Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Feature selection”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 955 records · Page 53Linked to original sources

[Elective mutism in children: literature review].

This paper presents contemporary opinions about selective mutism in children, including epidemiology, etiology, clinical features and therapy. This is the first extensive review on this topic in Polish literature. The essential feature of selective mutism is persistent failure to speak in social situations, where speaking is expected (e.g., in school), despite speaking in other situations (e.g., at home). The authors present the diagnostic criteria according do DSM-IV and suggested by other authors. Clinical characteristics of this disorder were also presented, including personality traits and behaviour of mutistic children (different at home and in unfamiliar environment) and comorbidity of selective mutism. Etiology of this disorder seems to be multifactorial. The important etiological factors are: minimal brain dysfunction, somatic or psychological trauma, particularly during the speech development and a family structure, especially the mother-child relation. The authors emphasize that mutism in children is a heterogeneous symptom and present several models of mutism. The paper describes also different methods of treatment (e.g., behavioral, psychodynamic, family therapy and some case reports on pharmacotherapy); and long-term prognosis.

Age of Onset↗

[Treatment results and prognostic significance of selected clinical and laboratory features in children diagnosed with malignant lymphoma].

The clinical picture and results of treating malignant lymphoma in children, diagnosed and treated at the Pediatric Institute of Pomeranian Medical Academy in Szczecin during the period between May 1979 and February 1992, were analyzed. The studied group consisted of 33 children (23 boys, 10 girls) aged between 41-169 months (median 112 months, mean 110 months) having Hodgkin's lymphoma (HL), and 35 children (26 boys, 9 girls), aged between 35-171 months (median 101 months, mean 104 months), with non-Hodgkin's lymphoma (NHL). Till 1987 the children with HL were treated according to MOPP program, and since 1988 with MVPP/B-DOPA. Two children were treated according to COMP and ABVD programs. The NHL children were treated till 1985 with LSA2L2 or COAMP, and from 1986 with BFM-NHL 86 with the modification of methotrexate doses. The duration of observation involving HL cases ranged from 2 to 156 months (median 76, mean 78 months), that covering NHL cases from 3 to 153 months (median 28 months, mean 44 months). It was proved that the results of HL treatment in the Pediatric Institute of Pomeranian Medical Academy in Szczecin were comparable with the results of other centers. The probability of event free survival (EFS) for the whole group was 0.818, for children treated by MOPP program was 0.888, for children with MVPP/B-DOPA was 0.900. Unfortunately, the results of NHL treatment in our center in Szczecin are worse than those of other hematologic-oncologic institutions. The EFS was 0.550. The reason why our results were poor in treating NHL in our center was: delay in beginning the remission-inducing treatment because of diagnostic difficulties (especially in smaller hospitals): prolongation of the first remission-inducing therapy over 14 days, mainly due to generalized infection, generalized diathesis haemorrhagica with bleeding from the alimentary tract, and finally the need of modifying the treatment program BFM 86 concerning primarily the lowering of methotrexate doses from 5 g/m2 to 0.5 g/m2. That was necessary in view of our inability of monitoring the level of methotrexate in blood. All of those findings suggest the necessity of: 1) earlier proper diagnosis; the physicians taking care of children should be aware of high incidence of such neoplasms in children, especially with atypical clinical presentation; 2) full realization of the therapeutic program (particularly remission-inducing one). A general real improvement of the treatment conditions in hospitals is indispensable. The actual work has proved that in HL the detrimental prognostic factors included; the age above 10 years and histological type of nodular sclerosis. Children older than 10 years had lower EFS (1.0 vs 0.65; p < 0.05). EFS in histological type of nodular sclerosis was also lower namely (0.925 vs 0.600; p < 0.05). In the NHL group the bad prognostic factors were the age over 10 years and proliferation of T-cells. Patients older than 10 years displayed statistically lower EFS (0.709 vs 0.288; p < 0.05). The children with T-NHL had also lower EFS (0.621 vs 0.187; p < 0.05). It is necessary that the prognosis in these children should be substantially improved by elaborating treatment programme being adjusted to cope with the risk factors.

