Establishment of magnetocardiogromics platform with reference range for normal magnetocardiogram Authors: Yijing Guo, Jian Ma, Hong Shen, Guangya Zhang, Jiabin Zang, Yujie Zhang, Chengxing Shen Published: 2025 ABSTRACT AbstractBackgroudMagnetocardiogram is a promising technology for future heart disease diagnosis and prognosis. However, there is no unified standard of magnetocardiogram in the normal population for further study and comparison. This study aimed to establish the magnetocardiogromics platform that could provide a well-acknowledged parameter library and its reference range in the normal population for future study.Methods93 cases were enrolled in study set for parameter reference range determination. Another 111 cases were used as validation.ResultsWe built a library with 246 parameters with their reference range in the normal population. In addition, we established a standard process for future magnetocardiogram study.ConclusionsThis is the first time that we proposed the concept of magnetocardiogromics, which standardized the parameter library and its reference range as well as establishing a unified analysis platform.Clinical trial numberNot applicable.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12880-025-02010-7.Keywords: Heart, Magnetocardiogram, Omics, Reference range Background Magnetocardiogram (MCG) is a novel technology that records and analyzes changes in the cardiac magnetic field, whose development has become increasingly mature in recent years. With the development of quantum science, superconducting quantum interference devices(SQUID) and spin exchange relaxation free atomic magnetometer(SERF AM) based MCG can detect the magnetic field generated by myocardial cell excitation around the heart [1, 2]. Thus, it can record and process this cardiac magnetic field information, after which analyze its changing characteristics. Due to the close relationship between cardiac magnetic changes and electrocardiographic activity, both MCG and electrocardiography can reflect changes in cardiac electrophysiology. Because of the pathological heterogeneity of repolarization processes in the ischemic and peripheral regions of the myocardium, local damage currents appear at the junction of normal myocardium and ischemic areas, resulting in corresponding changes in MCG [3]. Owing to the high sensitivity of this technology, many studies have found that the early predictive value of MCG for myocardial ischemia, which showed promising diagnosibility compared with traditional clinical detection methods such as electrocardiography, echocardiography, and plate exercise testing [4, 5]. In addition, MCG only takes 5 to 10 min, which is radiation free and contrast agents free. Therefore, it has strong applicability in the different application scenarios including chest pain differential diagnosis, patient contrast agent allergies, and primary healthcare settings with limited large-scale diagnostic equipments [6]. Besides, research related to MCG also involves the diagnosis of arrhythmia and congenital heart disease in fetuses [7, 8]. Thus, the application prospects of MCG, a highly sensitive, non-contact, non-invasive, and radiation free examination method, in the diagnosis of heart diseases are very broad. However, the promotion and application of MCG are currently greatly limited. Firstly, MCG machines currently used by various medical centers and research institutions worldwide has different parameters and standards due to differences in technical routes, analysis methods, equipment performance, etc., which makes it impossible to have horizontal comparisons [9]. Secondly, current studies on MCG are mostly case-control studies, whose screening of the control group is often not strict enough. For example, it is difficult for the control group to simultaneously exclude large vessel obstruction and microcirculation disorders, which has a significant impact on the research results [10, 11]. Thirdly, most published research results are limited to a single center understanding of MCG. They often analyze cases based on a few parameters, making it difficult to form a unified standard for worldwide communication [12]. At the same time, the limitations of understanding MCG may also affect its multidimensional diagnosis of diseases. Therefore, the establishment of magnetocardiogromics can solve above problems by using a universal MCG language to describe each case comprehensively, thus better serving clinical practice and facilitating MCG communication and comparison worldwide. We first proposed the concept of magnetocardiogromics in this study, in order to conduct a complete analysis of MCG through big data, which provided a unified MCG language for communication and comparison of MCG researches from different centers and devices worldwide. The first step of this work urgently requires us to establish a parameter library for MCG and determine its reference range in the normal population. This can also promote the establishment of a standardized system for this advanced technology in the future clinical work. Methods Research subjects This study was approved by the institutional review board of the Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine (Approval Number: 2024-KY-249(K)). All selected