Intelligent digital tools for screening of brain connectivity and dementia risk estimation in people affected by mild cognitive impairment: the AI-Mind clinical study protocol.

More than 10 million Europeans show signs of mild cognitive impairment (MCI), a transitional stage between normal brain aging and dementia stage memory disorder. The path MCI takes can be divergent; while some maintain stability or even revert to cognitive norms, alarmingly, up to half of the cases progress to dementia within 5 years. Current diagnostic practice lacks the necessary screening tools to identify those at risk of progression. The European patient experience often involves a long journey from the initial signs of MCI to the eventual diagnosis of dementia. The trajectory is far from ideal. Here, we introduce the AI-Mind project, a pioneering initiative with an innovative approach to early risk assessment through the implementation of advanced artificial intelligence (AI) on multimodal data. The cutting-edge AI-based tools developed in the project aim not only to accelerate the diagnostic process but also to deliver highly accurate predictions regarding an individual's risk of developing dementia when prevention and intervention may still be possible. AI-Mind is a European Research and Innovation Action (RIA H2020-SC1-BHC-06-2020, No. 964220) financed between 2021 and 2026. First, the AI-Mind Connector identifies dysfunctional brain networks based on high-density magneto- and electroencephalography (M/EEG) recordings. Second, the AI-Mind Predictor predicts dementia risk using data from the Connector, enriched with computerized cognitive tests, genetic and protein biomarkers, as well as sociodemographic and clinical variables. AI-Mind is integrated within a network of major European initiatives, including The Virtual Brain, The Virtual Epileptic Patient, and EBRAINS AISBL service for sensitive data, HealthDataCloud, where big patient data are generated for advancing digital and virtual twin technology development. AI-Mind's innovation lies not only in its early prediction of dementia risk, but it also enables a virtual laboratory scenario for hypothesis-driven personalized intervention research. This article introduces the background of the AI-Mind project and its clinical study protocol, setting the stage for future scientific contributions.

Noninvasive mapping reveals recurrent and suddenly changing patterns in atrial fibrillation-a magnetocardiographic study.

OBJECTIVE: To study noninvasive magnetocardiographic (MCG) mapping of ongoing atrial fibrillation (AF) and, for the possible mapping patterns observed, to develop simplified but meaningful descriptors or parameters, providing a possible basis for future research and clinical use of the mappings. APPROACH: MCG mapping with simultaneous ECG was recorded during arrhythmia in patients representing a range of typical, clinically classical atrial arrhythmias. The recordings were assessed using MCG map animations, and a method to compute magnetic field map orientation (MFO) and its time course was created to facilitate presentation of the findings. All the data were segmented into four categories of ECG waveform regularity. MAIN RESULTS: In visual observation of the MCG animations, an abundance of clear spatial and temporal patterns with regularity were found, often perceived as rotations of the map. This rotation and its sudden reversals of direction were distinctly present in the time course of the MFO. The shortest segments with consistent rotation lasted for some hundreds of milliseconds, i.e. a couple of cycles, but segments lasting for tens of seconds were observed as well. In the ECG, all four categories of regularity were present. The rotation of the MFO was observed in all patients under study and regardless of the ECG categories. Further, a change in ECG category during a measurement was frequently, but not always, found to be simultaneous with a change in the rotation pattern of the MFO. Utilization of spatial information of MCG mapping could enable detection of both regularities and instantaneous phenomena during AF. SIGNIFICANCE: Cardiac mapping may offer a useful noninvasive means to study the mechanisms of AF, including superior temporal resolution.

Investigations of sensitivity and resolution of ECG and MCG in a realistically shaped thorax model.