Adolescent↗

[Analysis of the effect of selected polymers on morphodynamic features of cells in Blepharisma undulans].

The studies presented enable to evaluate the survival rate of the protozoa in contact with the studied polymers (foils: with polyethylene PE, polyvinyl chloride P-25, P-47, polyester E1 and irradiated collagen L7). It can also be established whether the change of certain morphodynamic parameters could be used to evaluate biological usefulness of plastics. Stereometric method and factor analysis were used in the studies. On the basis of the obtained data polyvinyl chloride P-47 was well assessed, while irradiated collagen L7 showed most negative features.

Animals↗

Triple-targeted oncolytic adenoviruses featuring the cox2 promoter, E1A transcomplementation, and serotype chimerism for enhanced selectivity for ovarian cancer cells.

Conditionally replicating adenoviruses (CRAd's) feature selective replication in and killing of tumor cells. Initial clinical studies with relatively attenuated early generation agents have resulted in promising safety and efficacy data. Nevertheless, increased specificity may be advantageous for an emerging generation of infectivity-enhanced CRAd's. Further, increased specificity could translate into a larger tolerated dose. An approach for increasing specificity is dual control of E1A expression. We constructed six CRAd's featuring two variants of the cyclo-oxygenase 2 (cox2) promoter, combined with three versions of E1A. Transcriptional targeting was supplemented with transductional targeting utilizing the serotype 3 knob. In vivo and in vitro results suggest that cox2 can be utilized for enhancing the specificity of E1A deletion mutants and that combination with the Delta24 mutation increases specificity without reducing potency. Combination with Delta2-Delta24 was specific but somewhat attenuated. The promoter variants behaved similarly, although the longer 1,554-bp version displayed a trend for improved specificity. Transcriptional modifications were compatible with transductional targeting and resulted in up to 100,000-fold increase in the therapeutic window for Ad5/3cox2Ld24 vs wild-type adenovirus. Thus, the proposed triple-targeting strategy may be useful for increasing the safety and efficacy of adenoviral gene therapy for ovarian cancer.

Adenoviridae↗

Naming from definition: the role of feature type and feature distinctiveness.

The present paper evaluates the contribution of feature type and feature distinctiveness to naming of living and nonliving things using a naming from definition task. Normal subjects read verbal descriptions containing features varying in type (i.e., sensory vs. functional) and distinctiveness (i.e., distinct vs. shared) and were asked to name the concept described and to select the three features that most contributed to their answer. Main results showed that sensory features were selected more often than functional features to support naming living things and that, independent of feature type, more distinct features were selected to support naming more often than shared features. Results are discussed considering the implications for understanding naming and for neuropsychological evaluation.

Adult↗

Polymyositis with cytochrome oxidase negative muscle fibres. Early quadriceps weakness and poor response to immunosuppressive therapy.

We evaluated 10 patients with histologically typical polymyositis except for an excess of muscle fibres with absent cytochrome oxidase (COX) staining. No biopsies had vacuoles or congophilic material in muscle fibres. All patients presented with a history of slowly progressive weakness. The average age of onset was 9 years older than a group of polymyositis patients with normal COX staining of muscle fibres. Selective weakness of knee extension was a prominent and disabling feature in most patients. Serum creatine kinase was usually mildly elevated (363 +/- 115 U/l) but at levels lower than those in other patients with polymyositis. Mitochondrial DNA analysis showed multiple deletions in 90% of muscles from patients with excessive numbers of COX-negative muscle fibres, a prevalence significantly greater than the 22% figure for controls (P = 0.005). As a group, the patients responded poorly to immunosuppressive therapy. We conclude that patients with polymyositis and an excess of COX-negative muscle fibres, but no inclusion bodies, have common features including selective quadriceps weakness, mitochondrial pathology by histochemical and DNA analysis and a poor response to immunosuppressive therapy. Some of these features are shared with inclusion body myositis (IBM) and this entity cannot be entirely excluded as vacuoles may not be present in all muscle tissue in IBM patients. Evaluation of the COX activity in muscle fibres of patients with inflammatory myopathies provides useful prognostic information regarding the likelihood of improved strength after immunosuppressive treatment.

