A portable prototype magnetometer to differentiate ischemic and non-ischemic heart disease in patients with chest pain

The magnetocardiogram

The magnetic field produced by the heart's electrical activity is called the magnetocardiogram (MCG). The first 20 years of MCG research established most of the concepts, instrumentation, and computational algorithms in the field. Additional insights into fundamental mechanisms of biomagnetism were gained by studying isolated hearts or even isolated pieces of cardiac tissue. Much effort has gone into calculating the MCG using computer models, including solving the inverse problem of deducing the bioelectric sources from biomagnetic measurements. Recently, most magnetocardiographic research has focused on clinical applications, driven in part by new technologies to measure weak biomagnetic fields.

Survey of Transfer Learning Approaches in the Machine Learning of Digital Health Sensing Data

Machine learning and digital health sensing data have led to numerous research achievements aimed at improving digital health technology. However, using machine learning in digital health poses challenges related to data availability, such as incomplete, unstructured, and fragmented data, as well as issues related to data privacy, security, and data format standardization. Furthermore, there is a risk of bias and discrimination in machine learning models. Thus, developing an accurate prediction model from scratch can be an expensive and complicated task that often requires extensive experiments and complex computations. Transfer learning methods have emerged as a feasible solution to address these issues by transferring knowledge from a previously trained task to develop high-performance prediction models for a new task. This survey paper provides a comprehensive study of the effectiveness of transfer learning for digital health applications to enhance the accuracy and efficiency of diagnoses and prognoses, as well as to improve healthcare services. The first part of this survey paper presents and discusses the most common digital health sensing technologies as valuable data resources for machine learning applications, including transfer learning. The second part discusses the meaning of transfer learning, clarifying the categories and types of knowledge transfer. It also explains transfer learning methods and strategies, and their role in addressing the challenges in developing accurate machine learning models, specifically on digital health sensing data. These methods include feature extraction, fine-tuning, domain adaptation, multitask learning, federated learning, and few-/single-/zero-shot learning. This survey paper highlights the key features of each transfer learning method and strategy, and discusses the limitations and challenges of using transfer learning for digital health applications. Overall, this paper is a comprehensive survey of transfer learning methods on digital health sensing data which aims to inspire researchers to gain knowledge of transfer learning approaches and their applications in digital health, enhance the current transfer learning approaches in digital health, develop new transfer learning strategies to overcome the current limitations, and apply them to a variety of digital health technologies.

Magnetocardiography-based coronary artery disease severity assessment and localization using spatiotemporal features

Objective.This study aimed to develop an automatic and accurate method for severity assessment and localization of coronary artery disease (CAD) based on an optically pumped magnetometer magnetocardiography (MCG) system.Approach.We proposed spatiotemporal features based on the MCG one-dimensional signals, including amplitude, correlation, local binary pattern, and shape features. To estimate the severity of CAD, we classified the stenosis as absence or mild, moderate, or severe cases and extracted a subset of features suitable for assessment. To localize CAD, we classified CAD groups according to the location of the stenosis, including the left anterior descending artery (LAD), left circumflex artery (LCX), and right coronary artery (RCA), and separately extracted a subset of features suitable for determining the three CAD locations.Main results.For CAD severity assessment, a support vector machine (SVM) achieved the best result, with an accuracy of 75.1%, precision of 73.9%, sensitivity of 67.0%, specificity of 88.8%, F1-score of 69.8%, and area under the curve of 0.876. The highest accuracy and corresponding model for determining locations LAD, LCX, and RCA were 94.3% for the SVM, 84.4% for a discriminant analysis model, and 84.9% for the discriminant analysis model.Significance. The developed method enables the implementation of an automated system for severity assessment and localization of CAD. The amplitude and correlation features were key factors for severity assessment and localization. The proposed machine learning method can provide clinicians with an automatic and accurate diagnostic tool for interpreting MCG data related to CAD, possibly promoting clinical acceptance.

Electromagnetic field and cardiovascular diseases: A state-of-the-art review of diagnostic, therapeutic, and predictive values

Despite encouraging advances in early diagnosis and treatment, cardiovascular diseases (CVDs) remained a leading cause of morbidity and mortality worldwide. Increasing evidence has shown that the electromagnetic field (EMF) influences many biological processes, which has attracted much attention for its potential therapeutic and diagnostic modalities in multiple diseases, such as musculoskeletal disorders and neurodegenerative diseases. Nonionizing EMF has been studied as a therapeutic or diagnostic tool in CVDs. In this review, we summarize the current literature ranging from in vitro to clinical studies focusing on the therapeutic potential (external EMF) and diagnostic potential (internal EMF generated from the heart) of EMF in CVDs. First, we provided an overview of the therapeutic potential of EMF and associated mechanisms in the context of CVDs, including cardiac arrhythmia, myocardial ischemia, atherosclerosis, and hypertension. Furthermore, we investigated the diagnostic and predictive value of magnetocardiography in CVDs. Finally, we discussed the critical steps necessary to translate this promising approach into clinical practice.

