Volumetric analyses of central nervous system neoplasm based on MRI

Direct Evaluation of Treatment Response in Brain Metastatic Disease with Deep Neuroevolution

Cancer centers have an urgent and unmet clinical and research need for AI that can guide patient management. A core component of advancing cancer treatment research is assessing response to therapy. Doing so by hand, for example, as per RECIST or RANO criteria, is tedious and time-consuming, and can miss important tumor response information. Most notably, the prevalent response criteria often exclude lesions, the non-target lesions, altogether. We wish to assess change in a holistic fashion that includes all lesions, obtaining simple, informative, and automated assessments of tumor progression or regression. Because genetic sub-types of cancer can be fairly specific and patient enrollment in therapy trials is often limited in number and accrual rate, we wish to make response assessments with small training sets. Deep neuroevolution (DNE) is a novel radiology artificial intelligence (AI) optimization approach that performs well on small training sets. Here, we use a DNE parameter search to optimize a convolutional neural network (CNN) that predicts progression versus regression of metastatic brain disease. We analyzed 50 pairs of MRI contrast-enhanced images as our training set. Half of these pairs, separated in time, qualified as disease progression, while the other 25 image pairs constituted regression. We trained the parameters of a CNN via "mutations" that consisted of random CNN weight adjustments and evaluated mutation "fitness" as summed training set accuracy. We then incorporated the best mutations into the next generation's CNN, repeating this process for approximately 50,000 generations. We applied the CNNs to our training set, as well as a separate testing set with the same class balance of 25 progression and 25 regression cases. DNE achieved monotonic convergence to 100% training set accuracy. DNE also converged monotonically to 100% testing set accuracy. We have thus shown that DNE can accurately classify brain metastatic disease progression versus regression. Future work will extend the input from 2D image slices to full 3D volumes, and include the category of "no change." We believe that an approach such as ours can ultimately provide a useful and informative complement to RANO/RECIST assessment and volumetric AI analysis.

Intuitionistic Center-Free FCM Clustering for MR Brain Image Segmentation

In this paper, an intuitionistic center-free fuzzy c-means clustering method (ICFFCM) is proposed for magnetic resonance (MR) brain image segmentation. First, in order to suppress the effect of noise in MR brain images, a pixel-to-pixel similarity with spatial information is defined. Then, for the purpose of handling the vagueness in MR brain images as well as the uncertainty in clustering process, a pixel-to-cluster similarity measure is defined by employing the intuitionistic fuzzy membership function. These two similarities are used to modify the center-free FCM so that the ability of the method for MR brain image segmentation could be improved. Second, on the basis of the improved center-free FCM method, a local information term, which is also intuitionistic and center-free, is appended to the objective function. This generates the final proposed ICFFCM. The consideration of local information further enhances the robustness of ICFFCM to the noise in MR brain images. Experimental results on the simulated and real MR brain image datasets show that ICFFCM is effective and robust. Moreover, ICFFCM could outperform several fuzzy-clustering-based methods and could achieve comparable results to the standard published methods like statistical parametric mapping and FMRIB automated segmentation tool.

Regional infant brain development: an MRI-based morphometric analysis in 3 to 13 month olds

Elucidation of infant brain development is a critically important goal given the enduring impact of these early processes on various domains including later cognition and language. Although infants' whole-brain growth rates have long been available, regional growth rates have not been reported systematically. Accordingly, relatively less is known about the dynamics and organization of typically developing infant brains. Here we report global and regional volumetric growth of cerebrum, cerebellum, and brainstem with gender dimorphism, in 33 cross-sectional scans, over 3 to 13 months, using T1-weighted 3-dimensional spoiled gradient echo images and detailed semi-automated brain segmentation. Except for the midbrain and lateral ventricles, all absolute volumes of brain regions showed significant growth, with 6 different patterns of volumetric change. When normalized to the whole brain, the regional increase was characterized by 5 differential patterns. The putamen, cerebellar hemispheres, and total cerebellum were the only regions that showed positive growth in the normalized brain. Our results show region-specific patterns of volumetric change and contribute to the systematic understanding of infant brain development. This study greatly expands our knowledge of normal development and in future may provide a basis for identifying early deviation above and beyond normative variation that might signal higher risk for neurological disorders.