individuals have signed informed consent forms. This study was a retrospective, observational study that explored the reference range of specific MCG parameters in the normal population. The study set of the normal range of MCG included 93 cardiovascular normal population subjects who met the following inclusion and exclusion criteria during outpatient or inpatient in the Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine from February 2023 to February 2024. Inclusion criteria for the study set: ① Age ≥ 18 years and ≤ 80 years; ② Coronary angiography or CT angiography did not show ≥ 50% stenosis; ③Complete caIMR or cardiac magnetic resonance imaging(MRI) or Compresearcuted Tomography(CT) of myocardial perfusion imaging to exclude myocardial microcirculation disorders; ④ No significant abnormalities were found in cardiac ultrasound and 24-hour Holter; ⑤ Voluntarily participate in this study. Exclusion criteria: ① Previous history of coronary heart disease, valvular heart disease, cardiomyopathy, arrhythmia, heart failure, or other underlying cardiac conditions; ② There are implanted metal materials in the body (such as pacemakers, metal internal fixation after fractures, etc.); ③ Serious complications or comorbidities of other organs or systems (such as tumors, liver and kidney failure, etc.); ④ Other situations that are not suitable for participation in this study. The validation set of normal values for MCG included 111 cases who completed physical examinations at the Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine during the same period and met the following inclusion and exclusion criteria. Inclusion criteria for validation set: ① Age ≥ 18 years and ≤ 80 years; ② Completed electrocardiogram, cardiac ultrasound, and troponin I test showed no significant abnormalities; ③ No cardiovascular related symptoms and signs; ④ Voluntarily participate in this study. The exclusion criteria are the same as the study set. Collection of general clinical data In the study, we collected general clinical data of the candidates through the electronic medical record system, including demographic data (age, gender, occupation, marital status, etc.), body mass index, smoking and drinking history. Collection of MCG data We collected the results of selected individuals’ MCG information through an electronic medical record system. The selected participants received MCG test by using a domestically produced unshielded 9-channel MCG instrument (MD-U092001, Shanghai MEDI Instruments Ltd., Shanghai, China). During the examination, the subjects were placed in a supine position, while a superconducting quantum interferometer was used to collect MCG signals and obtain MCG data and images. After MCG signals collection, we would get three MCG images as regular MCG studies did including isomagnetic map, current density map, and waveform map. Then we developed a parameter library, which could thoroughly describe the MCG characteristics of these three images (Supplementary Table 1). This library described the characteristics of the isomagnetic map, current density map, and waveform map from multiple perspectives, including position, morphology, direction, and waveform, which let us quantify the MCG data. We used Labview2012 software to process and analyze the obtained cardiac magnetic data and images. Statistical analysis SPSS21.0 software was used for statistical analysis. The measurement data was represented by x ± s. And group t-test was used for comparison between two groups. Enumeration data was represented by ratios. Fisher’s exact probability method and chi-square test were used for inter group comparisons. The percentile method was applied to determine the range of reference values for MCG parameters, which was presented as P2.5-P97.5. The coefficient of variation is the ratio of standard deviation to mean. Pearson correlation coefficient was performed in correlation analysis. The difference is statistically significant with P < 0.05. Results General clinical data The study group consisted of 93 individuals who were cardiovascular normal, while the validation group consisted of 111 individuals who underwent physical examination. The general clinical data of both groups were detailed in Table 1. Table 1The general clinical data of study group and validation groupSubjectStudy group(n = 93)Validation group(n = 111)t valueP value Age(x ± s)59.15 ± 12.3941.95 ± 9.6911.016< 0.001 Male39(41.94%)60(54.05%)-0.093BMI(kg/m2, x ± s)22.31 ± 4.7321.82 ± 5.040.7110.478Occupation-0.154 On board50(53.76%)71(63.96%) Retired43(46.24%)40(36.04%)Marrige-0.708 Married79(54.95%)92(82.88%) Others14(15.05%)19(17.12%)Smoking22(23.66%)18(16.22%)-0.216Alcohol abuse21(22.58%)20(18.02%)-0.484 The general clinical data of study group and validation group Grouping of MCG parameters and their normal range The parameters of the MCG were divided into three groups based on the isomagnetic map, current density map, and waveform map (Figures.1.), with a total of 246 parameters. The reference value range, presented as P2.5-P97.5, with their meaning and the extraction methods was shown in Supplementary Table 1. Representative MCG chart in the normal population. (a) Representative isomagnetic map. (b) Representative current density map. (c) Representative