DNA, Mitochondrial↗

Clinico-pathological features of a series of 11 oncocytic endocrine tumours of the pancreas.

Oncocytic changes may occur very infrequently in neuroendocrine tumours. To estimate the prevalence, pathological features and clinical behaviour of oncocytic endocrine tumours of the pancreas, we reviewed a series of 227 lesions from two institutions. Eleven cases with predominant oncocytic features were selected, representing 4.85% of the whole series. The morphological features and immunophenotype of such tumours did not differ from conventional endocrine pancreatic tumours, except for the presence of abundant eosinophilic and granular cytoplasm. Anti-mitochondrial antigen was positive in all cases tested, and by electron microscopy, numerous mitochondria were observed in the cytoplasm of tumour cells. The majority of cases were nonfunctioning, and in most cases, pathologic signs of malignancy, leading to a diagnosis of endocrine carcinoma, were observed. In addition, the three nonmalignant cases matched the criteria of well-differentiated tumours of uncertain malignant potential. Nearly 50% of the cases were clinically aggressive, and lymph node and liver metastases were present at the time of diagnosis in a minority of cases. Therefore, oncocytic endocrine pancreatic tumours represent a peculiar morphological and clinical variant characterised by frequent hormonal inactivity and malignant behaviour, which should be considered in the differential diagnosis, especially when dealing with a metastatic lesion.

Adenoma, Oxyphilic↗

Basic points in and practical aspects of the application of diagnostic morphometry.

The number of applications of diagnostic morphometry is rapidly increasing. We have routinely used this technique for more than four years in our laboratory. Essential points in the application are: the type of lesions to which it is applied, multivariate analysis instead of single variate analysis and selection by a skilled pathologist of the relevant microscopical features. Such selective morphometry is useful in up to four percent of the total material entering an average pathology laboratory; in teaching laboratories the percentage may be considerably higher (e.g. 10-12%). Most of the applications are found in tumourpathology, but applications in non-tumor pathology are also promising. It is expected that diagnostic morphometry will become widely available and accepted in the next three to five years. Special symposia, diagnostic courses and organization of interested pathologists within the framework of the diagnostic morphometry workgroup of the European Society of Pathology is expected to be of help in this direction. Highly automatic, yet interactive, methods will become available within the next five to ten years. Further acceptance of these methods in most laboratories by 1990 is likely. A modular setup of quantitative microscopy equipment therefore is advisable. Turn-key, dedicated instruments for well-defined diagnostic problems are another (future) possibility. The diagnostic morphometry process as a product requires support and quality control. The pathologist is responsible for this and diagnostic morphometry should be seen as an integrated part of the diagnostic process.

Diagnosis↗

Ontogeny and phylogeny of femoro-tibial characters in humans and hominid fossils: functional influence and genetic determinism.

