Martin L. Silbiger

MRI segmentation: methods and applications.

The current literature on MRI segmentation methods is reviewed. Particular emphasis is placed on the relative merits of single image versus multispectral segmentation, and supervised versus unsupervised segmentation methods. Image pre-processing and registration are discussed, as well as methods of validation. The application of MRI segmentation for tumor volume measurements during the course of therapy is presented here as an example, illustrating problems associated with inter- and intra-observer variations inherent to supervised methods.

MRI: Stability of three supervised segmentation techniques

Supervised segmentation methods from three families of pattern recognition techniques were used to segment multispectral MRI data. Studied were the maximum likelihood method (MLM), k-nearest neighbors (k-NN), and a back-propagation artificial neural net (ANN). Performance was measured in terms of execution speed, and stability for the selection of training data, namely, region of interest (ROI) selection, and interslice and interpatient classifications. MLM proved to have the smallest execution times, but demonstrated the least stability. k-NN showed the best stability for training data selection. To evaluate the segmentation techniques, multispectral images were used of normal volunteers and patients with gliomas, the latter with and without MR contrast material. All measures applied indicated that k-NN provides the best results.

Application of fuzzy c-means segmentation technique for tissue differentiation in MR images of a hemorrhagic glioblastoma multiforme.

The application of a raw data-based, operator-independent MR segmentation technique to differentiate boundaries of tumor from edema or hemorrhage is demonstrated. A case of a glioblastoma multiforme with gross and histopathologic correlation is presented. The MR image data set was segmented into tissue classes based on three different MR weighted image parameters (T1-, proton density-, and T2-weighted) using unsupervised fuzzy c-means (FCM) clustering algorithm technique for pattern recognition. A radiological examination of the MR images and correlation with fuzzy clustering segmentations was performed. Results were confirmed by gross and histopathology which, to the best of our knowledge, reports the first application of this demanding approach. Based on the results of neuropathologic correlation, the application of FCM MR image segmentation to several MR images of a glioblastoma multiforme represents a viable technique for displaying diagnostically relevant tissue contrast information used in 3D volume reconstruction. With this technique, it is possible to generate segmentation images that display clinically important neuroanatomic and neuropathologic tissue contrast information from raw MR image data.

MONITORING BRAIN TUMOR RESPONSE TO THERAPY USING MRI SEGMENTATION

The performance evaluation of a semi-supervised fuzzy c-means (SFCM) clustering method for monitoring brain tumor volume changes during the course of routine clinical radiation-therapeutic and chemo-therapeutic regimens is presented. The tumor volume determined using the SFCM method was compared with the volume estimates obtained using three other methods: (a) a k nearest neighbor (kNN) classifier, b) a grey level thresholding and seed growing (ISG-SG) method and c) a manual pixel labeling (GT) method for ground truth estimation. The SFCM and kNN methods are applied to the multispectral, contrast enhanced T1, proton density, and T2 weighted, magnetic resonance images (MRI) whereas the ISG-SG and GT methods are applied only to the contrast enhanced T1 weighted image. Estimations of tumor volume were made on eight patient cases with follow-up MRI scans performed over a 32 week interval during treatment. The tumor cases studied include one meningioma, two brain metastases and five gliomas. Comparisons with manually labeled ground truth estimations showed that there is a limited agreement between the segmentation methods for absolute tumor volume measurements when using images of patients after treatment. The average intraobserver reproducibility for the SFCM, kNN and ISG-SG methods was found to be 5.8%, 6.6% and 8.9%, respectively. The average of the interobserver reproducibility of these methods was found to be 5.5%, 6.5% and 11.4%, respectively. For the measurement of relative change of tumor volume as required for the response assessment, the multi-spectral methods kNN and SFCM are therefore preferred over the seedgrowing method.

Tree-structured nonlinear filter and wavelet transform for microcalcification segmentation in mammography

The development of an extensive array of algorithms for both image enhancement and feature extraction for microcalcification cluster detection is reported. Specific emphasis is placed on image detail preservation and automatic or operator independent methods to enhance the sensitivity and specificity of detection and that should allow standardization of breast screening procedures. Image enhancement methods include both novel tree structured non-linear filters with fixed parameters and adaptive order statistic filters designed to further improve detail preservation. Novel feature extraction methods developed include both two channel tree structured wavelet transform and three channel quadrature mirror filter banks with multiresolution decomposition and reconstruction specifically tailored to extract MCCs. These methods were evaluated using fifteen representative digitized mammograms where similar sensitivity (true positive (TP) detection rate 100%) and specificity (0.1 - 0.2 average false positive (FP) MCCs/image) was observed but with varying degrees of detail preservation important for characterization of MCCs. The image enhancement step proved to be very critical to minimize image noise and associated FP detection rates for MCCs or individual microcalcifications.