wave pattern map Evaluation of MCG parameters We calculated the coefficient of variation in order to evaluate the reliability of these MCG parameters, which represented the degree of dispersion of these indicators. We found 51 parameters with absolute value of coefficient of variation more than 1 while the rest less than 1 (Fig. 2a). Due to the significant difference of age between the study group and validation group, we also calculated the correlation coefficient between MCG parameters and age. More than half parameters showed the absolute value of correlation coefficient less than 0.1 when connected them with age, while only 22 parameters showed the absolute value of correlation coefficient more than 0.2 (Fig. 2b). Evaluation of MCG parameters in the study. (a) The absolute value of coefficient of variation of parameters. (b) The absolute value of correlation coefficient between parameters and age Validation set results Comparing all MCG parameter values in the validation set with the reference range mentioned above, 236 parameters were able to achieve the reference range with conformity rate of over 90%, accounting for 95.93% (236/246). Among them, 46 parameters were completely within the reference range, accounting for 18.70% (46/246). The specific 10 parameters that met the conformity rate of less than 90% were shown in Fig. 3. Horizontally comparing the MCG parameters of each validation set case, 99.10% (110/111) of cases had over 90% parameters within the reference range. 10 parameters that met the conformity rate of less than 90% Schematic diagram for Establishing magnetocardiogromics In magnetocardiogromics, we first established a parameter library. In future research, researchers will extract parameters based on standardized algorithms and use statistical methods to quickly obtain diagnostic models after retrieving case MCG information. At the same time, researchers will add labels to the corresponding parameters in this model, which provide specific information of specific parameters for other researchers in the same field (Fig. 4). Schematic diagram of the standardization research platform for MCG based on magnetocardiogromics Discussion This study first proposed the concept of magnetocardiogromics, which is a research platform based on omics analysis of MCG. We first developed 246 parameters based on the isomagnetic map, current density map, and wave pattern map to describe the MCG information we obtained. These parameters described the shape, direction, and position of the isomagnetic map, the direction and position of the current density map, as well as the waveform and phase of the wave pattern map. Basically, they could provide a comprehensive description of the MCG information we currently obtained in the form of parameters. After obtaining parameter information, this study also accurately enrolled the normal myocardial perfusion population into research analysis for the first time. Previous studies on myocardial ischemia often focused on the ischemia caused by large vessel (coronary artery) occlusion, which was the population with coronary heart disease. According to the 2024 ESC Chronic Coronary Syndromes guideline, in addition to epicardial vascular structure or dysfunction, coronary microcirculation structure or dysfunction also affected myocardial perfusion in chronic coronary syndrome [13]. Therefore, during the enrollment stage of this study, we did not only excluded individuals with coronary artery stenosis exceeding 50%, but also excluded myocardial microcirculation disorders through caIMR, cardiac MRI, or CT myocardial perfusion imaging, which ensured the accuracy of our enrollment population. Due to various reasons that limited the number of people who could complete the above-mentioned examinations in daily clinical work, we carefully enrolled nearly 100 cases in the study group and obtained a reference range for all parameters. Then we calculated the coefficient of variation showing great reliability of them, which provided a solid foundation for magnetocardiogromics building. In addition, another 111 cases were included in the validation group through relatively simple examinations, further verifying the reliability of the reference values. In this study, it would be best if we could use the same inclusion criteria to select another group of people for internal validation, or use the same inclusion criteria for external validation in other centers. However, it was difficult and impracticable to have all normal people to receive coronary angiography, cardiac magnetic resonance imaging or computed tomography. Thus, we recruited the validation group for reference range validation from physical examination center who received non-invasive examinations without indication of related cardiac disease. After comparing two groups, we found that the vast majority of the range in the validation group matched the study group. However, the difference of age between two groups might raise the concern of bias or interference. We applied the correlation analysis that showed little correlation between MCG parameters and age. This proved that our parameter reference range had good age extrapolation ability. In addition, the calculation of the coefficient of variation had given