Hannu Eskola

Sensitivity distributions of EEG and MEG measurements

It is generally believed that because the skull has low conductivity to electric current but is transparent to magnetic fields, the measurement sensitivity of the magnetoencephalography (MEG) in the brain region should be more concentrated than that of the electroencephalography (EEG). It is also believed that the information recorded by these techniques is very different. If this were indeed the case, it might be possible to justify the cost of MEG instrumentation which is at least 25 times higher than that of EEG instrumentation. The localization of measurement sensitivity using these techniques was evaluated quantitatively in an inhomogeneous spherical head model using a new concept called half-sensitivity volume (HSV). It is shown that the planar gradiometer has a far smaller HSV than the axial gradiometer. However, using the EEG it is possible to achieve even smaller HSVs than with whole-head planar gradiometer MEG devices. The micro-superconducting quantum interference device (SQUID) MEG device does have HSVs comparable to those of the EEG. The sensitivity distribution of planar gradiometers, however, closely resembles that of dipolar EEG leads and, therefore, the MEG and EEG record the electric activity of the brain in a very similar way.

Conductivity of living intracranial tissues

Resistivity values were measured from living human brain tissue in nine patients. A monopolar needle electrode was used with a measurement frequency of 50 kHz. Mean values were 3.51 Ohms m for grey matter and 3.91 Ohms m for white matter. Cerebrospiral fluid had a mean value of 0.80 Ohms m. Values for tumour tissues were dependent on the type of tumour and ranged from 2.30 to 9.70 Ohms m.

Characterization of breast cancer types by texture analysis of magnetic resonance images.

RATIONALE AND OBJECTIVES This novel study aims to investigate texture parameters in distinguishing healthy breast tissue and breast cancer in breast magnetic resonance imaging (MRI). A specific aim was to identify possible differences in the texture characteristics of histological types (lobular and ductal) of invasive breast cancer and to determine the value of these differences for computer-assisted lesion classification. MATERIALS AND METHODS Twenty patients (mean age 50.6 + or - SD 10.6; range 37-70 years), with histopathologically proven invasive breast cancer (10 lobular and 10 ductal) were included in this preliminary study. The median MRI lesion size was 25 mm (range, 7-60 mm). The selected T1-weighted precontrast, post-contrast, and subtracted images were analyzed and classified with texture analysis (TA) software MaZda and additional statistical tests were used for testing the parameters separability. RESULTS All classification methods employed were able to differentiate between cancer and healthy breast tissue and also invasive lobular and ductal carcinoma with classification accuracy varying between 80% and 100%, depending on the used imaging series and the type of region of interest. We found several parameters to be significantly different between the regions of interest studied. The co-occurrence matrix based parameters proved to be superior to other texture parameters used. CONCLUSIONS The results of this study indicate that MRI TA differentiates breast cancer from normal tissue and may be able to distinguish between two histological types of breast cancer providing more accurate characterization of breast lesions thereby offering a new tool for radiological analysis of breast MRI.

Semi-automatic tool for segmentation and volumetric analysis of medical images

Segmentation software is described, developed for medical image processing and run on Windows. The software applies basic image processing techniques through a graphical user interface. For particular applications, such as brain lesion segmentation, the software enables the combination of different segmentation techniques to improve its efficiency. The program is applied for magnetic resonance imaging, computed tomography and optical images of cryosections. The software can be utilised in numerous applications, including pre-processing for three-dimensional presentations, volumetric analysis and construction of volume conductor models.

Effect of EEG electrode density on dipole localization accuracy using two realistically shaped skull resistivity models.

The effect of number of EEG electrodes on the dipole localization was studied by comparing the results obtained using the 10-20 and 10-10 electrode systems. Two anatomically detailed models with resistivity values of 177.6 Ωm and 67.0 Ωm for the skull were applied. Simulated potential values generated by current dipoles were applied to different combinations of the volume conductors and electrode systems. High and low resistivity models differed slightly in favour of the lower skull resistivity model when dipole localization was based on noiseless data. The localization errors were approximately three times larger using low resistivity model for generating the potentials, but applying high resistivity model for the inverse solution. The difference between the two electrode systems was minor in favour of the 10-10 electrode system when simulated, noiseless potentials were used. In the presence of noise the dipole localization algorithm operated more accurately using the denser electrode system. In conclusion, increasing the number of recording electrodes seems to improve the localization accuracy in the presence of noise. The absolute skull resistivity value also affects the accuracy, but using an incorrect value in modelling calculations seems to be the most serious source of error.

Applicability of semi-automatic segmentation for volumetric analysis of brain lesions

This project involves the development of a fast semi-automatic segmentation procedure to make an accurate volumetric estimation of brain lesions. This method has been applied in the segmentation of demyelination plaques in Multiple Sclerosis (MS) and right cerebral hemispheric infarctions in patients with neglect. The developed segmentation method includes several image processing techniques, such as image enhancement, amplitude segmentation, and region growing. The entire program operates on a PC-based computer and applies graphical user interfaces. Twenty three patients with MS and 43 patients with right cerebral hemisphere infarctions were studied on a 0.5 T MRI unit. The MS plaques and cerebral infarctions were thereafter segmented. The volumetric accuracy of the program was demonstrated by segmenting Magnetic Resonance (MR) images of fluid filled syringes. The relative error of the total volume measurement based on the MR images of syringes was 1.5%. Also the repeatability test was carried out as inter-and intra-observer study in which MS plaques of six randomly selected patients were segmented. These tests indicated 7% variability in the inter-observer study and 4% variability in the intra-observer study. Average time used to segment and calculate the total plaque volumes for one patient was 10 min. This simple segmentation method can be utilized in the quantitation of anatomical structures, such as air cells in the sinonasal and temporal bone area, as well as in different pathological conditions, such as brain tumours, intracerebral haematomas and bony destructions.