Clinical magnetocardiography: the unshielded bet-past, present, and future

Magnetocardiography (MCG), which is nowadays 60 years old, has not yet been fully accepted as a clinical tool. Nevertheless, a large body of research and several clinical trials have demonstrated its reliability in providing additional diagnostic electrophysiological information if compared with conventional non-invasive electrocardiographic methods. Since the beginning, one major objective difficulty has been the need to clean the weak cardiac magnetic signals from the much higher environmental noise, especially that of urban and hospital environments. The obvious solution to record the magnetocardiogram in highly performant magnetically shielded rooms has provided the ideal setup for decades of research demonstrating the diagnostic potential of this technology. However, only a few clinical institutions have had the resources to install and run routinely such highly expensive and technically demanding systems. Therefore, increasing attempts have been made to develop cheaper alternatives to improve the magnetic signal-to-noise ratio allowing MCG in unshielded hospital environments. In this article, the most relevant milestones in the MCG's journey are reviewed, addressing the possible reasons beyond the currently long-lasting difficulty to reach a clinical breakthrough and leveraging the authors' personal experience since the early 1980s attempting to finally bring MCG to the patient's bedside for many years thus far. Their nearly four decades of foundational experimental and clinical research between shielded and unshielded solutions are summarized and referenced, following the original vision that MCG had to be intended as an unrivaled method for contactless assessment of the cardiac electrophysiology and as an advanced method for non-invasive electroanatomical imaging, through multimodal integration with other non-fluoroscopic imaging techniques. Whereas all the above accounts for the past, with the available innovative sensors and more affordable active shielding technologies, the present demonstrates that several novel systems have been developed and tested in multicenter clinical trials adopting both shielded and unshielded MCG built-in hospital environments. The future of MCG will mostly be dependent on the results from the ongoing progress in novel sensor technology, which is relatively soon foreseen to provide multiple alternatives for the construction of more compact, affordable, portable, and even wearable devices for unshielded MCG inside hospital environments and perhaps also for ambulatory patients.

Magnetocardiography for the detection of myocardial ischemia

Ischemic heart disease (IHD) continues to be a significant global public health concern and ranks among the leading causes of mortality worldwide. However, the identification of myocardial ischemia in patients suspected of having coronary artery disease (CAD) remains a challenging issue. Functional or stress testing is widely recognized as the gold standard method for diagnosing myocardial ischemia, but it is hindered by low diagnostic accuracy and limitations such as radiation exposure. Magnetocardiography (MCG) is a non-contact, non-invasive method that records magnetic fields produced by the electrical activity of the heart. Unlike electrocardiography (EKG) and other functional or stress testing, MCG offers numerous advantages. It is highly sensitive and can detect early signs of myocardial ischemia that may be missed by other diagnostic tools. This review aims to provide an extensive overview of the available evidence that establishes the utility of MCG as a valuable diagnostic tool for identifying myocardial ischemia, accompanied by a discussion of potential future research directions in this domain.

Sparse Representation for Cardiac Electrical Activity Imaging in Magnetocardiography

Signal sparsity has been widely discussed in communication system, cloud computing, multimedia processing and computational biology. Reconstructing the sparsely distributed current sources of the heart by means of non-invasive magnetocardiography (MCG) measurement and various optimization methods provides a new way to solve the inverse problem of the cardiac magnetic field. The problem of sparse source location of MCG is in the time series of MCG measurement caused by active sparse current source, can the spatiotemporal source be reconstructed accurately and effectively? For the above problem, the scientific contributions of the paper include: (1) A modified focal underdetermined system solver algorithm is proposed for a sparse solution, by combing with dynamic regularization factor and smoothed sparse constraint; (2) Lead field matrix is reduced by prior information of cardiac magnetic field map to reduce under-determination; (3) Spatiotemporal sources are reconstructed for non-invasive cardiac electrical activity imaging. The results of real MCG data demonstrate the effectiveness of this method for cardiac electrical activity imaging. The temporal and spatial changes of the current sources are similar to the depolarization and repolarization process of the ventricle.

