, K. Computer analysis of the corrected, orthogonal cardiogram. Archiv für Kreislaufforschung 67, 123–141 (1972).](https://link.springer.com/article/10.1007/BF02123139) ### Assessment of the spatial QRS-T angle by vectorcardiography: current data and perspectives [Voulgari C, Tentolouris N. Assessment of the Spatial QRS-T Angle by Vectorcardiography: Current Data and Perspectives. Curr Cardiol Rev. 2009 Nov;5(4):251-62. doi: 10.2174/157340309789317850. Erratum in: Curr Cardiol Rev. 2010 Nov;6(4):373.](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2842956/) ### Improved evaluation of left ventricular hypertrophy using the spatial QRS-T angle by electrocardiography [Maanja, M., Schlegel, T.T., Kozor, R. et al. Improved evaluation of left ventricular hypertrophy using the spatial QRS-T angle by electrocardiography. Sci Rep 12, 15106 (2022).](https://www.nature.com/articles/s41598-022-16712-3)