Editorial for the Special Issue "Sensing Brain Activity Using EEG and Machine Learning".

Sensing brain activity to reveal, analyze and recognize brain activity patterns has become a topic of great interest and ongoing research [...].

Integrating EEG and Machine Learning to Analyze Brain Changes during the Rehabilitation of Broca's Aphasia.

The fusion of electroencephalography (EEG) with machine learning is transforming rehabilitation. Our study introduces a neural network model proficient in distinguishing pre- and post-rehabilitation states in patients with Broca's aphasia, based on brain connectivity metrics derived from EEG recordings during verbal and spatial working memory tasks. The Granger causality (GC), phase-locking value (PLV), weighted phase-lag index (wPLI), mutual information (MI), and complex Pearson correlation coefficient (CPCC) across the delta, theta, and low- and high-gamma bands were used (excluding GC, which spanned the entire frequency spectrum). Across eight participants, employing leave-one-out validation for each, we evaluated the intersubject prediction accuracy across all connectivity methods and frequency bands. GC, MI theta, and PLV low-gamma emerged as the top performers, achieving 89.4%, 85.8%, and 82.7% accuracy in classifying verbal working memory task data. Intriguingly, measures designed to eliminate volume conduction exhibited the poorest performance in predicting rehabilitation-induced brain changes. This observation, coupled with variations in model performance across frequency bands, implies that different connectivity measures capture distinct brain processes involved in rehabilitation. The results of this paper contribute to current knowledge by presenting a clear strategy of utilizing limited data to achieve valid and meaningful results of machine learning on post-stroke rehabilitation EEG data, and they show that the differences in classification accuracy likely reflect distinct brain processes underlying rehabilitation after stroke.

Dynamic Connectivity Analysis Using Adaptive Window Size.

In this paper, we propose a new method to study and evaluate the time-varying brain network dynamics. The proposed RICI-imCPCC method (relative intersection of confidence intervals for the imaginary component of the complex Pearson correlation coefficient) is based on an adaptive window size and the imaginary part of the complex Pearson correlation coefficient. It reduces the weaknesses of the existing method of constant sliding window analysis with narrow and wide windows. These are the low temporal precision and low reliability for short connectivity periods for wide windows, and high susceptibility to noise for narrow windows, all resulting in low estimation accuracy. The proposed method overcomes these shortcomings by dynamically adjusting the window width using the RICI rule, which is based on the statistical properties of the area around the observed sample. In this paper, we compare the RICI-imCPCC with the existing constant sliding window analysis method and describe its advantages. First, the mathematical principles are established. Then, the comparison between the existing and the proposed method using synthetic and real electroencephalography (EEG) data is presented. The results show that the proposed RICI-imCPCC method has improved temporal resolution and estimation accuracy compared to the existing method and is less affected by the noise. The estimation error energy calculated for the RICI-imCPCC method on synthetic signals was lower by a factor of 1.22 compared to the error of the constant sliding window analysis using narrow window size imCPCC, by a factor of 2.87 compared to using wide window size imCPCC, by a factor of 6.69 compared to using narrow window size wPLI, and by a factor of 4.72 compared to using wide window size wPLI. Analysis of the real signals shows the ability of the proposed method to detect a P300 response and to detect a decrease in dynamic connectivity due to desynchronization and blockage of mu-rhythms.

Complex Pearson Correlation Coefficient for EEG Connectivity Analysis.

In the background of all human thinking-acting and reacting are sets of connections between different neurons or groups of neurons. We studied and evaluated these connections using electroencephalography (EEG) brain signals. In this paper, we propose the use of the complex Pearson correlation coefficient (CPCC), which provides information on connectivity with and without consideration of the volume conduction effect. Although the Pearson correlation coefficient is a widely accepted measure of the statistical relationships between random variables and the relationships between signals, it is not being used for EEG data analysis. Its meaning for EEG is not straightforward and rarely well understood. In this work, we compare it to the most commonly used undirected connectivity analysis methods, which are phase locking value (PLV) and weighted phase lag index (wPLI). First, the relationship between the measures is shown analytically. Then, it is illustrated by a practical comparison using synthetic and real EEG data. The relationships between the observed connectivity measures are described in terms of the correlation values between them, which are, for the absolute values of CPCC and PLV, not lower that 0.97, and for the imaginary component of CPCC and wPLI-not lower than 0.92, for all observed frequency bands. Results show that the CPCC includes information of both other measures balanced in a single complex-numbered index.

Image registration in dynamic renal MRI-current status and prospects.

