ConvDip: A Convolutional Neural Network for Better EEG Source Imaging

The electroencephalography (EEG) is a well-established non-invasive method in neuroscientific research and clinical diagnostics. It provides a high temporal but low spatial resolution of brain activity. To gain insight about the spatial dynamics of the EEG, one has to solve the inverse problem, i.e., finding the neural sources that give rise to the recorded EEG activity. The inverse problem is ill-posed, which means that more than one configuration of neural sources can evoke one and the same distribution of EEG activity on the scalp. Artificial neural networks have been previously used successfully to find either one or two dipole sources. These approaches, however, have never solved the inverse problem in a distributed dipole model with more than two dipole sources. We present ConvDip, a novel convolutional neural network (CNN) architecture, that solves the EEG inverse problem in a distributed dipole model based on simulated EEG data. We show that (1) ConvDip learned to produce inverse solutions from a single time point of EEG data and (2) outperforms state-of-the-art methods on all focused performance measures. (3) It is more flexible when dealing with varying number of sources, produces less ghost sources and misses less real sources than the comparison methods. It produces plausible inverse solutions for real EEG recordings from human participants. (4) The trained network needs <40 ms for a single prediction. Our results qualify ConvDip as an efficient and easy-to-apply novel method for source localization in EEG data, with high relevance for clinical applications, e.g., in epileptology and real-time applications.

On the issue of the electroencephalographic phenomenon «burst-suppression»: variants of outcomes and possible neurophysiological mechanisms

Objective: to evaluate the indicators of electrical activity of the brain using frequency- spectral analysis and data of three- dimensional localization of sources of pathological activity for an approach to the analysis of possible neurophysiological mechanisms of the brain of patients whose EEG recorded the phenomenon of ‘burst- suppression’.Material and methods: 45 electroencephalograms recorded in 22 patients (average age 51.05; 11 women, 11 men) were analyzed. In 12 patients, the EEG study was performed in dynamics from 1 to 8 times. At the time of the first registration, the ‘burst- suppression’phenomenon was recorded in the EEG of all patients. The level of wakefulness of all patients, with the exception of patients who were under anesthesia, was 3 points on the Glasgow coma scale.EEG recording was performed on electroencephalographs ‘Encephalan- EEGR-19/26’, ‘Mitsar- EEG-10/70–201’, ‘Mitsar- EEG-SmartBCI’, ‘Neuron- Spectrum-5’and ‘Neuron- Spectrum-65’in accordance with the International scheme of arrangement of electrodes 10–20 %. A frequency- spectral analysis of the power of the ‘burst’and ‘suppression’periods was carried out — the fast Fourier transform method was used. The program ‘BrainLoc 6.1’(Russia) was used for localization of equivalent dipole sources of pathological electrical activity of the ‘burst’period.Results: during the first EEG recording, the ‘burst- suppression’phenomenon was recorded in all patients. In seven patients, the ‘burst’period in the ‘burstsuppression’phenomenon was visually represented by slow-wave oscillations, in 15 patients, the ‘burst’periods resembled epileptiform discharges. In frequency- spectral analysis EEG in all patients in the ‘burst’period, the dominance of the power of slow-wave oscillations (mainly in the delta range) was noted. According to the program ‘BrainLoc 6.1’, equivalent dipole sources of pathological activity of the ‘burst’period were recorded at the level of the thalamus, in the medio- basal parts of the frontal and temporal lobes on both sides. A favorable outcome of the ‘burst- suppression’phenomenon was observed in only five patients of 22, all other patients had an unfavorable outcome.Conclusion: a favorable outcome of the ‘burst- suppression’phenomenon was observed only in patients under sevorane anesthesia and in some patients after acute poisoning with drugs that affect the central nervous system, while patients after brain anoxia had an unfavorable outcome. In prognostic terms, our data are comparable to the literature data. The changes revealed during the frequency-spectral analysis of the EEG in the form of the dominance of the power of slow-wave oscillations (mainly in the delta range), as well as the localization of the supposed generators of electrical activity in the ‘burst’ period at the level of the thalamus, in the mediobasal parts of the frontal and temporal lobes (according to the ‘BrainLoc 6.1’program), may to some extent be consistent with the data of experimental works and mathematical models of the ‘burst–suppression’phenomenon If the ‘burst- suppression’ phenomenon is detected during EEG registration, it is advisableto conduct a dynamic EEG study or EEG monitoring.