Motor stereotypies and volumetric brain alterations in children with Autistic Disorder

Motor stereotypies are defined as patterned, repetitive, purposeless movements. These stigmatizing motor behaviors represent one manifestation of the third core criterion for an Autistic Disorder (AD) diagnosis, and are becoming viewed as potential early markers of autism. Moreover, motor stereotypies might be a tangible expression of the underlying neurobiology of this neurodevelopmental disorder. In this study, we videoscored stereotypies recorded during semi-structured play sessions from school age children with AD. We examined the effect of severity and persistence over time of stereotypies on brain volumetric changes. Our findings confirmed that the brain volume of school age children with AD is, on average, larger than that of age-matched typically developing children. However, we have failed to detect any sign of volumetric differences in brain regions thought to be particularly linked to the pathophysiology of stereotypies. This negative finding may suggest that, at least with respect to motor stereotypies, functional rather than structural alterations might be the underpinning of these disruptive motor manifestations of autism.

Hybrid genetic and variational expectation-maximization algorithm for gaussian-mixture-model-based brain MR image segmentation

The expectation-maximization (EM) algorithm has been widely applied to the estimation of gaussian mixture model (GMM) in brain MR image segmentation. However, the EM algorithm is deterministic and intrinsically prone to overfitting the training data and being trapped in local optima. In this paper, we propose a hybrid genetic and variational EM (GA-VEM) algorithm for brain MR image segmentation. In this approach, the VEM algorithm is performed to estimate the GMM, and the GA is employed to initialize the hyperparameters of the conjugate prior distributions of GMM parameters involved in the VEM algorithm. Since GA has the potential to achieve global optimization and VEM can steadily avoid overfitting, the hybrid GA-VEM algorithm is capable of overcoming the drawbacks of traditional EM-based methods. We compared our approach to the EM-based, VEM-based, and GA-EM based segmentation algorithms, and the segmentation routines used in the statistical parametric mapping package and FMRIB Software Library in 20 low-resolution and 17 high-resolution brain MR studies. Our results show that the proposed approach can improve substantially the performance of brain MR image segmentation.

Change Detection & Characterization: A New Tool for Imaging Informatics and Cancer Research

Modern imaging systems are able to produce a rich and diverse array of information, regarding various facets of anatomy and function. The quantity of information produced by these systems is so bountiful, however, as to have the potential to become a hindrance to clinical assessment. In the context of serial image evaluation, computer-based change detection and characterization is one important mechanism to process the information produced by imaging systems, so as to reduce the quantity of data, direct the attention of the physician to regions of the data which are the most informative for their purposes, and present the data in the form in which it will be the most useful. Change detection and characterization algorithms may serve as a basis for the creation of an objective definition of progression, which will reduce inter and intra-observer variability, and facilitate earlier detection of disease and recurrence, which in turn may lead to improved outcomes. Decreased observer variability combined with increased acuity should make it easier to discover promising therapies. Quantitative measures of the response to these therapies should provide a means to compare the effectiveness of treatments under investigation. Change detection may be applicable to a broad range of cancers, in essentially all anatomical regions. The source of information upon which change detection comparisons may be based is likewise broad. Validation of algorithms for the longitudinal assessment of cancer patients is expected to be challenging, though not insurmountable, as the many facets of the problem mean that validation will likely need to be approached from a variety of vantage points. Change detection and characterization is quickly becoming a very active field of investigation, and it is expected that this burgeoning field will help to facilitate cancer care both in the clinic and research.

Detailed semiautomated MRI based morphometry of the neonatal brain: preliminary results

In the neonate, regional growth trajectories provide information about the coordinated development of cerebral substructures and help identify regional vulnerability by identifying times of faster growth. Segmentation of magnetic resonance images (MRI) has provided detailed information for the myelinated brain but few reports of regional neonatal brain growth exist. We report the method and preliminary results of detailed semiautomated segmentation of 12 normative neonatal brains (gestational age 31.1-42.6 weeks at time of MRI) using volumetric T1-weighted images. Accuracy was confirmed by expert review of every segmented image. In 5 brains, repeat segmentation resulted in intraclass correlation coefficients >0.9 (except for the right amygdala) and an average percent voxel overlap of 90.0%. Artifacts or image quality limited the number of regions segmented in 9/12 data sets and 1/12 was excluded from volumetric analysis due to ventriculomegaly. Brains were segmented into cerebral exterior (N = 8), cerebral lobes (N = 5), lateral ventricles (N = 8), cerebral cortex (N = 6), white matter (N = 6), corpus callosum (N = 7), deep central gray (N = 8), hippocampi (N = 8), amygdalae (N = 8), cerebellar hemispheres (N = 8), vermis (N = 8), midbrain (N = 8), pons (N = 8) and medulla (N = 8). Linear growth (P < 0.05) was identified in all regions except the cerebral white matter, medulla and ventricles. Striking differences in regional growth rates were noted. These preliminary results are consistent with the heterochronous nature of cerebral development and provide initial estimates of regional brain growth and therefore regional vulnerability in the perinatal time period.