Magnetic resonance imaging (MRI) modalities have achieved an increasingly important role in the clinical work-up of chronic kidney diseases (CKD). This comprises among others assessment of hemodynamic parameters by arterial spin labeling (ASL) or dynamic contrast-enhanced (DCE-) MRI. Especially in the latter, images or volumes of the kidney are acquired over time for up to several minutes. Therefore, they are hampered by motion, e.g., by pulsation, peristaltic, or breathing motion. This motion can hinder subsequent image analysis to estimate hemodynamic parameters like renal blood flow or glomerular filtration rate (GFR). To overcome motion artifacts in time-resolved renal MRI, a wide range of strategies have been proposed. Renal image registration approaches could be grouped into (1) image acquisition techniques, (2) post-processing methods, or (3) a combination of image acquisition and post-processing approaches. Despite decades of progress, the translation in clinical practice is still missing. The aim of the present article is to discuss the existing literature on renal image registration techniques and show today's limitations of the proposed techniques that hinder clinical translation. This paper includes transformation, criterion function, and search types as traditional components and emerging registration technologies based on deep learning. The current trend points towards faster registrations and more accurate results. However, a standardized evaluation of image registration in renal MRI is still missing.

Correction to: Image registration in dynamic renal MRI-current status and prospects.

The article Image registration in dynamic renal MRI-current status and prospects, written by Frank G. Zöllner, Amira Serifovic­Trbalic, Gordian Kabelitz, Marek Kocinski, Andrzej Materka and Peter Rogelj, was originally published electronically on the publisher's internet portal on 9 October 2019 without open access.With the author(s)' decision to opt for Open Choice the copyright of the article changed on 24 April 2020 to ©.

Distance deviation measure of contouring variability.

BACKGROUND: Several methods that are currently used for contouring analysis have problems providing reliable and/or meaningful results. In this paper a solution to these problems is proposed in a form of a novel measure, which was developed based on requirements defined for contouring studies. MATERIALS AND METHODS: The proposed distance deviation measure can be understood as an extension of the closest point measures in such a way that it does not measure only distances between points on contours but rather analyse deviation of distances to both/all contours from each image point/voxel. The obtained result is information rich, reliable and provided in a form of an image, enabling detailed topographic analysis. In addition to image representation, results can be further processed into angular representation for compact topographic analysis or into overall scalar estimates for quick assessment of contour disagreement. RESULTS: Distance deviation method is demonstrated on a multi observer contouring example with complex contour shapes, i.e., with pronounced extremes and void interior. The results are presented using the three proposed methods. CONCLUSIONS: The proposed method can detect and measure contour variation irrespective of contour complexity and number of contour segments, while the obtained results are easy to interpret. It can be used in various situations, regarding the presence of reference contour or multiple test contours.

Assessment of Model Accuracy in Eyes Open and Closed EEG Data: Effect of Data Pre-Processing and Validation Methods.

Eyes open and eyes closed data is often used to validate novel human brain activity classification methods. The cross-validation of models trained on minimally preprocessed data is frequently utilized, regardless of electroencephalography data comprised of data resulting from muscle activity and environmental noise, affecting classification accuracy. Moreover, electroencephalography data of a single subject is often divided into smaller parts, due to limited availability of large datasets. The most frequently used method for model validation is cross-validation, even though the results may be affected by overfitting to the specifics of brain activity of limited subjects. To test the effects of preprocessing and classifier validation on classification accuracy, we tested fourteen classification algorithms implemented in WEKA and MATLAB, tested on comprehensively and simply preprocessed electroencephalography data. Hold-out and cross-validation were used to compare the classification accuracy of eyes open and closed data. The data of 50 subjects, with four minutes of data with eyes closed and open each was used. The algorithms trained on simply preprocessed data were superior to the ones trained on comprehensively preprocessed data in cross-validation testing. The reverse was true when hold-out accuracy was examined. Significant increases in hold-out accuracy were observed if the data of different subjects was not strictly separated between the test and training datasets, showing the presence of overfitting. The results show that comprehensive data preprocessing can be advantageous for subject invariant classification, while higher subject-specific accuracy can be attained with simple preprocessing. Researchers should thus state the final intended use of their classifier.

Symmetric image registration.

This paper presents an original non-rigid image registration approach, which tends to improve the registration by establishing a symmetric image interdependence. In order to gather more information about the image transformation it measures the image similarity in both registration directions. The presented solution is based on the interaction between the images involved in the registration process. Images interact through forces, which according to Newton's action-reaction law form a symmetric relationship. These forces may transform both of the images, although in our implementation one of the images remains fixed. The experiments performed to demonstrate the advantages of the symmetric registration approach involve the registration of simple objects, the recovery of synthetic deformation, and the inter-patient registration of real images of the head. The results show that the symmetric approach improves both the registration consistency and the registration correctness.

Voxelwise mapping of magnetic resonance ventilation-perfusion ratio in a porcine model by multimodality registration: technical note.

