Comparison of Energy Minimization Methods for 3-D Brain Tissue Classification

This paper presents 3-D brain tissue classificationschemes using three recent promising energy minimizationmethods for Markov random fields: graph cuts, loopybelief propagation and tree-reweighted message passing.The classification is performed using the well knownfinite Gaussian mixture Markov Random Field model.Results from the above methods are compared with widelyused iterative conditional modes algorithm. Theevaluation is performed on a dataset containing simulatedT1-weighted MR brain volumes with varying noise andintensity non-uniformities. The comparisons are performedin terms of energies as well as based on ground truthsegmentations, using various quantitative metrics.

Assessment of a new MRI classification of spinal stenosis based on cross-sectional morphology of the dural sac : FM2

Introduction: Quantitative measures of degree of lumbar spinal stenosis (LSS) such as antero-posterior diameter of the canal or dural sac cross sectional area vary widely and do not correlate with clinical symptoms or results of surgical decompression. In an effort to improve quantification of stenosis we have developed a grading system based on the morphology of the dural sac and its contents as seen on T2 axial images. The grading comprises seven categories ranging form normal to the most severe stenosis and takes into account the ratio of rootlet/CSF content. Material and methods: Fifty T2 axial MRI images taken at disc level from twenty seven symptomatic lumbar spinal stenosis patients who underwent decompressive surgery were classified into seven categories by five observers and reclassified 2 weeks later by the same investigators. Intra- and inter-observer reliability of the classification were assessed using Cohen's and Fleiss' kappa statistics, respectively. Results: Generally, the morphology grading system itself was well adopted by the observers. Its success in application is strongly influenced by the identification of the dural sac. The average intraobserver Cohen's kappa was 0.53 ± 0.2. The inter-observer Fleiss' kappa was 0.38 ± 0.02 in the first rating and 0.3 ± 0.03 in the second rating repeated after two weeks. Discussion: In this attempt, the teaching of the observers was limited to an introduction to the general idea of the morphology grading system and one example MRI image per category. The identification of the dimension of the dural sac may be a difficult issue in absence of complete T1 T2 MRI image series as it was the case here. The similarity of the CSF to possibly present fat on T2 images was the main reason of mismatch in the assignment of the cases to a category. The Fleiss correlation factors of the five observers are fair and the proposed morphology grading system is promising.

Morphometric shape analysis using learning vector quantization neural networks - an example distinguishing two microtine vole species

Closely related species may be very difficult to distinguish morphologically, yet sometimes morphology is the only reasonable possibility for taxonomic classification. Here we present learning-vector-quantization artificial neural networks as a powerful tool to classify specimens on the basis of geometric morphometric shape measurements. As an example, we trained a neural network to distinguish between field and root voles from Procrustes transformed landmark coordinates on the dorsal side of the skull, which is so similar in these two species that the human eye cannot make this distinction. Properly trained neural networks misclassified only 3% of specimens. Therefore, we conclude that the capacity of learning vector quantization neural networks to analyse spatial coordinates is a powerful tool among the range of pattern recognition procedures that is available to employ the information content of geometric morphometrics.

EORTC-ROG expert opinion: radiotherapy volume and treatment guidelines for neoadjuvant radiation of adenocarcinomas of the gastroesophageal junction and the stomach.

PURPOSE: The Gastro-Intestinal Working Party of the EORTC Radiation Oncology Group (GIWP-ROG) developed guidelines for target volume definition in neoadjuvant radiation of adenocarcinomas of the gastroesophageal junction (GEJ) and the stomach. METHODS AND MATERIALS: Guidelines about the definition of the clinical target volume (CTV) are based on a systematic literature review of the location and frequency of local recurrences and lymph node involvement in adenocarcinomas of the GEJ and the stomach. Therefore, MEDLINE was searched up to August 2008. Guidelines concerning prescription, planning and treatment delivery are based on a consensus between the members of the GIWP-ROG. RESULTS: In order to support a curative resection of GEJ and gastric cancer, an individualized preoperative treatment volume based on tumour location has to include the primary tumour and the draining regional lymph nodes area. Therefore we recommend to use the 2nd English Edition of the Japanese Classification of Gastric Carcinoma of the Japanese Gastric Cancer Association which developed the concept of assigning tumours of the GEJ and the stomach to anatomically defined sub-sites corresponding respectively to a distinct lymphatic spread pattern. CONCLUSION: The GIWP-ROG defined guidelines for preoperative irradiation of adenocarcinomas of the GEJ and the stomach to reduce variability in the framework of future clinical trials.