Prediction of dipole sources and aeroacoustics field for tandem cylinder flow field based on DBEM/hybrid LES

In this paper, the DBEM/Hybrid LES（Directly Boundary Element Method/Hybrid Large Eddy Simulation）technique is applied to predict the aerodynamic noise generated by tandem circular cylinders immersed in a three-dimensional turbulent flow. Utilizing the Lighthill's Acoustic Analogy, the flow pressure fluctuation near the surface of the cylinder is converted into acoustic dipole sources. Taking the dipole sound sources as the actual sound sources, the aeroacoustic field is simulated and analyzed by DBEM. The research shows that: The strong dipole sources are distributed in the collision zone of the downstream cylindrical surface, where the upstream cylinder's shedding vortex colliding to downstream cylinder surface. Both of the amplitude-frequency response and the phase-frequency response of dipole acoustic source are obtained, which is helpful for further research on aerodynamics noise interference and suppression. Good comparisons are obtained between numerical results and BART (Basic Aerodynamic Research Tunnel) experimental data published by NASA.

A More Efficient Observation Way With Electric Dipole Sources

The Ex and Hz amplitude are monotonic functions of the resistivity, and numerical algorithm can be utilized to obtain the high-precision apparent resistivity for all frequencies. On this basis, we focus on the comparative analysis of the resolution for Ex field, Hz field, and Cagniard apparent resistivity to conductive and resistance targets, which proves that the Ex field may have a better resolution for the resistance bodies, and the Hz field can better identify the conductive target. Besides, the electromagnetic data is often distorted by the static effect, which seriously affects the application effect of the electromagnetic method. The influence of the static effect on the Ex field, Hz field, is also analyzed. The apparent resistivity based on the Ex field and Cagniard apparent resistivity are severely affected by the static effect which can cover deep anomalous bodies, but the apparent resistivity based on the Hz field is almost unaffected by the static effect. At last, a more efficient observation way is provided for the resistance and conductive targets, respectively.

Beamform processing for sonic imaging using monopole and dipole sources

New processing techniques are presented that enhance event signals for sonic imaging using monopole and dipole sources. The techniques use the azimuthally spaced receivers of a sonic logging tool. Sonic imaging, which is also known as borehole acoustic reflection surveys, uses a sonic logging tool in a fluid-filled borehole to image geologic structures. Signals from monopole and dipole sources are reflected from geologic interfaces and recorded by arrays of receivers of the same tool. Because the amplitudes of the event signals are very weak compared with the direct waves, borehole modes, and noise, the event signals are often difficult to extract. To enhance the weak event signals, beamforming techniques were developed to stack the waveforms from azimuthally spaced receivers of the tool for given azimuthal directions. For the incident P-waves from the monopole source, phase arrival times for the azimuthal receivers are time shifted for stacking using properties of wave propagation in the borehole. For the incident SH-waves from the dipole source, the signs of waveforms for the receivers are changed for specified azimuths. When the waveforms are stacked for the back azimuth of the event signals, the signal-to-noise ratio of the event signals is significantly improved because the event signals are enhanced whereas the direct waves are relatively smeared, and random noise is canceled. Therefore, the stacked waveforms also provide accurate back azimuths of the incident waves.

A Comparative Study on the Difference between the Multi-dipole Sources and Vector Synthesis Source

CSAMT exploration generally adopts a single dipole as the transmitter. The single dipole source has the apparent disadvantages–there are weak areas for all components, Ey and Hx are weak in the area where Ex and Hy are reliable. Moreover, it is hard to deploy the source with a specific direction in a rugged mountainous area. Given the shortcomings of the single dipole source, multi-dipole sources are introduced into CSAMT exploration. Although the dipole sources follow the principle of vector synthesis, the length of the source in actual exploration can last for several kilometers and the offset is generally a few kilometers. In this case, the source can no longer be regarded as a single dipole in the near-field zone. The electromagnetic field in this region becomes relatively complicated. We first compare the similarities and differences of electromagnetic field generated by vector synthesis source and multi-dipole source through the Ex radiation patterns. Then, we study the factors that affect electromagnetic response due to the substitution of the double-dipole source with the vector synthesis source. The measured EM fields is affected by the source length, frequency, the source angle, the offset, and the resistivity.Finally, we apply the double-dipole source to the 1D and 3D geological model and compare the difference between the electromagnetic field generated by the double-dipole source and that generated by the vector synthesis source. Usually, the difference is very obvious in the near-field zone, and is almost negligible in the far-field zone.