Acute mild traumatic brain injury is not associated with white matter change on diffusion tensor imaging

This study was designed to (i) evaluate the influence of age on diffusion tensor imaging measures of white matter assessed using tract-based spatial statistics; (ii) determine if mild traumatic brain injury is associated with microstructural changes in white matter, in the acute phase following injury, in a large homogenous sample that was carefully screened for pre-injury medical, psychiatric, or neurological problems; and (iii) examine if injury severity is related to white matter changes. Participants were 75 patients with acute mild traumatic brain injury (age = 37.2 ± 12.0 years, 45 males and 30 females) and 40 controls (age = 40.6 ± 12.2 yrs, 20 males and 20 females). Age effects were analysed by comparing control subgroups aged 31-40, 41-50, and 51-60 years against a group of 18-30-year-old control subjects. Widespread statistically significant areas of abnormal diffusion tensor measures were observed in older groups. Patients and controls were compared using age and gender as covariates and in age- and gender-matched subgroups. Subgroups of patients with more severe injuries were compared to age-and gender-matched controls. No significant differences were detected in patient-control or severity analyses (all P-value > 0.01). In this large, carefully screened sample, acute mild traumatic brain injury was not associated with diffusion tensor imaging abnormalities detectable with tract-based spatial statistics.

Segmentation of T1 MR scans for reconstruction of resistive head models.

This paper describes a segmentation method primarily developed for reconstructing resistive head models for electroencephalographic modelling purposes. The method was implemented by combining several image processing techniques, such as amplitude segmentation, region growing, and image fusion. Also a graphical user interface was developed to enable semiautomatic approach to the segmentation process. This method was developed especially for segmentation of the brain and skull from T1-weighted magnetic resonance images, but can also be applied in any segmentation procedure. The entire project was implemented successfully in a PC-based computer running the Unix/NeXTstep operating system.

Brain Iron Deposition and Sequence Characteristics in Parkinsonism: Comparison of SWI, T2* Maps, T2-Weighted-, and FLAIR-SPACE

Objectives:To compare quantitatively T2- and T2*-based magnetic resonance imaging sequences in patients with symptoms of Parkinson disease and to evaluate the information content of those sequences regarding brain iron concentration. Materials and Methods:We imaged 51 patients with symptoms of Parkinson disease on 3-T magnetic resonance imaging with T2-weighted sampling perfection with application optimized contrasts using different flip-angle evolution (SPACE), fluid attenuation inversion recovery (FLAIR)-SPACE, susceptibility-weighted imaging (SWI), and parametric T2* sequence (MapIt). Signal analysis was performed in 22 regions of interest in the brain. Results:Correlations (r2 = 0.82…0.96) with brain iron concentration were excellent. Contrast and tissue separability ratios were best in the T2* maps and FLAIR-SPACE, respectively. Good correlations of contrast were reached between SWI and both T2-weighted SPACE and FLAIR-SPACE. Their relation to quantitative T2* values was reminiscent of a quadratic curve shape. However, separation into gray and white matter revealed a linear positive and negative correlation, respectively. Conclusions:SWI showed potential in differentiating illnesses characterized by brain iron deposition. Closely similar information was given by T2-weighted SPACE and FLAIR-SPACE, whereas other sequence comparisons revealed dispersion from intersequence agreement.

Mild Traumatic Brain Injury: Tissue Texture Analysis Correlated to Neuropsychological and DTI Findings

RATIONALE AND OBJECTIVES The aim of this study was to evaluate whether texture analysis (TA) can detect subtle changes in cerebral tissue caused by mild traumatic brain injury (MTBI) and to determine whether these changes correlate with neuropsychological and diffusion tensor imaging (DTI) findings. MATERIALS AND METHODS Forty-two patients with MTBIs were imaged using 1.5T magnetic resonance imaging within 3 weeks after head injury. TA was performed for the regions corresponding to the mesencephalon, centrum semiovale, and corpus callosum. Using DTI, the fractional anisotropic and apparent diffusion coefficient values for the same regions were evaluated. The same analyses were performed on a group of 10 healthy volunteers. Patients also underwent a battery of neurocognitive tests within 6 weeks after injury. RESULTS TA revealed textural differences between the right and left hemispheres in patients with MTBIs, whereas differences were minimal in healthy controls. A significant correlation was found between scores on memory tests and texture parameters (sum of squares, sum entropy, inverse difference moment, and sum average) in patients in the area of the mesencephalon and the genu of the corpus callosum. Significant correlations were also found between texture parameters for the left mesencephalon and both fractional anisotropic and apparent diffusion coefficient values. CONCLUSIONS The data suggest that heterogeneous texture and abnormal DTI patterns in the area of the mesencephalon may be linked with verbal memory deficits among patients with MTBIs. Therefore, TA combined with DTI in patients with MTBIs may increase the ability to detect early and subtle neuropathologic changes.