RATIONALE AND OBJECTIVES: The purpose of this study was to demonstrate the construction of voxelwise ventilation-perfusion (V/Q) ratio maps in a porcine model by nonrigidly aligning the respective ventilation and perfusion images using a multimodality registration algorithm. MATERIALS AND METHODS: The first-pass contrast agent technique for a blood flow map and 3He used for ventilation imaging were performed using a normal porcine model. The registered 3He-ventilation image was then aligned to the blood flow map using a multimodality registration algorithm. The voxelwise V/Q ratios were calculated by dividing the registered 3He-ventilation image by the blood flow map. The V/Q ratios were also semi-logarithmically scatter-plotted against the number of voxels. RESULTS: From perfusion magnetic resonance images, a voxel-by-voxel blood flow map was produced. Registered 3He ventilation image was successfully obtained as well as V/Q ratio map. Plots of the V/Q ratios obtained by this registration approach were similar to the logarithmic normal distribution. CONCLUSION: Registration of MR perfusion and ventilation images can potentially enable quantitative evaluation of regional pulmonary function and thus yield deeper insight into the physiology and pathophysiology of the lung.

Uncertainties of target volume delineation in MRI guided adaptive brachytherapy of cervix cancer: a multi-institutional study.

BACKGROUND AND AIM: We aimed to quantify target volume delineation uncertainties in cervix cancer image guided adaptive brachytherapy (IGABT). MATERIALS AND METHODS: Ten radiation oncologists delineated gross tumour volume (GTV), high- and intermediate-risk clinical target volume (HR CTV, IR CTV) in six patients. Their contours were compared with two reference delineations (STAPLE-Simultaneous Truth and Performance Level Estimation and EC- expert consensus) by calculating volumetric and planar conformity index (VCI and PCI) and inter-delineation distances (IDD). RESULTS: VCISTAPLE and VCIEC were 0.76 and 0.72 for HR CTV, 0.77 and 0.68 for IR CTV and 0.59 and 0.58 for GTV. Variation was most prominent caudally and cranially in all target volumes and posterolaterally in IR CTV. IDDSTAPLE and IDDEC for HR CTV (3.6±3.5 and 3.8±3.4 mm) were significantly lower than for GTV (4.8±4.2 and 4.2±3.5 mm) and IR CTV (4.7±5.2 and 5.2±5.6 mm) (p<0.05). CONCLUSIONS: Due to lower delineation uncertainties when compared to GTV and IR CTV, HR CTV may be considered most robust volume for dose prescription and optimization in cervix cancer IGABT. Adequate imaging, training and use of contouring recommendations are main strategies to minimize delineation uncertainties.

Comparison of 3D MRI with high sampling efficiency and 2D multiplanar MRI for contouring in cervix cancer brachytherapy.

BACKGROUND: MRI sequences with short scanning times may improve accessibility of image guided adaptive brachytherapy (IGABT) of cervix cancer. We assessed the value of 3D MRI for contouring by comparing it to 2D multi-planar MRI. PATIENTS AND METHODS: In 14 patients, 2D and 3D pelvic MRI were obtained at IGABT. High risk clinical target volume (HR CTV) was delineated by 2 experienced radiation oncologists, using the conventional (2D MRI-based) and test (3D MRI-based) approach. The value of 3D MRI for contouring was evaluated by using the inter-approach and inter-observer analysis of volumetric and topographic contouring uncertainties. To assess the magnitude of deviation from the conventional approach when using the test approach, the inter-approach analysis of contouring uncertainties was carried out for both observers. In addition, to assess reliability of 3D MRI for contouring, the impact of contouring approach on the magnitude of inter-observer delineation uncertainties was analysed. RESULTS: No approach- or observer - specific differences in HR CTV sizes, volume overlap, or distances between contours were identified. When averaged over all delineated slices, the distances between contours in the inter-approach analysis were 2.6 (Standard deviation (SD) 0.4) mm and 2.8 (0.7) mm for observers 1 and 2, respectively. The magnitude of topographic and volumetric inter-observer contouring uncertainties, as obtained on the conventional approach, was maintained on the test approach. This variation was comparable to the inter-approach uncertainties with distances between contours of 3.1 (SD 0.8) and 3.0 (SD 0.7) mm on conventional and test approach, respectively. Variation was most pronounced at caudal HR CTV levels in both approaches and observers. CONCLUSIONS: 3D MRI could potentially replace multiplanar 2D MRI in cervix cancer IGABT, shortening the overall MRI scanning time and facilitating the contouring process, thus making this treatment method more widely employed.

Assessment of 3D DCE-MRI of the kidneys using non-rigid image registration and segmentation of voxel time courses.

We have applied automated image analysis methods in the assessment of human kidney perfusion based on 3D dynamic contrast-enhanced MRI data. This approach consists of non-rigid 3D image registration of the moving kidney followed by k-means clustering of the voxel time courses with split between left and right kidney. This method was applied to four data sets acquired from healthy volunteers, using 1.5 T (2 exams) and 3 T scanners (2 exams). The proposed registration method reduced motion artifacts in the image time series and improved further analysis of the DCE-MRI data. The subsequent clustering to segment the kidney compartments was in agreement with manually delineations (similarity score of 0.96) in the same motion corrected images. The resulting mean intensity time curves clearly show the successive transition of contrast agent through kidney compartments (cortex, medulla, and pelvis). The proposed method for motion correction and kidney compartment segmentation might improve the validity and usefulness of further model-based pharmacokinetic analysis of kidney function in patients.