In Silico Study on Tumor-Size-Dependent Thermal Profiles inside an Anthropomorphic Female Breast Phantom Subjected to Multi-Dipole Antenna Array

Electromagnetic hyperthermia as a potent adjuvant for conventional cancer therapies can be considered valuable in modern oncology, as its task is to thermally destroy cancer cells exposed to high-frequency electromagnetic fields. Hyperthermia treatment planning based on computer in silico simulations has the potential to improve the localized heating of breast tissues through the use of the phased-array dipole applicators. Herein, we intended to improve our understanding of temperature estimation in an anatomically accurate female breast phantom embedded with a tumor, particularly when it is exposed to an eight-element dipole antenna matrix surrounding the breast tissues. The Maxwell equations coupled with the modified Pennes’ bioheat equation was solved in the modelled breast tissues using the finite-difference time-domain (FDTD) engine. The microwave (MW) applicators around the object were modelled with shortened half-wavelength dipole antennas operating at the same 1 GHz frequency, but with different input power and phases for the dipole sources. The total input power of an eight-dipole antenna matrix was set at 8 W so that the temperature in the breast tumor did not exceed 42 °C. Finding the optimal setting for each dipole antenna from the matrix was our primary objective. Such a procedure should form the basis of any successful hyperthermia treatment planning. We applied the algorithm of multi for multi-objective optimization for the power and phases for the dipole sources in terms of maximizing the specific absorption rate (SAR) parameter inside the breast tumor while minimizing this parameter in the healthy tissues. Electro-thermal simulations were performed for tumors of different radii to confirm the reliable operation of the given optimization procedure. In the next step, thermal profiles for tumors of various sizes were calculated for the optimal parameters of dipole sources. The computed results showed that larger tumors heated better than smaller tumors; however, the procedure worked well regardless of the tumor size. This verifies the effectiveness of the applied optimization method, regardless of the various stages of breast tumor development.

Solutions of P-SV and SV-P waves in single-well imaging through reciprocity relations

Single-well imaging (SWI) is a borehole measurement technique for detecting geologic structures outside a borehole by using arrays of receivers to record the waves reflected from these structures. The asymptotic solutions of P-P, SV-SV, and SH-SH waves in SWI have been obtained, but the P-SV and SV-P waves have been ignored in previous studies. It is necessary to know when these conversion waves are large and to estimate these waves when they are not ignorable. The analytical solutions of P-SV and SV-P waves are first derived using reciprocity relations between virtual concentrated forces and acoustic sources. The analytical results agree well with finite-difference solutions for monopole and dipole sources. We have found that, for a given source-receiver offset, the detected converted waves first increase and then decrease with the increase of the source-reflector distance. The relative amplitudes of converted waves to reflected waves are larger than 20% and cannot be ignored when the ratio of source-reflector distance to source-receiver offset is smaller than four. However, when the source-reflector distance is 10 or more times the source-receiver offset and the dip angle of the reflector is smaller than 10°, the amplitudes of the converted waves are smaller than 5% of the reflected waves. Furthermore, the relative amplitudes of converted waves increase with the dip angle of the reflector. The analytical solutions of converted waves are useful for simulating SWI wavefields and checking the inversion results in field logging.

Multiscale Modeling of EEG/MEG Response of a Compact Cluster of Tightly Spaced Pyramidal Neocortical Neurons

In this study, the boundary element fast multipole method or BEM-FMM is applied to model compact clusters of tightly spaced pyramidal neocortical neurons firing simultaneously and coupled with a high-resolution macroscopic head model. The algorithm is capable of processing a very large number of surface-based unknowns along with a virtually unlimited number of elementary microscopic current dipole sources distributed within the neuronal arbor.The realistic cluster size may be as large as 10,000 individual neurons, while the overall computation times do not exceed several minutes on a standard server. Using this approach, we attempt to establish how well the conventional lumped-dipole model used in electroencephalography/magnetoencephalography (EEG/MEG) analysis approximates a compact cluster of realistic neurons situated either in a gyrus (EEG response dominance) or in a sulcus (MEG response dominance).