Assessing heart disease using a novel magnetocardiography device

The aim of this paper is to present the use of a portable, unshielded magnetocardiograph (MCG) and identify key characteristics of MCG scans that could be used in future studies to identify parameters that are sensitive to cardiac pathology. We recruited 50 patients with confirmed myocardial infarction (MI) within the past 12 weeks and 46 volunteers with no history of cardiac disease. A set of 38 parameters were extracted from MCG features including both signals from the sensor array and from magnetic images obtained from the device and principal component analysis was used to concentrate the information contained in these parameters into uncorrelated predictors. Linear fits of these parameters were then used to examine the ability of MCG to distinguish between sub-groups of patients. In the fist instance, the primary aim of this study was to ensure that MCG has a basic ability to separate a highly polarised patient group (young controls from post infarction patients) and to identify parameters that could be used in future studies to build a formal diagnostic tool kit. Parameters that parameterised left ventricular ejection fraction (LVEF) were identified and an example is presented to show differential low and high ejection fractions.

Diagnostic accuracy of the magnetocardiograph for patients with suspected acute coronary syndrome

Background

We aimed to estimate the diagnostic accuracy of the VitalScan magnetocardiograph (MCG) for suspected acute coronary syndrome (ACS).

Methods

We undertook a prospective cohort study evaluating the diagnostic accuracy of the MCG in adults with suspected ACS. The reference standard of ACS was determined by an independent adjudication committee based on 30-day investigations and events. The cohort was split into a training sample, to derive the MCG algorithm and an algorithm combining MCG with a modified Manchester Acute Coronary Syndrome (MACS) clinical probability score, and a validation sample, to estimate diagnostic accuracy.

Results

We recruited 756 participants and analysed data from 680 (293 training, 387 validation), of whom 96 (14%) had ACS. In the training sample, the respective area under the receiver operating characteristic (AUROC) curves were the following: MCG 0.66 (95% CI 0.58 to 0.74), MACS 0.64 (95% CI 0.54 to 0.73) and MCG+MACS 0.70 (95% CI 0.63 to 0.77). MCG specificity was 0.16 (95% CI 0.12 to 0.21) at the threshold achieving acceptable sensitivity for rule-out (>0.98). In the validation sample (n=387), the respective AUROCs were the following: MCG 0.56 (95% CI 0.48 to 0.64), MACS 0.69 (95% CI 0.61 to 0.77) and MCG+MACS 0.64 (95% CI 0.56 to 0.72). MCG sensitivity was 0.89 (95% CI 0.77 to 0.95) and specificity 0.15 (95% CI 0.12 to 0.20) at the rule-out threshold. MCG+MACS sensitivity was 0.85 (95% CI 0.73 to 0.92) and specificity 0.30 (95% CI 0.25 to 0.35).

Conclusion

The VitalScan MCG is currently unable to accurately rule out ACS and is not yet ready for use in clinical practice. Further developmental research is required.

Predictive value of unshielded magnetocardiographic mapping to differentiate atrial fibrillation patients from healthy subjects

BACKGROUND

P-wave duration, its dispersion and signal-averaged ECG, are currently used markers of vulnerability to atrial fibrillation (AF). However, since tangential atrial currents are better detectable at the body surface as magnetic than electric signals, we investigated the accuracy of magnetocardiographic mapping (MCG), recorded in unshielded clinical environments, as predictor of AF occurrence.

METHODS

MCG recordings, in sinus rhythm (SR), of 71 AF patients and 75 controls were retrospectively analyzed. Beside electric and magnetic P-wave and PR interval duration, two MCG P-wave subintervals, defined P-dep and P-rep, were measured, basing on the point of inversion of atrial magnetic field (MF). Eight parameters were calculated from inverse solution with "Effective Magnetic Dipole (EMD) model" and 5 from "MF Extrema" analysis. Discriminant analysis (DA) was used to assess MCG predictive accuracy to differentiate AF patients from controls.

RESULTS

All but one (P-rep) intervals were significantly longer in AF patients. At univariate analysis, three EMD parameters differed significantly: in AF patients, the dipole-angle-elevation angular speed was lower during P-dep (p < 0.05) and higher during P-rep (p < 0.001) intervals. The space-trajectory during P-rep and the angle-dynamics during P-dep were higher (p < 0.05), whereas ratio-dynamics P-dep was lower (p < 0.01), in AF. At DA, with a combination of MCG and clinical parameters, 81.5% accuracy in differentiating AF patients from controls was achieved. At Cox-regression, the angle-dynamics P-dep was an independent predictor of AF recurrences (p = 0.037).

CONCLUSIONS

Quantitative analysis of atrial MF dynamics in SR and the solution of the inverse problem provide new sensitive markers of vulnerability to AF.