Imaging the Developing Epileptic Brain

Accurate identification of cortical malformations in children with epilepsy can be crucial for successful clinical management. Although standard head-coil magnetic resonance imaging (MRI) at 1.5 tesla (T) can be used to view the macrostructure of the brain, phased array technology at both 1.5 and 3T significantly improves signal-to-noise ratio (SNR). As a result, spatial resolution and contrast can be optimized to increase visual detection of subtle macrostructural changes that occur with small epileptogenic lesions. In addition, these improvements in SNR allow more accurate quantitative analysis of brain macrostructure and more accurate assessment of brain microstructure using newer sophisticated imaging techniques. For example, phased array imaging enables more accurate diffusion tensor imaging (DTI), and 3T imaging, when combined with phased array technology, enables more informative diffusion spectroscopic imaging (DSI). Recent technological improvements therefore result in improved lesion detection and enable assessment of cerebral growth trajectories and associated longitudinal changes in tissue microstructural organization that occur in association with various types of epilepsy. This article presents a brief comparison of imaging techniques currently in use, both clinically and experimentally, to diagnose, treat, and increase our understanding of the neuropathology of epilepsy in the developing brain.

A review of the automated detection of change in serial imaging studies of the brain

Serial imaging is frequently performed on patients with diseases of the brain, to track and observe changes. Magnetic resonance imaging provides very detailed and rich information, and is therefore used frequently for this application. The data provided by MR can be so plentiful; however, that it obfuscates the information the radiologist seeks. A system which could reduce the large quantity of primitive data to a smaller and more informative subset of data, emphasizing change, would be useful. This article discusses motivating factors for the production of an automated process to this effect, and reviews the approaches of previous authors. The discussion is focused on brain tumors and multiple sclerosis, but many of the ideas are applicable to other disease processes, as well.

Dissociations of cerebral cortex, subcortical and cerebral white matter volumes in autistic boys

High-functioning autistic and normal school-age boys were compared using a whole-brain morphometric profile that includes both total brain volume and volumes of all major brain regions. We performed MRI-based morphometric analysis on the brains of 17 autistic and 15 control subjects, all male with normal intelligence, aged 7-11 years. Clinical neuroradiologists judged the brains of all subjects to be clinically normal. The entire brain was segmented into cerebrum, cerebellum, brainstem and ventricles. The cerebrum was subdivided into cerebral cortex, cerebral white matter, hippocampus-amygdala, caudate nucleus, globus pallidus plus putamen, and diencephalon (thalamus plus ventral diencephalon). Volumes were derived for each region and compared between groups both before and after adjustment for variation in total brain volume. Factor analysis was then used to group brain regions based on their intercorrelations. Volumes were significantly different between groups overall; and diencephalon, cerebral white matter, cerebellum and globus pallidus-putamen were significantly larger in the autistic group. Brain volumes were not significantly different overall after adjustment for total brain size, but this analysis approached significance and effect sizes and univariate comparisons remained notable for three regions, although not all in the same direction: cerebral white matter showed a trend towards being disproportionately larger in autistic boys, while cerebral cortex and hippocampus-amygdala showed trends toward being disproportionately smaller. Factor analysis of all brain region volumes yielded three factors, with central white matter grouping alone, and with cerebral cortex and hippocampus-amygdala grouping separately from other grey matter regions. This morphometric profile of the autistic brain suggests that there is an overall increase in brain volumes compared with controls. Additionally, results suggest that there may be differential effects driving white matter to be larger and cerebral cortex and hippocampus-amygdala to be relatively smaller in the autistic than in the typically developing brain. The cause of this apparent dissociation of cerebral cortical regions from subcortical regions and of cortical white from grey matter is unknown, and merits further investigation.

