Hybrid artificial neural network segmentation of precise and accurate inversion recovery (PAIR) images from normal human brain

A Novel Radial Basis Neural Network-Leveraged Fast Training Method for Identifying Organs in MR Images

We propose a new method for fast organ classification and segmentation of abdominal magnetic resonance (MR) images. Magnetic resonance imaging (MRI) is a new type of high-tech imaging examination fashion in recent years. Recognition of specific target areas (organs) based on MR images is one of the key issues in computer-aided diagnosis of medical images. Artificial neural network technology has made significant progress in image processing based on the multimodal MR attributes of each pixel in MR images. However, with the generation of large-scale data, there are few studies on the rapid processing of large-scale MRI data. To address this deficiency, we present a fast radial basis function artificial neural network (Fast-RBF) algorithm. The importance of our efforts is as follows: (1) The proposed algorithm achieves fast processing of large-scale image data by introducing the ε-insensitive loss function, the structural risk term, and the core-set principle. We apply this algorithm to the identification of specific target areas in MR images. (2) For each abdominal MRI case, we use four MR sequences (fat, water, in-phase (IP), and opposed-phase (OP)) and the position coordinates (x, y) of each pixel as the input of the algorithm. We use three classifiers to identify the liver and kidneys in the MR images. Experiments show that the proposed method achieves a higher precision in the recognition of specific regions of medical images and has better adaptability in the case of large-scale datasets than the traditional RBF algorithm.

Medical image analysis with artificial neural networks

Given that neural networks have been widely reported in the research community of medical imaging, we provide a focused literature survey on recent neural network developments in computer-aided diagnosis, medical image segmentation and edge detection towards visual content analysis, and medical image registration for its pre-processing and post-processing, with the aims of increasing awareness of how neural networks can be applied to these areas and to provide a foundation for further research and practical development. Representative techniques and algorithms are explained in detail to provide inspiring examples illustrating: (i) how a known neural network with fixed structure and training procedure could be applied to resolve a medical imaging problem; (ii) how medical images could be analysed, processed, and characterised by neural networks; and (iii) how neural networks could be expanded further to resolve problems relevant to medical imaging. In the concluding section, a highlight of comparisons among many neural network applications is included to provide a global view on computational intelligence with neural networks in medical imaging.

MR image segmentation using a power transformation approach

This study proposes a segmentation method for brain MR images using a distribution transformation approach. The method extends traditional Gaussian mixtures expectation-maximization segmentation to a power transformed version of mixed intensity distributions, which includes Gaussian mixtures as a special case. As MR intensities tend to exhibit non-Gaussianity due to partial volume effects, the proposed method is designed to fit non-Gaussian tissue intensity distributions. One advantage of the method is that it is intuitively appealing and computationally simple. To avoid performance degradation caused by intensity inhomogeneity, different methods for correcting bias fields were applied prior to image segmentation, and their correction effects on the segmentation results were examined in the empirical study. The partitions of brain tissues (i.e., gray and white matter) resulting from the method were validated and evaluated against manual segmentation results based on 38 real T1-weighted image volumes from the internet brain segmentation repository, and 18 simulated image volumes from BrainWeb. The Jaccard and Dice similarity indexes were computed to evaluate the performance of the proposed approach relative to the expert segmentations. Empirical results suggested that the proposed segmentation method yielded higher similarity measures for both gray matter and white matter as compared with those based on the traditional segmentation using the Gaussian mixtures approach.

Magnetic resonance imaging segmentation techniques using batch-type learning vector quantization algorithms

In this article, we propose batch-type learning vector quantization (LVQ) segmentation techniques for the magnetic resonance (MR) images. Magnetic resonance imaging (MRI) segmentation is an important technique to differentiate abnormal and normal tissues in MR image data. The proposed LVQ segmentation techniques are compared with the generalized Kohonen's competitive learning (GKCL) methods, which were proposed by Lin et al. [Magn Reson Imaging 21 (2003) 863-870]. Three MRI data sets of real cases are used in this article. The first case is from a 2-year-old girl who was diagnosed with retinoblastoma in her left eye. The second case is from a 55-year-old woman who developed complete left side oculomotor palsy immediately after a motor vehicle accident. The third case is from an 84-year-old man who was diagnosed with Alzheimer disease (AD). Our comparisons are based on sensitivity of algorithm parameters, the quality of MRI segmentation with the contrast-to-noise ratio and the accuracy of the region of interest tissue. Overall, the segmentation results from batch-type LVQ algorithms present good accuracy and quality of the segmentation images, and also flexibility of algorithm parameters in all the comparison consequences. The results support that the proposed batch-type LVQ algorithms are better than the previous GKCL algorithms. Specifically, the proposed fuzzy-soft LVQ algorithm works well in segmenting AD MRI data set to accurately measure the hippocampus volume in AD MR images.

Identification of human term and preterm labor using artificial neural networks on uterine electromyography data

OBJECTIVE

To use artificial neural networks (ANNs) on uterine electromyography (EMG) data to classify term/preterm labor/non-labor pregnant patients.

