Application of fuzzy-rule-based postprocessing to correlation methods in pattern recognition

One of the most important problems in optical pattern recognition by correlation is the appearance of sidelobes in the correlation plane, which causes false alarms. We present a method that eliminate sidelobes of up to a given height if certain conditions are satisfied. The method can be applied to any generalized synthetic discriminant function filter and is capable of rejecting lateral peaks that are even higher than the central correlation. Satisfactory results were obtained in both computer simulations and optical implementation.

False recognition following frontal lobe damage: The role of encoding factors

This paper reports a series of experiments on patient JB, a man with memory difficulties following damage to the left frontal lobe. The primary characteristic of JB's recognition memory impairment is a high level of false recognition together with a normal hit rate. The hypothesis that JB's false recognition reflects an over-reliance on familiarity is considered, but discounted on the basis that the false alarm rate is not affected by increasing the similarity between distracters and targets, and remains high when nonword stimuli are used. It is suggested, instead, that JB relies on a poorly focused memory description, which lacks item-specific detail but contains more general, low-level properties of the target items-these properties being held by many distracter items as well. This deficit is considered to arise because of damage to frontally mediated control processes involved in the selection of elements for memory encoding. An encoding deficit is supported by the fact that JB's false recognition is significantly reduced by orienting instructions, and is eliminated when his remote memory is subjected to recognition testing. In contrast, it is shown that manipulations at the level of retrieval (e.g. restricting the number of "old" responses) have little effect on his false recognition.

Dispersal (Entropy) and Recognition (Information) as Foundations of Emergence and Dissolvence

The objective of this essay is to reflect on a possible relation between entropy and emergence. A qualitative, relational approach is followed. We begin by highlighting that entropy includes the concept of dispersal, relevant to our enquiry. Emergence in complex systems arises from the coordinated behavior of their parts. Coordination in turn necessitates recognition between parts, i.e., information exchange. What will be argued here is that the scope of recognition processes between parts is increased when preceded by their dispersal, which multiplies the number of encounters and creates a richer potential for recognition. A process intrinsic to emergence is dissolvence (aka submergence or top-down constraints), which participates in the information-entropy interplay underlying the creation, evolution and breakdown of higher-level entities.

Chemodiversity and molecular plasticity: recognition processes as explored by property spaces.

In the last few years, a need to account for molecular flexibility in drug-design methodologies has emerged, even if the dynamic behavior of molecular properties is seldom made explicit. For a flexible molecule, it is indeed possible to compute different values for a given conformation-dependent property and the ensemble of such values defines a property space that can be used to describe its molecular variability; a most representative case is the lipophilicity space. In this review, a number of applications of lipophilicity space and other property spaces are presented, showing that this concept can be fruitfully exploited: to investigate the constraints exerted by media of different levels of structural organization, to examine processes of molecular recognition and binding at an atomic level, to derive informative descriptors to be included in quantitative structure--activity relationships and to analyze protein simulations extracting the relevant information. Much molecular information is neglected in the descriptors used by medicinal chemists, while the concept of property space can fill this gap by accounting for the often-disregarded dynamic behavior of both small ligands and biomacromolecules. Property space also introduces some innovative concepts such as molecular sensitivity and plasticity, which appear best suited to explore the ability of a molecule to adapt itself to the environment variously modulating its property and conformational profiles. Globally, such concepts can enhance our understanding of biological phenomena providing fruitful descriptors in drug-design and pharmaceutical sciences.

Functional analysis of HLA-A*0201/Melan-A peptide multimer+ CD8+ T cells isolated from an HLA-A*0201- donor: exploring tumor antigen allorestricted recognition.

Recent studies in mouse models have suggested that genetic transfer of tumor antigen-specific high affinity T cell receptors (TCR) into host lymphocytes could be a viable strategy for the rapid induction of tumor-specific immunity. A previously proposed approach for the isolation of such TCRs consists in circumventing tolerance to self-restricting HLA/peptide complexes by deriving them from PMBCs of allogenic donors. Towards this aim, we used fluorescent HLA-A2 class-I/peptide soluble multimers to isolate A2-restricted CD8+ T cells specific for a previously described Melan-A peptide enhanced analog (Melan-A 26-35 A27L) from an HLA-A*0201 (A2) negative donor. We isolated two distinct groups of Melan-A 26-35 A27L-specific clones. Clones from the first group recognized the analog peptide with high avidity but showed very low recognition of Melan-A parental peptides. In contrast, clones from the second group efficiently recognized Melan-A parental peptides. Surprisingly however, most clones recognized not only A2+ Melan-A+ targets, but also A2+ Melan-A- targets suggesting that they can also recognize endogenous peptides other than Melan-A. In addition, one clone showed full cross-recognition of an antigenically unrelated peptide. Together, our data show that HLA-A2/peptide multimers can be successfully used for the isolation of allorestricted CD8+ T cells reactive with tumor antigen-derived peptides. However, as the cross-reactivity of these apparently peptide-specific allorestricted TCRs is presently unpredictable, a careful in vitro analysis of their reactivity to the host's normal cells is recommended.