Retrospective correction of MR intensity inhomogeneity by information minimization

In this paper, the problem of retrospective correction of intensity inhomogeneity in magnetic resonance (MR) images is addressed. A novel model-based correction method is proposed, based on the assumption that an image corrupted by intensity inhomogeneity contains more information than the corresponding uncorrupted image. The image degradation process is described by a linear model, consisting of a multiplicative and an additive component which are modeled by a combination of smoothly varying basis functions. The degraded image is corrected by the inverse of the image degradation model. The parameters of this model are optimized such that the information of the corrected image is minimized while the global intensity statistic is preserved. The method was quantitatively evaluated and compared to other methods on a number of simulated and real MR images and proved to be effective, reliable, and computationally attractive. The method can be widely applied to different types of MR images because it solely uses the information that is naturally present in an image, without making assumptions on its spatial and intensity distribution. Besides, the method requires no preprocessing, parameter setting, nor user interaction. Consequently, the proposed method may be a valuable tool in MR image analysis.

Mapping therapeutic response in a patient with malignant glioma

Short-interval scanning of patients offers a detailed understanding of the natural progression of tumor tissue, as revealed through imaging markers such as contrast enhancement and edema, prior to therapy. Following treatment, short-interval scanning can also provide evidence of attenuation of growth rates. We present a longitudinal imaging study of a patient with glioblastoma multiforme (GBM) scanned 15 times in 104 days on a 3 T MR scanner. Images were analyzed independently by two automated algorithms capable of creating detailed maps of tumor changes as well as volumetric analysis. The algorithms, a nearest-neighbor-based tissue segmentation and a surface-modeling algorithm, tracked the patient's response to temozolomide, showing an attenuation of growth. The need for surrogate imaging end-points, of which growth rates are an example, is discussed. Further, the strengths of these algorithms, the insight gained by short-interval scanning, and the need for a better understanding of imaging markers are also described.

Automatic segmentation of subcortical brain structures in MR images using information fusion

This paper reports a new automated method for the segmentation of internal cerebral structures using an information fusion technique. The information is provided both by images and expert knowledge, and consists in morphological, topological, and tissue constitution data. All this ambiguous, complementary and redundant information is managed using a three-step fusion scheme based on fuzzy logic. The information is first modeled into a common theoretical frame managing its imprecision and incertitude. The models are then fused and a decision is taken in order to reduce the imprecision and to increase the certainty in the location of the structures. The whole process is illustrated on the segmentation of thalamus, putamen, and head of the caudate nucleus from expert knowledge and magnetic resonance images, in a protocol involving 14 healthy volunteers. The quantitative validation is achieved by comparing computed, manually segmented structures and published data by means of indexes assessing the accuracy of volume estimation and spatial location. Results suggest a consistent volume estimation with respect to the expert quantification and published data, and a high spatial similarity of the segmented and computed structures. This method is generic and applicable to any structure that can be defined by expert knowledge and morphological images.

Etiology of neuroanatomical correlates of reading disability

The heritable nature of reading disability has been well documented (DeFries & Alarcón, 1996), and possible abnormalities of brain structures have been associated with the disorder (Filipek, 1995). However, the etiology of individual differences in morphological brain measures has not been examined extensively. The purpose of this study was to apply behavioral genetic methods to assess the etiology of individual differences in neuroanatomical structures. Measures of reading performance, cognitive ability, and magnetic resonance imaging scans were obtained from 25 monozygotic (MZ) and 23 same-sex dizygotic (DZ) twin pairs with reading disability, and 9 MZ and 9 DZ control twin pairs participating in the Colorado Learning Disabilities Research Center. Results obtained from multiple regression analyses (DeFries & Fulker, 1985, 1988) of these twin data indicated that individual differences in the size of most cortical and subcortical structures were largely due to heritable influences. Moreover, estimates of heritability did not change appreciably after controlling for IQ and total brain size.

Tumor volume measurements of acoustic neuromas with three-dimensional constructive interference in steady state and conventional spin-echo MR imaging

The purpose was to compare three-dimensional (3D) constructive interference in steady state (CISS) and conventional spin-echo (SE) MR imaging in tumor volume measurements of acoustic neuromas. Twenty-two patients with acoustic neuromas were examined using high-resolution 3D-CISS and SE imaging at a 1.5-T system. Tumor volume determined by SE imaging with the ellipsoid formula was overestimated by 692 mm(3)(35%) on average as compared with that at 3D-CISS with the voxel-count method (the reference standard). Intra- and interobserver variations in SE imaging were poor as compared with 3D-CISS imaging. However, tumor volume results with SE imaging showed a high correlation with those using 3D-CISS imaging (P <. 0001). On the basis of diameters shown on SE images, the tumor volume could be assessed using the following equation (P <.0001): (Tumor volume) = -26.407 + 0.387 x (maximum diameter along the pyramid) x(maximum diameter perpendicular to the pyramid) x (maximum height). J. Magn. Reson. Imaging 2000;12:826-832.