MATERIALS AND METHODS

A total of 134 term and 51 preterm women (all ultimately delivered spontaneously) were included. Uterine EMG was measured trans-abdominally using surface electrodes. "Bursts" of elevated uterine EMG, corresponding to uterine contractions, were quantified by finding the means and/or standard deviations of the power spectrum (PS) peak frequency, burst duration, number of bursts per unit time, and total burst activity. Measurement-to-delivery (MTD) time was noted for each patient. Term and preterm patient groups were sub-divided, resulting in the following categories: [term-laboring (TL): n = 75; preterm-laboring (PTL): n = 13] and [term-non-laboring (TN): n = 59; preterm-non-laboring (PTN): n = 38], with labor assessed using clinical determinations. ANN was then used on the calculated uterine EMG data to algorithmically and objectively classify patients into labor and non-labor. The percent of correctly categorized patients was found. Comparison between ANN-sorted groups was then performed using Student's t test (with p < 0.05 significant).

RESULTS

In total, 59/75 (79%) of TL patients, 12/13 (92%) of PTL patients, 51/59 (86%) of TN patients, and 27/38 (71%) of PTN patients were correctly classified.

CONCLUSION

ANNs, used with uterine EMG data, can effectively classify term/preterm labor/non-labor patients.

Prediction of total cerebral tissue volumes in normal appearing brain from sub-sampled segmentation volumes

The need for anatomical coverage and multi-spectral information must be balanced against examination and processing time to ensure high-quality, feasible imaging protocols for clinical research of cerebral development in normal-appearing brains. The focus of this study was to create and assess models to estimate total cerebral volumes of gray matter, white matter, and cerebrospinal fluid (CSF) from anatomically defined sub-samples of full clinical examinations. Pediatric patients (18F, 11M; aged 1.7 to 18.7, median 5.2 years) underwent a clinical imaging protocol consisting of 3 mm contiguous T1-, T2-, PD-, and FLAIR-weighted images after obtaining informed consent. Magnetic resonance imaging (MRI) sets were registered, RF-corrected, and then analyzed with a hybrid neural network segmentation and classification algorithm to identify normal brain parenchyma. The correlation between the image subsets and the total cerebral volumes of gray matter, white matter and CSF were examined through linear regression analyses. Five sub-sampled sets were defined and assessed in each patient to produce estimation models which were all significantly correlated (p < 0.001) with the total cerebral volumes of gray matter, white matter, and CSF. Volumes were estimated from as little as a single representative slice requiring minimal processing time, 27 min, but with an average estimation error of approximately 6%. Larger sub-samples of approximately three-quarters of the full cerebral volume required much more processing time, 2 h and 4 min, but produced estimates with an average error less than 2%. This study demonstrated that investigators can choose the amount of cerebrum sampled to optimize the acquisition and processing time against the degree of accuracy needed in the total cerebral volume estimates.

Self-organizing neural network analyses of cardiac data in depression

OBJECTIVE

To determine if an unsupervised self-organizing neural network could create a clinically meaningful distinction of 'depression' versus 'no depression' based on cardiac time-series data.

DESIGN

A self-organizing map (SOM) was used to separate the time-series of 84 subjects into groups based on characteristics of the data alone.

MATERIALS AND METHODS

Analyses included natural log transformations and two types of filtering to enhance characteristics of the data as well as classifications of unprocessed data. A Pearson chi(2) analysis was performed to determine if the SOM groups bore any relation to the binary clinical groups.

RESULTS

Overall correct SOM classifications ranged from 54 to 70.2% with two classifications being clinically meaningful.

CONCLUSIONS

SOM classifications of cardiac time-series data with enhanced ultradian variations and cardiac data recorded around the interval when a person was in bed were useful in differentiating clinically meaningful subgroups with and without depression.

Computer-assisted enhanced volumetric segmentation magnetic resonance imaging data using a mixture of artificial neural networks

An accurate computer-assisted method able to perform regional segmentation on 3D single modality images and measure its volume is designed using a mixture of unsupervised and supervised artificial neural networks. Firstly, an unsupervised artificial neural network is used to estimate representative textures that appear in the images. The region of interest of the resultant images is selected by means of a multi-layer perceptron after a training using a single sample slice, which contains a central portion of the 3D region of interest. The method was applied to magnetic resonance imaging data collected from an experimental acute inflammatory model (T(2) weighted) and from a clinical study of human Alzheimer's disease (T(1) weighted) to evaluate the proposed method. In the first case, a high correlation and parallelism was registered between the volumetric measurements, of the injured and healthy tissue, by the proposed method with respect to the manual measurements (r = 0.82 and p < 0.05) and to the histopathological studies (r = 0.87 and p < 0.05). The method was also applied to the clinical studies, and similar results were derived of the manual and semi-automatic volumetric measurement of both hippocampus and the corpus callosum (0.95 and 0.88).

