This paper presents a new approach to investigate the spatial correlation of brain alpha activity in functional magnetic resonance imaging (fMRI) and electroencephalography (EEG). To avoid potential problems of simultaneous fMRI and EEG acquisitions in imaging brain alpha activity, data from each modality were acquired separately under a “three conditions” setup where one of the conditions involved closing eyes and relaxing, thus making it conducive to generation of alpha activity. The other two conditions -- eyes open in a lighted room or engaged in a mental arithmetic task, were designed to attenuate alpha activity. Using the Mixture Density Independent Component Analysis (MD-ICA) that incorporates flexible non-linearity functions into the conventional ICA framework, we could identify the spatiotemporal components of fMRI activations and EEG activities associated with the alpha rhythm. The sources of the individual EEG alpha activity component were localized by a Maximum Entropy (ME) method that solves an inverse problem in the framework of a classical four-sphere head model. The resulting dipole sources of EEG alpha activity were spatially transformed to 3D MRIs of the subject and compared to fMRI ICA-determined alpha activity maps.
This paper presents an automatic tissue segmentation methodology for High-Resolution Ultrasonic Transmission Tomography (HUTT) imagery of biological organs. This method combines a recent segmentation approach: the L-level set active contours algorithm with unsupervised clustering using the agglomerative hierarchical k-means algorithm. The active contours algorithm has been recently explored as a powerful tool for image segmentation since it automatically decomposes a given image into 2L segment classes by utilizing L level set functions and finding the optimal boundaries of the 2L segment classes so that the pixel feature values of each segment are as homogeneous as possible. Unfortunately, the algorithm is often trapped at local minima due to the intrinsic non-convexity of the cost function, especially for noisy data. To overcome this problem, we introduce a multi-stage multi-resolution analysis that optimizes the active contours at successive resolutions of the image data. The resulting segments are then re-clustered by subsequent agglomerative hierarchical k-means clustering that seeks the optimal clusters yielding the minimum within-cluster distance in the feature space. The preliminary studies reported here indicate that this proposed methodology can enhance the accuracy of soft tissue segmentation and provide fully automatic tissue differentiation without any user intervention except for specifying the number of level set functions L.
Recently it was shown that soft tissue can be differentiated with spectral unmixing and detection methods that utilize multi-band information obtained from a High-Resolution Ultrasonic Transmission Tomography (HUTT) system. In this study, we focus on tissue differentiation using the spectral target detection method based on Constrained Energy Minimization (CEM). We have developed a new tissue differentiation method called “CEM filter bank”. Statistical inference on the output of each CEM filter of a filter bank is used to make a decision based on the maximum statistical significance rather than the magnitude of each CEM filter output. We validate this method through 3-D inter/intra-phantom soft tissue classification where target profiles obtained from an arbitrary single slice are used for differentiation in multiple tomographic slices. Also spectral coherence between target and object profiles of an identical tissue at different slices and phantoms is evaluated by conventional cross-correlation analysis. The performance of the proposed classifier is assessed using Receiver Operating Characteristic (ROC) analysis. Finally we apply our method to classify tiny structures inside a beef kidney such as Styrofoam balls (~1mm), chicken tissue (~5mm), and vessel-duct structures.