Prediction of cross-recognition of peptide-HLA A2 by Melan-A-specific cytotoxic T lymphocytes using three-dimensional quantitative structure-activity relationships.

The cross-recognition of peptides by cytotoxic T lymphocytes is a key element in immunology and in particular in peptide based immunotherapy. Here we develop three-dimensional (3D) quantitative structure-activity relationships (QSARs) to predict cross-recognition by Melan-A-specific cytotoxic T lymphocytes of peptides bound to HLA A*0201 (hereafter referred to as HLA A2). First, we predict the structure of a set of self- and pathogen-derived peptides bound to HLA A2 using a previously developed ab initio structure prediction approach [Fagerberg et al., J. Mol. Biol., 521-46 (2006)]. Second, shape and electrostatic energy calculations are performed on a 3D grid to produce similarity matrices which are combined with a genetic neural network method [So et al., J. Med. Chem., 4347-59 (1997)] to generate 3D-QSAR models. The models are extensively validated using several different approaches. During the model generation, the leave-one-out cross-validated correlation coefficient (q (2)) is used as the fitness criterion and all obtained models are evaluated based on their q (2) values. Moreover, the best model obtained for a partitioned data set is evaluated by its correlation coefficient (r = 0.92 for the external test set). The physical relevance of all models is tested using a functional dependence analysis and the robustness of the models obtained for the entire data set is confirmed using y-randomization. Finally, the validated models are tested for their utility in the setting of rational peptide design: their ability to discriminate between peptides that only contain side chain substitutions in a single secondary anchor position is evaluated. In addition, the predicted cross-recognition of the mono-substituted peptides is confirmed experimentally in chromium-release assays. These results underline the utility of 3D-QSARs in peptide mimetic design and suggest that the properties of the unbound epitope are sufficient to capture most of the information to determine the cross-recognition.

Recognition of major histoincompatibilities after transplantation with marrow from HLA closely matched donors.

BACKGROUND: To determine the extent to which major histoincompatibilities are recognized after bone marrow transplantation, we characterized the specificity of the cytotoxic T lymphocytes isolated during graft-versus-host disease. We studied three patients transplanted with marrow from donors who were histoincompatible for different types of HLA antigens. METHODS: Patient 1 was mismatched for one "ABDR-antigen" (HLA-A2 versus A3). Two patients were mismatched for antigens that would usually not be taken into account by standard selection procedures: patient 2 was mismatched for an "HLA-A subtype" (A*0213 versus A*0201), whereas patient 3 was mismatched for HLA-C (HLA-C*0501 versus HLA-C*0701). All three HLA class I mismatches were detected by a pretransplant cytotoxic precursor test. RESULTS: Analysis of the specificity of the cytotoxic T lymphocyte clones isolated after transplantation showed that the incompatibilities detected by the pretransplant cytotoxic precursor assay were the targets recognized during graft-versus-host disease. CONCLUSIONS: Independent of whether the incompatibility consisted of a "full" mismatch, a "subtype" mismatch, or an HLA-C mismatch, all clones recognized the incompatible HLA molecule. In addition, some of these clones had undergone antigen selection and were clearly of higher specificity than the ones established before transplantation, indicating that they had been participating directly in the antihost immune response.

Low TCR avidity and lack of tumor cell recognition in CD8(+) T cells primed with the CEA-analogue CAP1-6D peptide.

The use of "altered peptide ligands" (APL), epitopes designed for exerting increased immunogenicity as compared with native determinants, represents nowadays one of the most utilized strategies for overcoming immune tolerance to self-antigens and boosting anti-tumor T cell-mediated immune responses. However, the actual ability of APL-primed T cells to cross-recognize natural epitopes expressed by tumor cells remains a crucial concern. In the present study, we show that CAP1-6D, a superagonist analogue of a carcinoembriyonic antigen (CEA)-derived HLA-A*0201-restricted epitope widely used in clinical setting, reproducibly promotes the generation of low-affinity CD8(+) T cells lacking the ability to recognized CEA-expressing colorectal carcinoma (CRC) cells. Short-term T cell cultures, obtained by priming peripheral blood mononuclear cells from HLA-A*0201(+) healthy donors or CRC patients with CAP1-6D, were indeed found to heterogeneously cross-react with saturating concentrations of the native peptide CAP1, but to fail constantly lysing or recognizing through IFN- gamma release CEA(+)CRC cells. Characterization of anti-CAP1-6D T cell avidity, gained through peptide titration, CD8-dependency assay, and staining with mutated tetramers (D227K/T228A), revealed that anti-CAP1-6D T cells exerted a differential interaction with the two CEA epitopes, i.e., displaying high affinity/CD8-independency toward the APL and low affinity/CD8-dependency toward the native CAP1 peptide. Our data demonstrate that the efficient detection of self-antigen expressed by tumors could be a feature of high avidity CD8-independent T cells, and underline the need for extensive analysis of tumor cross-recognition prior to any clinical usage of APL as anti-cancer vaccines.