Using MRI to examine brain-behavior relationships in males with attention deficit disorder with hyperactivity

OBJECTIVE

The relationship between neuropsychological measures of inhibition and sustained attention and structural brain differences in the regions of the caudate and the frontal region was examined in males with attention deficit disorder with hyperactivity (ADD/H).

METHOD

Ten males with ADD/H (aged 8-17) and 11 male controls (aged 9-18) participated in a neuropsychological evaluation and had a magnetic resonance imaging scan.

RESULTS

As had been reported previously by these authors, the children with ADD/H were found to have reversed asymmetry of the head of the caudate, smaller volume of the left caudate head, and smaller volume of the white matter of the right frontal lobe. Children with ADD/H were found to score more poorly on measures of inhibition and sustained attention but not on measures of IQ, achievement, or motor speed. Comparison of neuropsychological measures and brain structure measures indicated a significant relationship between reversed caudate asymmetry and measures of inhibition and externalizing behavior; i.e., children with reversed caudate asymmetry performed more poorly on measures of inhibition regardless of group membership. Poorer performance on sustained attention tasks was related to smaller volume of the right-hemispheric white matter.

CONCLUSIONS

There is emerging evidence that compromised brain morphology of selected regions is related to behavioral measures of inhibition and attention.

A twin MRI study of size variations in human brain

Although it is well known that there is considerable variation among individuals in the size of the human brain, the etiology of less extreme individual differences in brain size is largely unknown. We present here data from the first large twin sample (N=132 individuals) in which the size of brain structures has been measured. As part of an ongoing project examining the brain correlates of reading disability (RD), whole brain morphometric analyses of structural magnetic response image (MRI) scans were performed on a sample of adolescent twins. Specifically, there were 25 monozygotic (MZ) and 23 dizygotic (DZ) pairs in which at least one member of each pair had RD and 9 MZ and 9 DZ pairs in which neither member had RD. We first factor-analyzed volume data for 13 individual brain structures, comprising all of the neocortex and most of the subcortex. This analysis yielded two factors ("cortical" and "subcortical") that accounted for 64% of the variance. We next tested whether genetic and environmental influences on brain size variations varied for these two factors or by hemisphere. We computed intraclass correlations within MZ and DZ pairs in each sample for the cortical and subcortical factor scores, for left and right neocortex, and for the total cerebral volume. All five MZ correlations were substantial (r's=.78 to.98) and significant in both samples, as well as being larger than the corresponding DZ correlations, (r's=0.32 to 0.65) in both samples. The MZ-DZ difference was significant for 3 variables in the RD sample and for one variable in the smaller control sample. These results indicate significant genetic influences on these variables. The magnitude of genetic influence did not vary markedly either for the 2 factors or the 2 hemispheres. There was also a positive correlation between brain size and full-scale IQ, consistent with the results of earlier studies. The total cerebral volume was moderately correlated (r=.42, p<.01, two-tailed) with full-scale IQ in the RD sample; there was a similar trend in the smaller control sample (r=.31, p<.07, two-tailed). Corrections of similar magnitude were found between the subcortical factor and full-scale IQ, whereas the results for the cortical factor (r=.16 and.13) were smaller and not significant. In sum, these results provide evidence for the heritability of individual differences in brain size which do not vary markedly by hemisphere or for neocortex relative to subcortex. Since there are also correlations between brain size and full-scale IQ in this sample, it is possible that genetic influences on brain size partly contribute to individual differences in IQ.

The effect of extent of resection on time to tumor progression and survival in patients with glioblastoma multiforme of the cerebral hemisphere

BACKGROUND

We retrospectively analyzed preoperative and postoperative radiographic tumor volumes in 92 patients who underwent hemispheric glioblastoma multiforme operations (107) to determine the factors that affect time to tumor progression (TTP) and overall survival.

METHODS

Quantification of tumor volumes was based on a previously described method involving computerized image analysis of contrast enhancing tumor on computerized tomography or magnetic resonance imaging scans.