An investigation of mountain method clustering for large data sets

The Mountain Method of clustering was introduced by Yager and Filev and refined for practical use by Chiu. The approach is based on density estimation in feature space with the highest peak extracted as a cluster center and a new density estimation created for extraction of the next cluster center. The process is repeated until a stopping condition is met. The Chiu version of this approach has been implemented in the Matlab Fuzzy Logic Tool@?. In this paper, we develop an alternate implementation that allows large data sets to be processed effectively. Methods to set the parameters required by the algorithm are also given. Magnetic resonance images of the human brain are used as a test domain. Comparisons with the Matlab implementation show that our new approach is considerably more practical in terms of the time required to cluster, as well as better at producing partitions of the data that correspond to those expected. Comparisons are also made to the fuzzy c-means clustering algorithm, which show that our improved mountain method is a viable competitor, producing excellent partitions of large data sets.

Digital mammography: M-channel quadrature mirror filters (QMFs) for microcalcification extraction.

Multiresolution methods are reported for feature extraction in breast cancer screening using digital mammography. The initial application is directed at the detection of microcalcification clusters (MCCs). Quadrature mirror filter (QMF) banks, using both two and three channel are proposed for the first time for both multiresolution decomposition and reconstruction. These filters are specifically tailored for automatic extraction of MCCs. The QMF multiresolution methods are compared to two channel tree structured wavelet transforms (TSWTs) methods previously reported. The QMF filters are preceded by an advanced tree structured nonlinear filter for noise suppression, prior to feature extraction, in order to minimize the false positive (FP) detection rate in digital mammography. The relative performance of these methods were evaluated using both simulated images and fifteen representative digitized mammograms containing biopsy proven microcalcification clusters. Similar high sensitivity (true positive (TP) detection rate (100%) and high specificity (0.6 average false positive (FP) MCCs/image) were observed, substantially better than more traditional approaches using single scale filters. The three channel QMF method, however, demonstrated better detail preservation of MCCs extracted compared to the two channel method. Detail preservation is important for the characterization of MCCs or individual microcalcifications in cancer screening.

Composite and classified color display in MR imaging of the female pelvis

Because of its superior soft-tissue-imaging capabilities, MRI has proved to be an excellent modality for visualizing the contents of the female pelvis. In an effort to potentially improve gynecological MRI studies, we have applied color composite techniques to sets of spin-echo and gradient-echo gray-tone MR images obtained from various individuals. For composite generation, based on tissue region of interest calculated mean pixel intensity values, various colors were applied to spatially aligned images using a DEC MicroVAX II computer with interactive digital language (IDL) so that tissue contrast patterns could be optimized in the final image. The IDL procedures, which are similar to those used in NASAs LANDSAT image processing system, allowed the generation of single composite images displaying the combined information present in a series of spatially aligned images acquired using different pulse sequences. With our composite generation techniques, it was possible to generate seminatural-appearing color images of the female pelvis that possessed enhanced conspicuity of specific tissues and fluids. For comparison with color composites, classified images were also generated based on computer recognition and statistical separation of distinct tissue intensity patterns in an image set using the maximum likelihood processing algorithm.

Quantitative SPECT imaging: influence of object size.

In positron emission tomography (PET) the measured radionuclide concentration or recovery coefficient (RC) in the tranverse plane has been shown to be dependent on source size. The RC dependence on object size was therefore experimentally measured for gamma camera SPECT system to determine the influence of additional factors such as: (a) geometrical spatial resolution of each collimator type, (b) scattering effects at each photon energy due to surrounding background, (c) influence of choice of reconstruction filter and (d) photon penetration through the collimator septa. Data were acquired for different collimator types using 99Tcm (140 keV) and 131I (364 keV) to partly differentiate between photon scattering and collimator penetration effects. Results obtained demonstrated that thick septa collimators with low penetration fraction (theoretical leakage < 3%) are required to minimize the measured response dependence on source size and to obtain a relatively unique relationship independent of photon energy and background activity. Measurements of this type are important for quantitative SPECT imaging in radioimmunoimaging or radioimmunotherapy.

Intervertebral Disc Calcification in Childhood

No lumbar punctures had ever been performed. On physical examination, she was afebrile, apprehensive, mentally retarded, and in no distress. She held her neck rigidly in slight flexion and resisted manipulation. No other unusual physical findings were present. There was no lymphadenopathy, otitis, or tonsillitis, and Kernig and Brudzinski signs were absent. The results of routine laboratory studies were within normal limits. A twenty-four hour urine study for abnormal amino acids was unremarkable. Studies