Solving the inverse problem of electrocardiography (ECG) and magnetocardiography (MCG) is often referred to as cardiac source imaging. Spatial properties of ECG and MCG as imaging systems are, however, not well known. In this modelling study, we investigate the sensitivity and point-spread function (PSF) of ECG, MCG, and combined ECG+MCG as a function of source position and orientation, globally around the ventricles: signal topographies are modelled using a realistically-shaped volume conductor model, and the inverse problem is solved using a distributed source model and linear source estimation with minimal use of prior information. The results show that the sensitivity depends not only on the modality but also on the location and orientation of the source and that the sensitivity distribution is clearly reflected in the PSF. MCG can better characterize tangential anterior sources (with respect to the heart surface), while ECG excels with normally-oriented and posterior sources. Compared to either modality used alone, the sensitivity of combined ECG+MCG is less dependent on source orientation per source location, leading to better source estimates. Thus, for maximal sensitivity and optimal source estimation, the electric and magnetic measurements should be combined.

Comparison of minimum-norm estimation and beamforming in electrocardiography with acute ischemia.

In the electrocardiographic (ECG) inverse problem, the electrical activity of the heart is estimated from measured electrocardiogram. A model of thorax conductivities and a model of the cardiac generator is required for the ECG inverse problem. Limitations and errors in methods, models, and data will lead to errors in the estimates. However, in experimental applications, the use of limited or erroneous models is often inevitable due to necessary model simplifications and the difficulty of obtaining accurate 3D anatomical imaging data. In this work, we focus on two methods for solving the inverse problem of ECG in the case of acute ischemia: minimum-norm (MN) estimation and linearly constrained minimum-variance beamforming. We study how these methods perform with different sizes of ischemia and with erroneous conductivity models. The results indicate that the beamformer can localize small ischemia given an accurate model, but it cannot be used for estimating the size of ischemia. The MN estimator is tolerant to geometry errors and excels in estimating the size of ischemia, although the beamformer performs better with accurate model and small ischemia.

Interatrial conduction in patients with paroxysmal atrial fibrillation and in healthy subjects.

BACKGROUND: Altered atrial conduction is linked to susceptibility to atrial fibrillation (AF). Whether signal propagation to left atrium (LA) during sinus rhythm differs between patients with paroxysmal lone AF and healthy subjects is not known. METHODS: In 107 patients with lone paroxysmal AF (age 45±12 years) and 94 controls 99-channel magnetocardiography (MCG) was recorded over anterior chest. The duration of the atrial wave (Pd) and the MCG maps over time intervals corresponding to early and later LA activations were determined. Based on magnetic field orientations in LA maps, MCG atrial waves were classified into 3 types which are related to distinct interatrial conduction routes: Type 1 to Bachmann bundle, Type 2 to margin of fossa ovalis or multisite, and Type 3 to coronary sinus ostial connections. RESULTS: Pd was longer in AF patients than in controls (112±13 vs. 104±13; p<0.001), which was most obvious in Type 1 wave (109±12 vs. 102±11 ms, p=0.003). The distribution of the atrial wave types differed between AF patients and controls: Type 1 occurred in 67% and Type 2 in 20% of controls whereas Type 1 occurred in 54% and Type 2 in 42% of AF patients, p<0.01 for difference. CONCLUSIONS: Susceptibility to paroxysmal lone AF is associated with propagation of atrial signal to LA via margin of fossa ovalis or multiple pathways. When conduction occurs via Bachmann bundle, it is related with prolonged atrial activation. Thus altered and alternative conduction pathways may contribute to pathogenesis of lone AF.

Reversal of atrial remodeling after cardioversion of persistent atrial fibrillation measured with magnetocardiography.