us a deeper understanding of these parameters, in which with conservative reference ranges might have higher clinical value. In the future, we will organize more healthy people and centers to join our research, which has the same inclusion criteria for internal and external validation. In that way, we will have more confidence in the appropriate parameter ranges, while parameters with high coefficients of variation or those that do not meet the validation criteria will be eliminated eventually from our magnetocardiogromics platform, In previous studies on MCG, researchers typically used two methods. The first method was to select a fixed number of parameters based on their experience for analysis. Taking ischemic heart disease as an example, most related studies focus on the T wave or TT segments of the isomagnetic field due to their stable morphology in various phases of MCG [14, 15]. However, the approach of attributing MCG information to several parameters in such studies was bound to losing most of the information. In that way, we did not know the exact position of this discarded information in studying diseases, which greatly reduces the accuracy of the research. The differences in parameters among various centers also render the research incomparable and non-repeatable [16–18]. The second research method was usually based on machine learning to analyze the parameters of MCG. Compared to the first approach, this type of scheme could obtain more MCG information, theoretically enabling better construction of MCG diagnostic models for various types of heart diseases [19, 20]. In this case, it would pose a great obstacle to our further understanding of MCG information. We cannot understand the actual meaning and clinical value of the parameters involved in the diagnostic model established by the machine, as well as analyze and quantify certain information with special significance. Thus, the actual normal reference range for MCG parameters in the normal population was desperately to be determined. In our study, it firstly proposed the concept of magnetocardiogromics, which standardized the parameter library and analysis methods of MCG as well as establishing a unified analysis platform for MCG. In summary, the magnetocardiogromics proposed in this study was promising and had great potential for future development. Next, we will continue to identify and fill in gaps in parameter development, as well as supplement them with meaningful parameters obtained from machine learning. In terms of parameter labeling, we will imitate the methods of omics in bioinformatics to annotate graphics, bands, diseases, and other methods for subsequent analysis. The disease diagnosis and prognosis model will be analyzed using omics methods. We believe that the establishment of magnetocardiogromics will provide the most standardized, comparable, and high-yield platform for researchers in this field in the future. Conclusion This study is the first to propose the concept of magnetocardiogromics, which establishes the MCG parameter library as well as the reference range in the normal population. It provides a unified research method and comparative approach for future MCG related research, which is of pioneering significance. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements We acknowledge the support received from National Key Research and Development Program of China. Abbreviations Magnetocardiogram Spin exchange relaxation free atomic magnetometer Superconducting quantum interference devices Author contributions Y. G.: Investigation, Formal analysis, Data curation, Writing original draft.J. M.: Validation, Resources, Writing-review and editing, Supervision.H. S.: Data curation, Conceptualization, Writing-review and editing.G. Z.: Data curation, ValidationJ. Z.: Data curation, InvestigationY. Z.: Methology, ValidationC. S.: Conceptualization, Methodology, Validation, Writing-review and editing, Supervision, Project administration. Funding This research is supported by National Key Research and Development Program of China (2022YFC2407000), and Diagnosis and Treatment Technology Promotion and Management Optimization Project of Shanghai Shenkang Hospital Development Center (No. SDHC2023101). Data availability All data generated or analyzed during this study are included in this published article and its supplementary information files. Other specific materials are with the corresponding author and can be obtained by mail if necessary. Declarations Ethics approval and consent to participate This study was conducted according to the tenets of the Declaration of Helsinki. Informed consent was obtained from each patient. This study had approval from our hospital’s ethics review committee and each patient provided signed informed consent. Consent for publication All participants provided written informed consent. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Yijing Guo, Jian Ma and Hong Shen contributed equally to this study. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement All data generated or analyzed during this study are included in this published article and its supplementary information files. Other specific materials are with the corresponding author and can be obtained by mail if necessary.