Three different human femoro-tibial characters are selected as functionally relevant and derived hominid characters: femoral bicondylar angle, shape of the femoral distal epiphysis, and the tibial insertion of the lateral meniscus. The timing and mode of formation of these characters are investigated during human ontogeny and are shown to differ considerably. The available hominid fossils (Australopithecus afarensis and early Homo) are interpreted in the light of this ontogenetic analysis with the conclusion that, during hominid evolution, different modes of selection of these features must have occurred. In modern humans, the femoral bicondylar angle proves to be an epigenetic functional feature, which develops during early childhood growth. It is present in all australopithecines and we suggest that it developed following a change in their locomotor behavior and not upon a genomic change: the early practice of bipedal walking, with adducted knee joints, in the locomotor repertoire of infant australopithecines, was sufficient to promote this angle. Later in hominid evolution, the knee joint evolved from having a single insertion of the lateral meniscus on the tibia to a double one. While Australopithecus afarensis exhibits a single insertion, early Homo clearly exhibits a double insertion of the lateral meniscus on the tibia. The double insertion restricts the mobility of the meniscus on the tibial plateau, indicating a habitual practice of full extension movements of the knee joint. Among modern humans, the posterior insertion of the lateral meniscus appears early in fetal life. Consequently in early Homo, this new selected feature developed directly as a result of a genomic change. The derived shape of human distal femoral epiphysis includes a prominence of the lateral lip of the femoral trochlea, an elliptical profile of the lateral condyle, and an anteroposterior lengthening of the epiphysis. Analysis of human fetal and neonatal distal epiphyses shows that the prominence of the lateral lip of the trochlea arises before any use, and thus appears to be genetically determined. However, the postnatal development of this joint shows that this feature is also modified epigenetically by use. It is argued that the hominid femoro-patellar joint would have been reshaped following the process of genetic assimilation (Waddington [1942] Nature 3811:563-565). The prominence of the lateral lip of the femoral trochlea was probably selected following a two-staged process-first epigenetic, then genetic. Far from being a Lamarckian explanation, this concept applies precisely to adaptive characters that are induced by an external stimulus during a single lifetime and are replaced through natural selection by genetically based equivalent characters. The nature of the structures involved in the studied features is shown to be an important parameter determining their mode of development and selection.

Adaptation, Physiological↗

Auditory selective attention in young and elderly adults: the selection of single versus conjoint features.

The effects of aging on the behavioral and neurophysiological correlates of auditory selective attention were investigated when selection was between either unidimensional or multidimensional stimuli. Attentional processes were studied by recording event-related brain potentials (ERPs) from 16 young (M = 22 years) and 16 elderly adults (M = 74 years) while they detected target tones based on a single location cue or a conjunction of location and pitch cues. Performance declined from the single- to the conjoint-cue task for both age groups but more so for the elderly. The ERP data showed that both age groups used a hierarchical processing strategy to perform the conjoint-cue task, but processing of the pitch dimension took longer for the elderly than for the young. The ERP data also showed that the scalp distribution of a late aspect of the waveform was more restricted in both anterior and posterior directions for the elderly. This suggests that frontal-lobe dependent attentional processes may be less efficacious with aging.

Acoustic Stimulation↗

Anatomic feature-based registration for patient set-up in head and neck cancer radiotherapy.

Modern radiotherapy equipment is capable of delivering high precision conformal dose distributions relative to isocentre. One of the barriers to precise treatments is accurate patient re-positioning before each fraction of treatment. At Massachusetts General Hospital, we perform daily patient alignment using radiographs, which are captured by flat panel imaging devices and sent to an analysis program. A trained therapist manually selects anatomically significant features in the skeleton, and couch movement is computed based on the image coordinates of the features. The current procedure takes about 5 to 10 min and significantly affects the efficiency requirement in a busy clinic. This work presents our effort to develop an improved, semi-automatic procedure that uses the manually selected features from the first treatment fraction to automatically locate the same features on the second and subsequent fractions. An implementation of this semi-automatic procedure is currently in clinical use for head and neck tumour sites. Radiographs collected from 510 patient set-ups were used to test this algorithm. A mean difference of 1.5 mm between manual and automatic localization of individual features and a mean difference of 0.8 mm for overall set-up were seen.

Head and Neck Neoplasms↗

On the use of neural network techniques to analyze sleep EEG data. Third communication: robustification of the classificator by applying an algorithm obtained from 9 different networks.