BACKGROUND: Atrial fibrillation (AF) causes electrical, functional, and structural changes in the atria. We examined electrophysiologic remodeling caused by AF and its reversal noninvasively by applying a new atrial signal analysis based on magnetocardiography (MCG). METHODS: In 26 patients with persistent AF, MCG, signal-averaged electrocardiography (SAECG), and echocardiography were performed immediately after electrical cardioversion (CV), and repeated after 1 month in 15 patients who remained in sinus rhythm (SR). Twenty-four matched subjects without history of AF served as controls. P-wave duration (Pd) and dispersion (standard deviation of Pd values in individual channels) and root mean square amplitudes of the P wave over the last 40 ms portions (RMS40) were determined. RESULTS: In MCG Pd was longer (122.8 +/- 18.2 ms vs 101.5 +/- 14.6 ms, P < 0.01) and RMS40 was higher (60.4 +/- 28.2 vs 46.9 +/- 19.1 fT) in AF patients immediately after CV as compared to the controls. In SAECG Pd dispersion was increased in AF patients. Mitral A-wave velocity and left atrial (LA) contraction were decreased and LA diameter was increased (all P < 0.01). After 1 month, Pd in MCG still remained longer and LA diameter greater (both P < 0.05), while RMS40 in MCG, Pd dispersion in SAECG, mitral A-wave velocity, and LA contraction were recovered. CONCLUSIONS: Magnetocardiographically detected atrial electrophysiologic alterations in persistent AF diminish rapidly although incompletely during maintained SR after CV. This might be related to the known early high and late lower, but still existent tendency to AF relapses.

Non-invasive detection of conduction pathways to left atrium using magnetocardiography: validation by intra-cardiac electroanatomic mapping.

AIMS: Alteration in conduction from right to left atrium (LA) is linked to susceptibility to atrial fibrillation (AF). We examined whether different inter-atrial conduction pathways can be identified non-invasively by magnetocardiographic mapping (MCG). METHODS AND RESULTS: In 27 patients undergoing catheter ablation of paroxysmal AF, LA activation sequence was determined during sinus rhythm using invasive electroanatomic mapping. Before this, 99-channel magnetocardiography was recorded over anterior chest. The orientation of the magnetic fields during the early (40-70 ms from P onset) and later part (last 50%) of LA depolarization was determined using pseudocurrent conversion. Breakthrough of electrical activation to LA occurred through Bachmann bundle (BB) in 14, margin of fossa ovalis (FO) in 3, coronary sinus ostial region (CS) in 2, and their combinations in 10 cases by invasive reference in total of 29 different P-waves. Based on the combination of pseudocurrent angles over early and late parts of LA activation, the MCG maps were divided to three types. These types correctly identified the LA breakthrough sites to BB, CS, FO, or their combinations in 27 of 29 (93%) cases. CONCLUSION: Magnetocardiographic mapping seems capable of distinguishing inter-atrial conduction pathways. Recognizing the inter-atrial conduction pattern may assist in understanding the pathogenesis of AF and identifying the subgroups for patient-tailored therapy.

High-resolution signal-averaged analysis of atrial electromagnetic characteristics in patients with paroxysmal lone atrial fibrillation.

BACKGROUND: Abnormalities in the electromagnetic signal of the atria during sinus rhythm could serve as markers of triggering foci or substrate for atrial fibrillation (AF). We examined atrial electrophysiologic properties noninvasively by using magnetocardiographic mapping (MCG) in patients with paroxysmal lone AF to find whether any difference exists between those who have frequent triggers of AF and who don't. METHODS: MCG was recorded over anterior chest during sinus rhythm in 80 patients with paroxysmal lone AF (44 +/- 12 years, 61 males) and 80 matched controls. Atrial wave duration (Pd) and root mean square amplitudes of the last 40 ms (RMS40) of the averaged filtered atrial complex were determined automatically. Patients expressing atrial arrhythmias triggering AF episodes were classified as focal AF. RESULTS: The Pd was 109 ms in patients and 104 ms in controls (P = 0.007). In focal AF (72%) the Pd was slightly prolonged and its proportion of the PR interval was larger, but RMS40 was normal compared to controls. In other patients, the Pd was close to controls, but the RMS40 was reduced (59 +/- 17 vs74 +/- 36 fT, P = 0.006). Pd and atrial RMS amplitudes were unrelated to duration of AF history or frequency of recurrences. CONCLUSION: Clinical subclasses of lone AF seem to possess distinct signal profiles of atrial depolarization. Differences in electrophysiological properties between these subclasses may reflect pathogenetic variation and could have implications on diagnostics and therapy.