In clinical problems, numerous factors are usually involved in a medical syndrome. New advances in medicine provide a broad range of diagnosis methods to cover all aspects of a disease. However, huge amounts of raw in...
详细信息
In clinical problems, numerous factors are usually involved in a medical syndrome. New advances in medicine provide a broad range of diagnosis methods to cover all aspects of a disease. However, huge amounts of raw information may confuse clinicians and decrease decision accuracy. Computerized knowledge extraction is an active area of research in medical informatics. This paper suggests a new medical data mining approach using an advanced swarm intelligence data mining algorithm. Considering medical knowledge discovery difficulties, this approach addresses common issues such as missing value management and interactive rule extraction. Here, surgery candidate selection in temporal lobe epilepsy is the main target application. However, the general idea can be applied to other medical knowledge discovery problems. Experimental results show noticeable performance improvement in the final rule-set quality while the method is flexible and fast.
Despite its potential advantages for fMRI analysis, fuzzy C-means (FCM) clustering suffers from limitations such as the need for a priori knowledge of the number of clusters, and unknown statistical significance and i...
详细信息
Despite its potential advantages for fMRI analysis, fuzzy C-means (FCM) clustering suffers from limitations such as the need for a priori knowledge of the number of clusters, and unknown statistical significance and instability of the results. We propose a randomization-based method to control the false positive rate and estimate statistical significance of the FCM results. Using this novel approach, we develop an fMRI activation detection method. The ability of the method in controlling the false positive rate is shown by analysis of false positives in activation maps of resting-state fMRI data. Controlling the false positive rate in FCM allows comparison of different fuzzy clustering methods, using different feature spaces, to other fMRI detection methods. In this paper, using simulation and real fMRI data, we compare a novel feature space that takes the variability of the hemodynamic response function into account (HRF-based feature space) to the conventional cross-correlation analysis and FCM using the cross-correlation feature space.
In this paper, a cluster-based framework is introduced for comparing analysis methods of functional magnetic resonance images (fMRI). In the proposed framework, fMRI data is replaced with a feature space and each meth...
详细信息
In this paper, a cluster-based framework is introduced for comparing analysis methods of functional magnetic resonance images (fMRI). In the proposed framework, fMRI data is replaced with a feature space and each method considered as a clustering method in the new space. As a result, different methods can be compared by means of a cluster validity measure. The feature space is computed using a non-parametric method (principal component analysis-PCA). Four subjects have been analyzed with three methods and the proposed cluster-based framework has evaluated performance of the methods. The results are identical to those of the modified receiver operating characteristics (ROC). This validates the proposed approach.
Automatic segmentation of brain tissues is crucial to many medical imaging applications. We use a multi-resolution analysis and a power transform to extend the well-known Gaussian mixture model expectation maximizatio...
详细信息
Automatic segmentation of brain tissues is crucial to many medical imaging applications. We use a multi-resolution analysis and a power transform to extend the well-known Gaussian mixture model expectation maximization based algorithm for segmentation of white matter, gray matter, and cerebrospinal fluid from T1-weighted magnetic resonance images (MRI) of the brain. Experimental results with near 4000 synthetic and real images are included. The results illustrate that the proposed method outperforms six existing methods.
Magnetic resonance spectroscopic imaging (MRSI) is a non-invasive technique for assessing biochemical fingerprint of tissue composition. The need to differentiate between normal and abnormal tissues and determine type...
详细信息
Magnetic resonance spectroscopic imaging (MRSI) is a non-invasive technique for assessing biochemical fingerprint of tissue composition. The need to differentiate between normal and abnormal tissues and determine type of abnormality before biopsy or surgery motivated development and application of MRSI. There are several technical reasons that make the brain easier than other organs to be examined with MRSI. This work presents our proposed methods and results for the analysis of the brain spectra of patients with three tumor types (malignant glioma, astrocytoma, and oligodendroglioma). After extracting features from MRSI data using wavelet and wavelet packets, we use artificial neural networks to determine the abnormal spectra and the type of abnormality. We evaluated the proposed methods using clinical and simulated MRSI data and biopsy results. The MRSI analysis results were correct 97% of the time when classifying the spectra of the clinical MRSI data into normal tissue, tumor, and radiation necrosis. They were correct 72% and 83% of the time when determining tumor types using the clinical and simulated MRSI data, respectively.
A new method based on generalized likelihood ratio test (GLRT) for activation detection in multi-subject studies of functional MRI (fMRI) is proposed. In this method, we test the correlation between the fMRI time seri...
详细信息
A new method based on generalized likelihood ratio test (GLRT) for activation detection in multi-subject studies of functional MRI (fMRI) is proposed. In this method, we test the correlation between the fMRI time series of different subjects and the bases of a signal subspace which increases the flexibility of method in detecting different shapes of hemodynamic response. The proposed multivariate method can be applied to group studies where the conventional cross-correlation method cannot be used due to its univariate property. This method is applied to both experimental and simulated fMRI data and the results are compared to those of general linear model (GLM). We show that the proposed method detects more significant activated regions in analyzing experimental data and more true voxels in simulated data
暂无评论