On the efficiency and sensitivity of a pyramidal classification algorithm

In this paper we propose a Pyramidal Classification Algorithm,which together with an appropriate aggregation index producesan indexed pseudo-hierarchy (in the strict sense) withoutinversions nor crossings. The computer implementation of thealgorithm makes it possible to carry out some simulation testsby Monte Carlo methods in order to study the efficiency andsensitivity of the pyramidal methods of the Maximum, Minimumand UPGMA. The results shown in this paper may help to choosebetween the three classification methods proposed, in order toobtain the classification that best fits the original structureof the population, provided we have an a priori informationconcerning this structure.

Adaptive model selection using empirical complexities

Given $n$ independent replicates of a jointly distributed pair $(X,Y)\in {\cal R}^d \times {\cal R}$, we wish to select from a fixed sequence of model classes ${\cal F}_1, {\cal F}_2, \ldots$ a deterministic prediction rule $f: {\cal R}^d \to {\cal R}$ whose risk is small. We investigate the possibility of empirically assessingthe {\em complexity} of each model class, that is, the actual difficulty of the estimation problem within each class. The estimated complexities are in turn used to define an adaptive model selection procedure, which is based on complexity penalized empirical risk.The available data are divided into two parts. The first is used to form an empirical cover of each model class, and the second is used to select a candidate rule from each cover based on empirical risk. The covering radii are determined empirically to optimize a tight upper bound on the estimation error. An estimate is chosen from the list of candidates in order to minimize the sum of class complexity and empirical risk. A distinguishing feature of the approach is that the complexity of each model class is assessed empirically, based on the size of its empirical cover.Finite sample performance bounds are established for the estimates, and these bounds are applied to several non-parametric estimation problems. The estimates are shown to achieve a favorable tradeoff between approximation and estimation error, and to perform as well as if the distribution-dependent complexities of the model classes were known beforehand. In addition, it is shown that the estimate can be consistent,and even possess near optimal rates of convergence, when each model class has an infinite VC or pseudo dimension.For regression estimation with squared loss we modify our estimate to achieve a faster rate of convergence.

Classification of cell death: recommendations of the Nomenclature Committee on Cell Death 2009

Different types of cell death are often defined by morphological criteria, without a clear reference to precise biochemical mechanisms. The Nomenclature Committee on Cell Death (NCCD) proposes unified criteria for the definition of cell death and of its different morphologies, while formulating several caveats against the misuse of words and concepts that slow down progress in the area of cell death research. Authors, reviewers and editors of scientific periodicals are invited to abandon expressions like 'percentage apoptosis' and to replace them with more accurate descriptions of the biochemical and cellular parameters that are actually measured. Moreover, at the present stage, it should be accepted that caspase-independent mechanisms can cooperate with (or substitute for) caspases in the execution of lethal signaling pathways and that 'autophagic cell death' is a type of cell death occurring together with (but not necessarily by) autophagic vacuolization. This study details the 2009 recommendations of the NCCD on the use of cell death-related terminology including 'entosis', 'mitotic catastrophe', 'necrosis', 'necroptosis' and 'pyroptosis'.

Description and phylogenetic analysis of the Calycopidina (Lepidoptera, Lycaenidae, Theclinae, Eumaeini): a subtribe of detritivores