RESULTS

Among the variables analyzed, preoperative Karnofsky Performance Status (KPS) (p < 0.05), chemotherapy (p < 0.05), percent of resection (POR) (p < 0.001), and volume of residual disease (VRD) (p < 0.001) had a significant effect on TTP. Factors that affected survival were age (p < 0.05), preoperative KPS (p = 0.05), postoperative KPS (p < 0.005), POR (p < 0.0005), and VRD (p < 0.0001). Greater resections did not compromise the quality of life, and patients without any residual disease had a better postoperative KPS than those patients who received less than total resections.

CONCLUSIONS

The extent of tumor removal and the amount of residual tumor volume, documented on postoperative imaging studies, are highly significant factors affecting the median time to tumor progression and median survival for patients with glioblastoma multiforme of the cerebral hemisphere.

Segmentation and measurement of the cortex from 3-D MR images using coupled-surfaces propagation

The cortex is the outermost thin layer of gray matter in the brain; geometric measurement of the cortex helps in understanding brain anatomy and function. In the quantitative analysis of the cortex from MR images, extracting the structure and obtaining a representation for various measurements are key steps. While manual segmentation is tedious and labor intensive, automatic reliable efficient segmentation and measurement of the cortex remain challenging problems, due to its convoluted nature. Here we present a new approach of coupled-surfaces propagation, using level set methods to address such problems. Our method is motivated by the nearly constant thickness of the cortical mantle and takes this tight coupling as an important constraint. By evolving two embedded surfaces simultaneously, each driven by its own image-derived information while maintaining the coupling, a final representation of the cortical bounding surfaces and an automatic segmentation of the cortex are achieved. Characteristics of the cortex, such as cortical surface area, surface curvature, and cortical thickness, are then evaluated. The level set implementation of surface propagation offers the advantage of easy initialization, computational efficiency, and the ability to capture deep sulcal folds. Results and validation from various experiments on both simulated and real three-dimensional (3-D) MR images are provided.

Brain morphometry in reading-disabled twins

OBJECTIVE

To test for brain structure differences in reading disability (RD) by means of MRI-based morphometry.

BACKGROUND

Consensus is lacking on the brain structural correlates of RD. The current study reports on a wider set of structures in the largest sample yet studied, controlling for age, gender, IQ, and attention deficit hyperactivity disorder (ADHD).

METHODS

A case-control study was performed that was comprised of 75 individuals with RD (mean age, 17.43+/-4.29 years) and 22 control subjects without RD (mean age, 18.69+/-3.75 years), each a single member of a twin pair. The two groups were similar in age, gender, and handedness, but differed in full-scale IQ (FSIQ), with the RD group having a lower mean FSIQ (101.8+/-9.9 versus 118.3+/-10.3). Using three group-by-structure analyses of covariance, groups were compared in terms of volume (in cubic centimeters) of major neocortical subdivisions, subcortical structures, and midsagittal areas (in square millimeters) of three subdivisions of the corpus callosum.

RESULTS

Controlling for age, gender, and IQ, the authors found a significant group-by-structure interaction for the major neocortical subdivisions (p = 0.002), reflecting a different developmental pattern in the RD group, with the insula and anterior superior neocortex being smaller and the retrocallosal cortex being larger in the RD group. In contrast, they found no group main or interaction effects for the subcortical or callosal structures. The pattern of results was essentially the same in subjects without ADHD.

CONCLUSIONS

Most brain structures do not differ in size in RD, but cortical development is altered subtly. This study replicates in a larger sample previous findings of insular differences in RD and demonstrates further that those differences are not attributable to comorbid ADHD.

Choroid plexus carcinomas in childhood: clinical features and prognostic factors

OBJECTIVE

Choroid plexus carcinomas are rare tumors with dismal prognosis. The role of surgery has been well established, but the benefit of either chemotherapy or radiotherapy remains controversial. To determine prognostic factors and effects of different therapeutic modalities on the outcome, we have reviewed the French experience of choroid plexus carcinoma.

METHODS

Twenty-two children were registered in the Société Française d'Oncologie Pédiatrique between 1984 and 1995. All these children underwent surgical resection of the primary tumor. The intent of postoperative treatment was to delay or to avoid radiation therapy. Nineteen children received postoperative treatment, with chemotherapy in 17 and radiation therapy in 2. Two responding patients underwent high-dose chemotherapy with stem cell rescue.