Maximum likelihood Linear Programming Data Fusion for Speaker Recognition

Biometric system performance can be improved by means of data fusion. Several kinds of information can be fused in order to obtain a more accurate classification (identification or verification) of an input sample. In this paper we present a method for computing the weights in a weighted sum fusion for score combinations, by means of a likelihood model. The maximum likelihood estimation is set as a linear programming problem. The scores are derived from a GMM classifier working on a different feature extractor. Our experimental results assesed the robustness of the system in front a changes on time (different sessions) and robustness in front a change of microphone. The improvements obtained were significantly better (error bars of two standard deviations) than a uniform weighted sum or a uniform weighted product or the best single classifier. The proposed method scales computationaly with the number of scores to be fussioned as the simplex method for linear programming.

Application of the Mutual Information Minimization to speaker recognition / identification improvement

In this paper we propose the inversion of nonlinear distortions in order to improve the recognition rates of a speaker recognizer system. We study the effect of saturations on the test signals, trying to take into account real situations where the training material has been recorded in a controlled situation but the testing signals present some mismatch with the input signal level (saturations). The experimental results for speaker recognition shows that a combination of several strategies can improve the recognition rates with saturated test sentences from 80% to 89.39%, while the results with clean speech (without saturation) is 87.76% for one microphone, and for speaker identification can reduce the minimum detection cost function with saturated test sentences from 6.42% to 4.15%, while the results with clean speech (without saturation) is 5.74% for one microphone and 7.02% for the other one.

Improving a leaves automatic recognition process using PCA

In this work we present a simulation of a recognition process with perimeter characterization of a simple plant leaves as a unique discriminating parameter. Data coding allowing for independence of leaves size and orientation may penalize performance recognition for some varieties. Border description sequences are then used, and Principal Component Analysis (PCA) is applied in order to study which is the best number of components for the classification task, implemented by means of a Support Vector Machine (SVM) System. Obtained results are satisfactory, and compared with [4] our system improves the recognition success, diminishing the variance at the same time.

Automatic Recognition of Leaves by Shape Detection Pre-Processing with Ica

In this work we present a simulation of a recognition process with perimeter characterization of a simple plant leaves as a unique discriminating parameter. Data coding allowing for independence of leaves size and orientation may penalize performance recognition for some varieties. Border description sequences are then used to characterize the leaves. Independent Component Analysis (ICA) is then applied in order to study which is the best number of components to be considered for the classification task, implemented by means of an Artificial Neural Network (ANN). Obtained results with ICA as a pre-processing tool are satisfactory, and compared with some references our system improves the recognition success up to 80.8% depending on the number of considered independent components.

Empirical mode decomposition-based face recognition system

In this work we explore the multivariate empirical mode decomposition combined with a Neural Network classifier as technique for face recognition tasks. Images are simultaneously decomposed by means of EMD and then the distance between the modes of the image and the modes of the representative image of each class is calculated using three different distance measures. Then, a neural network is trained using 10- fold cross validation in order to derive a classifier. Preliminary results (over 98 % of classification rate) are satisfactory and will justify a deep investigation on how to apply mEMD for face recognition.

Auditory localisation and recognition after left inferior collicular lesion: effects of work load on cortical activation patterns (fMRI study)

Evidence from neuropsychological and activation studies (Clarke et al., 2oo0, Maeder et al., 2000) suggests that sound recognitionand localisation are processed by two anatomically and functionally distinct cortical networks. We report here on a case of a patientthat had an interruption of auditory information and we show: i) the effects of this interruption on cortical auditory processing; ii)the effect of the workload on activation pattern.A 36 year old man suffered from a small left mesencephalic haemotrhage, due to cavernous angioma; the let% inferior colliculuswas resected in the surgical approach of the vascular malformation. In the acute stage, the patient complained of auditoryhallucinations and of auditory loss in right ear, while tonal audiometry was normal. At 12 months, auditory recognition, auditorylocalisation (assessed by lTD and IID cues) and auditory motion perception were normal (Clarke et al., 2000), while verbal dichoticlistening was deficient on the right side.Sound recognition and sound localisation activation patterns were investigated with fMRI, using a passive and an activeparadigm. In normal subjects, distinct cortical networks were involved in sound recognition and localisation, both in passive andactive paradigm (Maeder et al., 2OOOa, 2000b).Passive listening of environmental and spatial stimuli as compared to rest strongly activated right auditory cortex, but failed toactivate left primary auditory cortex. The specialised networks for sound recognition and localisation could not be visual&d onthe right and only minimally on the left convexity. A very different activation pattern was obtained in the active condition wherea motor response was required. Workload not only increased the activation of the right auditory cortex, but also allowed theactivation of the left primary auditory cortex. The specialised networks for sound recognition and localisation were almostcompletely present in both hemispheres.These results show that increasing the workload can i) help to recruit cortical region in the auditory deafferented hemisphere;and ii) lead to processing auditory information within specific cortical networks.References:Clarke et al. (2000). Neuropsychologia 38: 797-807.Mae.der et al. (2OOOa), Neuroimage 11: S52.Maeder et al. (2OOOb), Neuroimage 11: S33