Description and phylogenetic analysis of the Calycopidina (Lepidoptera, Lycaenidae, Theclinae, Eumaeini): a subtribe of detritivores. The purpose of this paper is to establish a phylogenetic basis for a new Eumaeini subtribe that includes those lycaenid genera in which detritivory has been recorded. Morphological characters were coded for 82 species of the previously proposed "Lamprospilus Section" of the Eumaeini (19 of these had coding identical to another species), and a phylogenetic analysis was performed using the 63 distinct ingroup terminal taxa and six outgroups belonging to four genera. Taxonomic results include the description in the Eumaeini of Calycopidina Duarte & Robbins new subtribe (type genus Calycopis Scudder, 1876), which contains Lamprospilus Geyer, Badecla Duarte & Robbins new genus (type species Thecla badaca Hewitson), Arzecla Duarte & Robbins new genus (type species Thecla arza Hewitson), Arumecla Robbins & Duarte, Camissecla Robbins & Duarte, Electrostrymon Clench, Rubroserrata K. Johnson & Kroenlein revalidated status, Ziegleria K. Johnson, Kisutam K. Johnson & Kroenlein revalidated status, and Calycopis. Previous "infratribe" names Angulopina K. Johnson & Kroenlein, 1993, and Calycopina K. Johnson & Kroenlein, 1993, are nomenclaturally unavailable and polyphyletic as proposed. New combinations include Badecla badaca (Hewitson), Badecla picentia (Hewitson), Badecla quadramacula (Austin & K. Johnson), Badecla lanckena (Schaus), Badecla argentinensis (K. Johnson & Kroenlein), Badecla clarissa (Draudt), Arzecla arza (Hewitson), Arzecla tarpa (Godman & Salvin), Arzecla canacha (Hewitson), Arzecla calatia (Hewitson), Arzecla tucumanensis (K. Johnson & Kroenlein), Arzecla sethon (Godman & Salvin), Arzecla nubilum (H. H. Druce), Arzecla paralus (Godman & Salvin), Arzecla taminella (Schaus), Arzecla albolineata (Lathy), Electrostrymon denarius (Butler & H.Druce), Electrostrymon guzanta (Schaus), Electrostrymon perisus (H. H. Druce), Rubroserrata mathewi (Hewitson), Rubroserrata ecbatana (Hewitson), Kisutam micandriana (K. Johnson), and Kisutam syllis (Godman & Salvin). The structure of the male genitalia lateral window, labides, and brush organs are described and discussed, as are the female genitalia signa of the corpus bursae and 8th abdominal tergum. Widespread wing pattern sexual dimorphism in the Calycopidina is noted and illustrated, and the presence of alternating dark and light bands on the ventral wings of both sexes is discussed. The evidence for detritivory in Lamprospilus, Badecla, Arzecla, Arumecla, Camissecla, Electrostrymon, Ziegleria, Kisutam, and Calycopis is summarized using the new classification.

Pattern of attack of a galling insect reveals an unexpected preference-performance linkage on medium-sized resources

Pattern of attack of a galling insect reveals an unexpected preference-performance linkage on medium-sized resources. The Plant Vigor Hypothesis (PVH) predicts oviposition preference and higher offspring performance on longer and fast-growing shoots, and although several studies have tested its predictions, long-term studies concerning the patterns of host selection by galling species are still lacking. The PVH was tested in this study using Bauhinia brevipes (Fabaceae) as the host of a leaf gall midge, Asphondylia microcapillata (Diptera, Cecidomyiidae) during three consecutive years. Shoots were collected from the same 80 plants between 2001 and 2003 and shoot length, number of healthy and galled leaves, gall number, and mortality factors were recorded. Nearly 600 galls were found on the 5,800 shoots collected. Medium-sized shoots supported from 46 to 70% of all galls, with greater gall survival rate in 2002 and 2003. A decrease in parasitism rate coupled with an increase in gall predation lead to a constant similar gall survivorship rate in all years (x = 22.7%). Although gall abundance varied among years (122 in 2001, 114 in 2002 and 359 in 2003) preference for longer shoots was not observed because the percentage of galled shoots and galled leaves were higher on medium shoot length classes in all years. The observed distribution of gall abundance and galled shoots were always greater than the expected distribution on medium shoot length classes. These findings do not support the PVH, and show that A. microcapillata can maximize the female preference and larval performance on medium-sized shoots of B. brevipes.

Automatic classification of MR scans in Alzheimer's disease.

To be diagnostically useful, structural MRI must reliably distinguish Alzheimer's disease (AD) from normal aging in individual scans. Recent advances in statistical learning theory have led to the application of support vector machines to MRI for detection of a variety of disease states. The aims of this study were to assess how successfully support vector machines assigned individual diagnoses and to determine whether data-sets combined from multiple scanners and different centres could be used to obtain effective classification of scans. We used linear support vector machines to classify the grey matter segment of T1-weighted MR scans from pathologically proven AD patients and cognitively normal elderly individuals obtained from two centres with different scanning equipment. Because the clinical diagnosis of mild AD is difficult we also tested the ability of support vector machines to differentiate control scans from patients without post-mortem confirmation. Finally we sought to use these methods to differentiate scans between patients suffering from AD from those with frontotemporal lobar degeneration. Up to 96% of pathologically verified AD patients were correctly classified using whole brain images. Data from different centres were successfully combined achieving comparable results from the separate analyses. Importantly, data from one centre could be used to train a support vector machine to accurately differentiate AD and normal ageing scans obtained from another centre with different subjects and different scanner equipment. Patients with mild, clinically probable AD and age/sex matched controls were correctly separated in 89% of cases which is compatible with published diagnosis rates in the best clinical centres. This method correctly assigned 89% of patients with post-mortem confirmed diagnosis of either AD or frontotemporal lobar degeneration to their respective group. Our study leads to three conclusions: Firstly, support vector machines successfully separate patients with AD from healthy aging subjects. Secondly, they perform well in the differential diagnosis of two different forms of dementia. Thirdly, the method is robust and can be generalized across different centres. This suggests an important role for computer based diagnostic image analysis for clinical practice.