This is the third communication on the use of neural network techniques to classify sleep stages. In our first communication we presented the algorithms and the selection of the feature space and its reduction by using evolutionary and genetic procedures. In our second communication we trained the evolutionary optimized networks on the basis of multiple subject data in context with some smoothing algorithms in analogy of Rechtschaffen and Kales (RK). In this third communication we could demonstrate that the robustness concerning individual specific features of automatically generated sleep profiles could be reasonably improved by an additional modification of the procedure used by SASCIA (Sleep Analysis System to Challenge Innovative Artificial Networks). The outputs of nine different networks that were created by the data of 9 different subjects were used simultaneously for classification. The medians of the values obtained in each output measure were selected for the allocation to a sleep stage. The fitness criteria of 16 automatically generated sleep profiles showed reasonable concordance with the expert profile. Even though in single cases the concordance between conventional RK classifications and automatically generated profiles were a few percentages lower, the average correct classification of the 12 classified subjects improved substantially, thus proving that the classifier is more robust against individuum-specific variability. Despite the fact that the expert generally employs three channels (EEG, EMG and EOG), at least to build up sleep profiles, the SASCIA system was able to produce profiles on the basis of only one EEG channel with 80% concordance and a correlation coefficient of 0.86. The feature selections were performed by genetic algorithms and the topologies of the networks were optimized by evolutionary algorithms. This algorithm will now be used for larger sample forward classification.

Algorithms↗

Selective brainstem injury in an asphyxiated newborn.

We report the clinical, radiological, and neuropathological features of selective brainstem injury in an asphyxiated term infant. Disproportionate injury to thalamus, basal ganglia, and brainstem with relative sparing of cortex and subcortical white matter is observed occasionally after acute total asphyxia. Although this entity has been well documented neuropathologically, the specific computed tomographic findings have not been reported previously to our knowledge, and the clinical correlation is recognized rarely.

Asphyxia Neonatorum↗

Relation between MRI features and dementia in cerebrovascular disease patients with leukoaraiosis: a longitudinal study.

We examined selective MRI features (localization and degree of white matter abnormalities, cortical and subcortical atrophy) in relation to cognitive decline in patients with cerebrovascular disease (CVD) and leukoaraiosis (LA). We enrolled 6 female and 18 male CVD patients (mean age 66.2 +/- 6.6 years) whose Magnetic Resonance Images (MRI) revealed LA and who displayed a history of stroke or TIA; none showed signs of cortical infarcts or normal pressure hydrocephalus. Two blind raters independently scored MRI scans with a high level of agreement. All patients underwent extensive clinico-neuropsychological assessment upon admission to the study and 19 were followed for an average of 48 +/- 7.6 months. Twelve patients were initially classified as non-demented and 12 as demented. Three years later, one in the former group had become demented and mental impairment had worsened for 6 patients in the latter group; these 7 subjects were labeled as "decliners". Ventricular indexes were significantly higher in the demented group and correlated with severity of mental impairment, while the degree of LA was similar in demented and non-demented subjects. Neither white matter lesions nor sulcal and ventricular enlargement differed statistically between decliners and non-decliners.

Aged↗

Cytopathology of malignant mesothelioma: a stepwise logistic regression analysis.

Twenty-four cytologic features, previously reported to be useful in the distinction of malignant mesothelioma, adenocarcinoma, and benign mesothelial proliferation in serous effusions were assessed. Forty-four cases of malignant mesotheliomas, 46 cases of metastatic adenocarcinomas, and 30 cases of benign mesothelial proliferations were examined for these parameters. When these cytologic features were subjected to a stepwise logistic regression analysis, five features were selected to distinguish malignant mesothelioma from adenocarcinoma. These were true papillary aggregates, multinucleation with atypia, cell-to-cell apposition, acinus-like structures, and balloon-like vacuolation, the latter two features being characteristic of adenocarcinoma. The four variables selected to distinguish malignant mesothelioma from benign mesothelial proliferations were nuclear pleomorphism, macronucleoli, cell-in-cell engulfment, and monolayer cell groups, the latter being a feature of benign proliferations. Using these selected variables, the logistic model correctly predicted 95.4% of cases of malignant mesothelioma versus 100% of adenocarcinoma and 100% of malignant mesotheliomas versus 90% of benign mesothelial proliferations. The results of regression analysis suggest that many of the previously described cytologic features are not important diagnostic discriminators.

