Symmetry difference electrical impedance tomography-a novel modality for anomaly detection

Exploratory study of a multifrequency EIT-based method for detecting intracranial abnormalities

Objective

The purpose of this paper is to compare the differences in the features of multifrequency electrical impedance tomography (MFEIT) images of human heads between healthy subjects and patients with brain diseases and to explore the possibility of applying MFEIT to intracranial abnormality detection.

Methods

Sixteen healthy volunteers and 8 patients with brain diseases were recruited as subjects, and the cerebral MFEIT data of 9 frequencies in the range of 21 kHz - 100 kHz of all subjects were acquired with an MFEIT system. MFEIT image sequences were obtained according to certain imaging algorithms, and the area ratio of the ROI (AR_ROI) and the mean value of the reconstructed resistivity change of the ROI (MVRRC_ROI) on both the left and right sides of these images were extracted. The geometric asymmetry index (GAI) and intensity asymmetry index (IAI) were further proposed to characterize the symmetry of MFEIT images based on the extracted indices and to statistically compare and analyze the differences between the two groups of subjects on MFEIT images.

Results

There were no significant differences in either the AR_ROI or the MVRRC_ROI between the two sides of the brains of healthy volunteers (p > 0.05); some of the MFEIT images mainly in the range of 30 kHz - 60 kHz of patients with brain diseases showed stronger resistivity distributions (larger area or stronger signal) that were approximately symmetric with the location of the lesions. However, statistical analysis showed that the AR_ROI and the MVRRC_ROI on the healthy sides of MFEIT images of patients with unilateral brain disease were not significantly different from those on the affected side (p > 0.05). The GAI and IAI were higher in all patients with brain diseases than in healthy volunteers except for 80 kHz (p < 0.05).

Conclusion

There were significant differences in the geometric symmetry and the signal intensity symmetry of the reconstructed targets in the MFEIT images between healthy volunteers and patients with brain diseases, and the above findings provide a reference for the rapid detection of intracranial abnormalities using MFEIT images and may provide a basis for further exploration of MFEIT for the detection of brain diseases.

Multi-frequency symmetry difference electrical impedance tomography with machine learning for human stroke diagnosis

OBJECTIVE

Multi-frequency symmetry difference electrical impedance tomography (MFSD-EIT) can robustly detect and identify unilateral perturbations in symmetric scenes. Here, an investigation is performed to assess if the algorithm can be successfully applied to identify the aetiology of stroke with the aid of machine learning.

METHODS

Anatomically realistic four-layer finite element method models of the head based on stroke patient images are developed and used to generate EIT data over a 5 Hz-100 Hz frequency range with and without bleed and clot lesions present. Reconstruction generates conductivity maps of each head at each frequency. Application of a quantitative metric assessing changes in symmetry across the sagittal plane of the reconstructed image and over the frequency range allows lesion detection and identification. The algorithm is applied to both simulated and human (n = 34 subjects) data. A classification algorithm is applied to the metric value in order to differentiate between normal, haemorrhage and clot values.

MAIN RESULTS

An average accuracy of 85% is achieved when MFSD-EIT with support vector machines (SVM) classification is used to identify and differentiate bleed from clot in human data, with 77% accuracy when differentiating normal from stroke in human data.

CONCLUSION

Applying a classification algorithm to metrics derived from MFSD-EIT images is a novel and promising technique for detection and identification of perturbations in static scenes.

SIGNIFICANCE

The MFSD-EIT algorithm used with machine learning gives promising results of lesion detection and identification in challenging conditions like stroke. The results imply feasible translation to human patients.

Bi-frequency symmetry difference electrical impedance tomography-a novel technique for perturbation detection in static scenes

OBJECTIVE

A novel method for the imaging of static scenes using electrical impedance tomography (EIT) is reported with implementation and validation using numerical and phantom models. The technique is applicable to regions featuring symmetry in the normal case, asymmetry in the presence of a perturbation, and where there is a known, frequency-dependent change in the electrical conductivity of the materials in the region.

APPROACH

The stroke diagnostic problem is used as a motivating sample application. The head is largely symmetrical across the sagittal plane. A haemorrhagic or ischaemic lesion located away from the sagittal plane will alter this natural symmetry, resulting in a symmetrical imbalance that can be detected using EIT. Specifically, application of EIT stimulation and measurement protocols at two distinct frequencies detects deviations in symmetry if an asymmetrically positioned lesion is present, with subsequent identification and localisation of the perturbation based on known frequency-dependent conductivity changes. Anatomically accurate computational models are used to demonstrate the feasibility of the proposed technique using different types, sizes, and locations of lesions with frequency-dependent (or independent) conductivity. Further, a realistic experimental head phantom is used to validate the technique using frequency-dependent perturbations emulating the key numerical simulations.

MAIN RESULTS

Lesion presence, type, and location are detectable using this novel technique. Results are presented in the form of images and corresponding robust quantitative metrics. Better detection is achieved for larger lesions, those further from the sagittal plane, and when measurements have a higher signal-to-noise ratio.

SIGNIFICANCE

Bi-frequency symmetry difference EIT is an exciting new modality of EIT with the ability to detect deviations in the symmetry of a region that occur due to the presence of a lesion. Notably, this modality does not require a time change in the region and thus may be used in static scenarios such as stroke detection.