RESULTS

The 5-year survival rate was 26%. The sole relevant prognostic factor was the extent of surgery. Patients with total or gross total resection had a 86% survival rate. Survival did not correlate with age, sex, delay between first appearance of symptoms and diagnosis, location of the primary tumor, tumor volume, or response to postoperative treatment. All but one patient with incomplete surgery had tumor recurrence within 2 to 23 months.

CONCLUSION

Choroid plexus carcinoma has a very poor prognosis when surgery is incomplete. Aggressive surgical resection of the tumor is necessary for survival. Although chemotherapy gives promising responses, local control remains the main challenge, and "second look" surgery has to be considered for patients with incomplete resection.

Quantitative computer-assisted analysis vs. visual estimation of MR imaging response of brain metastases to radiotherapy

Using both quantitative computer-assisted analysis and visual estimation the MR imaging response of brain metastases to whole-brain radiotherapy (WBRT) was prospectively studied in 49 patients. Compared to the computer-assisted analysis, the precision of the visual estimation (by two readers in conference) of tumor response was determined. Also the radiological response was related to the clinical response. Six weeks after WBRT follow-up MR was performed in 27 of 49 patients. Of the remaining 22 patients without follow-up MR, 9 had died or deteriorated due to neurologic decline. In four patients (15%) tumor increase was demonstrated by computer-assisted analysis; in the other 23 patients tumor remained stable (15%) or decreased (70%). In 11 of the 27 patients, a second follow-up MR was obtained 26 wk after WBRT demonstrating tumor increase in seven patients (64%). Complete discordance between both assessment methods was noticed in only 4 of 38 MR scans (27 scans 6 wk after WBRT and 11 scans 26 wk after WBRT) with small quantitative changes in tumor volume. In 26 of 38 (68%) observations 6 and 26 wk after WBRT, there was a positive correlation between change in tumor volume and change in the Neurologic Status. Visual estimation analysis seems an appropriate method of routine therapy monitoring in individual patients with brain metastases. The computer-assisted analysis should be considered if precise response monitoring becomes critical.

Structural abnormalities in brain magnetic resonance images of depressed children

OBJECTIVE

Brain magnetic resonance images (MRIs) of 65 children and adolescents who were hospitalized with depressive disorders (DD) were compared with the brain MRIs of 18 hospitalized psychiatric controls (PC) without a depressive disorder.

METHOD

Volumetric analyses were used to measure frontal lobe volumes (FLV), lateral ventricular volumes (VV), and total cerebral volumes (CV) for all subjects. To correct for differences in absolute cerebral volume associated with different body and head size, the ratios of FLV/CV and VV/CV were used to compare differences between the two groups. A multivariate analysis was used to control for the effects of several independent variables (age, sex, diagnosis).

RESULTS

Significant differences were seen in the FLV/CV ratio and the VV/CV ratio when the results were compared between the two groups (DD versus PC). The DD group had a significantly smaller FLV/CV ratio (t = 2.148, df = 79, p = .035) and a significantly larger VV/CV ratio (t = -2.093, df = 79, p = .040).

CONCLUSION

The findings are consistent with previous reports in depressed adults and may implicate the frontal lobes in the pathogenesis of early-onset depressive disorders.

Attention-deficit hyperactivity disorder: magnetic resonance imaging morphometric analysis of the corpus callosum

OBJECTIVE

The following study seeks to document possible differences in corpus callosal area and shape between children with attention-deficit hyperactivity disorder (ADHD) and controls.

METHODS

Fifteen carefully diagnosed right-handed male subjects with ADHD with overactivity symptomatology were compared to 15 right-handed male control subjects. The corpus callosum was divided into seven areas on the midsagittal slice of a magnetic resonance image with shape analysis also conducted.

RESULTS

An exploratory shape analysis showed no significant differences in shape between the groups. No group differences were found in the area, length, or anterior regions of the corpus callosum. The ADHD subjects were found to have significantly smaller posterior corpus callosum regions than the control group, with the splenium accounting for most of the variance between the groups.

CONCLUSIONS

The splenial area of the corpus callosum is smaller in children with ADHD than in a sample of normally developing children. These smaller areas may relate to commonly seen sustained attention deficits which in turn negatively impact on the development of more advanced levels of attention such as self-regulation. Further study of the regions surrounding the splenial area is suggested to determine whether they are correlated in size to the smaller corpus callosum.