Adenocarcinoma↗

Discovering functional gene expression patterns in the metabolic network of Escherichia coli with wavelets transforms.

BACKGROUND: Microarray technology produces gene expression data on a genomic scale for an endless variety of organisms and conditions. However, this vast amount of information needs to be extracted in a reasonable way and funneled into manageable and functionally meaningful patterns. Genes may be reasonably combined using knowledge about their interaction behaviour. On a proteomic level, biochemical research has elucidated an increasingly complete image of the metabolic architecture, especially for less complex organisms like the well studied bacterium Escherichia coli. RESULTS: We sought to discover central components of the metabolic network, regulated by the expression of associated genes under changing conditions. We mapped gene expression data from E. coli under aerobic and anaerobic conditions onto the enzymatic reaction nodes of its metabolic network. An adjacency matrix of the metabolites was created from this graph. A consecutive ones clustering method was used to obtain network clusters in the matrix. The wavelet method was applied on the adjacency matrices of these clusters to collect features for the classifier. With a feature extraction method the most discriminating features were selected. We yielded network sub-graphs from these top ranking features representing formate fermentation, in good agreement with the anaerobic response of hetero-fermentative bacteria. Furthermore, we found a switch in the starting point for NAD biosynthesis, and an adaptation of the l-aspartate metabolism, in accordance with its higher abundance under anaerobic conditions. CONCLUSION: We developed and tested a novel method, based on a combination of rationally chosen machine learning methods, to analyse gene expression data on the basis of interaction data, using a metabolic network of enzymes. As a case study, we applied our method to E. coli under oxygen deprived conditions and extracted physiologically relevant patterns that represent an adaptation of the cells to changing environmental conditions. In general, our concept may be transferred to network analyses on biological interaction data, when data for two comparable states of the associated nodes are made available.

Algorithms↗

Real-time signal processing in speech recognition and its potential use within the development of hearing aids.

Acoustic-phonetic features necessary for rule-based speech recognition are described. The selection of feature algorithms is based on their robustness to variability of dynamics in speech signals and to influence of environmental noise, and their suitability for real-time implementation using speech production or speech perception modelling. Features are estimated in real-time, using a 32-bit floating point signal processor DSP32. The features described are pitch, formants, segmentation and labelling. The paper indicates the potential use of these features in connection with future experiments leading to development of new hearing aids.

Algorithms↗

Clinicopathologic features in rectal cancer treated by local excision and postoperative radiation therapy.

We report the impact of selected clinicopathologic features on local failure and disease-free survival in 22 patients with localized, mobile, primary resectable rectal cancer treated with local excision and postoperative radiation therapy. Full thickness local excisions with negative margins were performed in 21 patients. One patient had a transanal snare excision of a T1 polyp. Postoperatively patients received 4500-4950 cGy (medial 4680 cGy) whole pelvis, and in 15 this was followed by a conedown of 360-1000 cGy (median 360 cGy). Two received 5-FU. Tumors were evaluated for size, gross appearance, distance from the ana verge, T stage, blood vessel invasion, lymphatic vessel invasion, and DNA content (ploidy, DNA index, and proliferation index). The median follow-up was 37 months (range 5-73). With increasing T stage there was a corresponding increase in local failure (T1: 0%, T2: 17%, and T3: 33%) and a decrease in disease-free survival (T1: 100%, T2: 67%, and T3: 50%). When accounting for the effect of T stage, tumors which were either BVI-or ulcerative were associated with an increase in local failure, and tumors which were < or = 3 cm, ulcerative, or nonaneuploid were associated with a decrease in disease-free survival. However, none of the differences reached statistical significance. Although other clinicopathologic features may have an impact, T stage remains the most reliable clinicopathologic feature by which to predict local failure and disease-free survival in patients with rectal cancer who undergo local excision and postoperative radiation therapy.

Adenocarcinoma↗