Generalized Kohonen’s competitive learning algorithms for ophthalmological MR image segmentation

Kohonen's self-organizing map is a two-layer feedforward competitive learning network. It has been used as a competitive learning clustering algorithm. In this paper, we generalize Kohonen's competitive learning (KCL) algorithm with fuzzy and fuzzy-soft types called fuzzy KCL (FKCL) and fuzzy-soft KCL (FSKCL). These generalized KCL algorithms fuse the competitive learning with soft competition and fuzzy c-means (FCM) membership functions. We then apply these generalized KCLs to MRI and MRA ophthalmological segmentations. These KCL-based MRI segmentation techniques are useful in reducing medical image noise effects using a learning mechanism. They may be particularly helpful in clinical diagnosis. Two real cases with MR image data recommended by an ophthalmologist are examined. First case is a patient with Retinoblastoma in her left eye, an inborn malignant neoplasm of the retina frequently metastasis beyond the lacrimal cribrosa. The second case is a patient with complete left side oculomotor palsy immediately after a motor vehicle accident. Her brain MRI with MRA, skull routine, orbital CT, and cerebral angiography did not reveal brainstem lesions, skull fractures, or vascular anomalies. These generalized KCL algorithms were used in segmenting the ophthalmological MRIs. KCL, FKCL and FSKCL comparisons are made. Overall, the FSKCL algorithm is recommended for use in MR image segmentation as an aid to small lesion diagnosis.

Effect of therapeutic ionizing radiation on the human brain

We test a hypothesis that fractionated radiation therapy within a therapeutic dose range is associated with a dose-related change in normal brain, detectable by quantitative magnetic resonance imaging. A total of 33 patients were examined by quantitative magnetic resonance imaging to measure brain tissue spin-lattice relaxation time (T1) before treatment, and at various times during and after radiation therapy. A T1 map was generated at each time point, and radiation therapy isodose contours were superimposed on the corresponding segmented T1 map. Changes in white matter and gray matter T1 were analyzed as a function of radiation therapy dose and time since treatment, controlling for patient age and tumor site. In white matter, a dose level of more than 20 Gy was associated with a dose-dependent decrease in T1 over time, which became significant 6 months after treatment. There was no significant change in T1 of gray matter over time, at radiation therapy doses of less than 60 Gy. However, GM in close proximity to the tumor had a lower T1 before therapy. Our results represent the first radiation dose-response data derived from pediatric brain in vivo. These findings confirm that white matter is more vulnerable to radiation-induced change than is gray matter, and suggest that T1 mapping is sensitive to radiation-related changes over a broad dose range (20 to 60 Gy). Human white matter T1 is not sensitive to radiation therapy of less than 20 Gy, and gray matter T1 is unchanged over the dose range used to treat human brain tumor. The reduction of gray matter T1 near the tumor could result from compression of cortical parenchyma near the growing tumor mass, or from tumor cell invasion directly into the parenchyma. If brain T1 is a surrogate for radiation effect, reducing the volume of normal white matter receiving more than 20 Gy could be an important treatment planning goal.

Current awareness

In order to keep subscribers up-to-date with the latest developments in their field, John Wiley & Sons are providing a current awareness service in each issue of the journal. The bibliography contains newly published material in the field of NMR in biomedicine. Each bibliography is divided into 9 sections: 1 Books, Reviews ' Symposia; 2 General; 3 Technology; 4 Brain and Nerves; 5 Neuropathology; 6 Cancer; 7 Cardiac, Vascular and Respiratory Systems; 8 Liver, Kidney and Other Organs; 9 Muscle and Orthopaedic. Within each section, articles are listed in alphabetical order with respect to author. If, in the preceding period, no publications are located relevant to any one of these headings, that section will be omitted.

Effect of ionizing radiation on the human brain: white matter and gray matter T1 in pediatric brain tumor patients treated with conformal radiation therapy

OBJECTIVE

To test a hypothesis that fractionated radiation therapy (RT) to less than 60 Gy is associated with a dose-related change in the spin-lattice relaxation time (T1) of normal brain tissue, and that such changes are detectable by quantitative MRI (qMRI).

METHODS

Each of 21 patients received a qMRI examination before treatment, and at several time points during and after RT. A map of brain T1 was calculated and segmented into white matter and gray matter at each time point. The RT isodose contours were then superimposed upon the T1 map, and changes in brain tissue T1 were analyzed as a function of radiation dose and time following treatment. We used a mixed-model analysis to analyze the longitudinal trend in brain T1 from the start of RT to 1 year later. Predictive factors evaluated included patient age and clinical variables, such as RT dose, time since treatment, and the use of an imaging contrast agent.

RESULTS

In white matter (WM), a dose level of greater than 20 Gy was associated with a dose-dependent decrease in T1 over time, which became significant about 3 months following treatment. In gray matter (GM), there was no significant change in T1 over time, as a function of RT doses < 60 Gy. However, GM in close proximity to the tumor had an inherently lower T1 before therapy. Neither use of a contrast agent nor a combination of chemotherapy plus steroids had a significant effect on brain T1.

CONCLUSION

Results suggest that T1 mapping may be sensitive to radiation-related changes in human brain tissue T1. WM T1 appears to be unaffected by RT at doses less than approximately 20 Gy; GM T1 does not change at doses less than 60 Gy. However, tumor appears to have an effect upon adjacent GM, even before treatment. Conformal RT may offer a substantial benefit to the patient, by minimizing the volume of normal brain exposed to greater than 20 Gy.