Generation of the primary hair follicle pattern.

Hair follicles are spaced apart from one another at regular intervals through the skin. Although follicles are predominantly epidermal structures, classical tissue recombination experiments indicated that the underlying dermis defines their location during development. Although many molecules involved in hair follicle formation have been identified, the molecular interactions that determine the emergent property of pattern formation have remained elusive. We have used embryonic skin cultures to dissect signaling responses and patterning outcomes as the skin spatially organizes itself. We find that ectodysplasin receptor (Edar)-bone morphogenetic protein (BMP) signaling and transcriptional interactions are central to generation of the primary hair follicle pattern, with restriction of responsiveness, rather than localization of an inducing ligand, being the key driver in this process. The crux of this patterning mechanism is rapid Edar-positive feedback in the epidermis coupled with induction of dermal BMP4/7. The BMPs in turn repress epidermal Edar and hence follicle fate. Edar activation also induces connective tissue growth factor, an inhibitor of BMP signaling, allowing BMP action only at a distance from their site of synthesis. Consistent with this model, transgenic hyperactivation of Edar signaling leads to widespread overproduction of hair follicles. This Edar-BMP activation-inhibition mechanism appears to operate alongside a labile prepattern, suggesting that Edar-mediated stabilization of beta-catenin active foci is a key event in determining definitive follicle locations.

Distributional patterns of the Neotropical genus Thecomyia Perty (Diptera, Sciomyzidae) and phylogenetic support

Distributional patterns of the Neotropical genus Thecomyia Perty (Diptera, Sciomyzidae) and phylogenetic support. The distributional pattern of the genus Thecomyia Perty, 1833 was defined using panbiogeographic tools, and analyzed based on the phylogeny of the group. This study sought to establish biogeographical homologies in the Neotropical region between different species of the genus, based on their distribution pattern and later corroboration through its phylogeny. Eight individual tracks and 16 generalized tracks were identified, established along nearly the entire swath of the Neotropics. Individual tracks are the basic units of a panbiogeographic study, and correspond to the hypothesis of minimum distribution of the organisms involved. The generalized tracks, obtained from the spatial congruence between two or more individual tracks, are important in the identification of smaller areas of endemism. Thus, we found evidence from the generalized tracks in support of previous classification for the Neotropical region. The Amazon domain is indicated as an area of outstanding importance in the diversification of the group, by the confluence of generalized tracks and biogeographic nodes in the region. Most of the generalized tracks and biogeographical nodes were congruent with the phylogenetic hypothesis of the genus, indicating support of the primary biogeographical homologies originally defined by the track analysis.

Saving our science from ourselves: the plight of biological classification

Saving our science from ourselves: the plight of biological classification. Biological classification ( nomenclature, taxonomy, and systematics) is being sold short. The desire for new technologies, faster and cheaper taxonomic descriptions, identifications, and revisions is symptomatic of a lack of appreciation and understanding of classification. The problem of gadget-driven science, a lack of best practice and the inability to accept classification as a descriptive and empirical science are discussed. The worst cases scenario is a future in which classifications are purely artificial and uninformative.

A data-dependent skeleton estimate and a scale-sensitive dimension for classification

The classical binary classification problem is investigatedwhen it is known in advance that the posterior probability function(or regression function) belongs to some class of functions. We introduceand analyze a method which effectively exploits this knowledge. The methodis based on minimizing the empirical risk over a carefully selected``skeleton'' of the class of regression functions. The skeleton is acovering of the class based on a data--dependent metric, especiallyfitted for classification. A new scale--sensitive dimension isintroduced which is more useful for the studied classification problemthan other, previously defined, dimension measures. This fact isdemonstrated by performance bounds for the skeleton estimate in termsof the new dimension.