High Classification Scores Research Articles

LangAware: A robust ML‐based platform for the quick and accurate assessment of neurodegenerative diseases

AbstractBackgroundAlthough Alzheimer's disease (AD) is associated with changes in spoken language, these have seldom been subjected to systematic analysis on a large scale. We evaluated the effectiveness of LangAware to detect the language indicators that are coupled with early AD, thus assisting with diagnosis. We evaluated LangAware using recordings of speech samples obtained from AD patients and matched healthy controls (NC) derived from various elicitation tasks in two languages, English and Greek*.MethodEnglish and Greek datasets were analyzed employing feature selection techniques to choose the most prominent multi‐level linguistic analysis features differentiating the AD from the NC group in both languages. The platform’s diagnostic performance was evaluated on its ability to classify "unseen" audio recordings employing these salient features.ResultEvaluation results indicated that LangAware achieved equally high classification scores for both English and Greek. Most significantly, these scores were achieved by employing a custom set of LangAware‐developed cross‐linguistic markers.ConclusionThe current evaluation verified the robustness of the platform’s predictive models using audio datasets in two languages. Based on the findings, we conclude that LangAware could provide a time and cost‐effective platform for cognitive screening across languages and across tasks pertaining to neurodegenerative diseases in a range of clinical settings. Such findings also advocate in favor of the robustness of LangAware platform in pursuing cognitive assessment on spontaneous speech across languages. *Peter Garrard and Antonis A. Mougias contributed to this work by providing data for the evaluation of the LangAware platform.

Graph-based non-maximal suppression for detecting products on the rack

Abstract Identification of stacked retail products from the images of racks of supermarket is a challenging computer vision problem. Regions with convolutional neural network features (R-CNN) generate (mostly overlapped) region proposals around the products on the rack. Subsequently, the region proposals are classified using a convolutional neural network. In the end, R-CNN implements a greedy non-maximal suppression (greedy-NMS) for disambiguating the overlapping proposals. Greedy-NMS discards the proposals (with lower classification scores) that are overlapped with the proposal with higher classification score. This greedy approach often eliminates the (geometrically) better fitted region proposals with (marginally) lower classification scores. This paper introduces a novel graph-based non-maximal suppression (G-NMS) that removes this critical bottleneck of greedy-NMS by looking not only at the classification scores but also at the product classes of the overlapping region proposals. G-NMS first determines the potential confidence scores (pc-scores) of the region proposals by defining the groups of overlapping regions. Subsequently, a directed acyclic graph (DAG) is strategically constructed with the proposals utilizing their pc-scores and overlapping groups. Eventually the maximum weighted path of the DAG provides the products that are present in the rack. The results of our extensive experiments confirm that the proposed scheme is better up to around 7% on one large In-house and three benchmark datasets of retail products. Additionally, the efficacy of our proposed GNMS is also analyzed on four benchmark datasets for detecting generic objects.

Alternative treatment with Johrei: A controlled randomized study evaluating seed physiological potential

BackgroundJohrei, a noncontact healing therapy that purportedly channels “universal energy”, is an easily taught technique with numerous positive reports of healing in human trials. However, studies on humans have been limited by small sample sizes, the placebo effect and baseline differences between groups. Overcoming some of these limitations, this study evaluated the effect of Johrei on the physiological potential of lettuce seeds. MethodsLettuce seeds were subjected to thermal stress and then treated with Johrei before and after two years of storage. The seedling emergence percentage, emergence speed index, classification score, percentage of viable seedlings and root and shoot dry mass of the treated seeds were determined. ResultsStress-treated seeds treated with Johrei had a greater seedling emergence percentage than those not treated with Johrei (p<0.0001). Johrei also resulted in a greater emergence speed index (p<0.0001), a higher percentage of viable seedlings with higher classification score (p<0.0001) and greater root and shoot dry mass regardless of the storage duration (p<0.0001). ConclusionsJohrei effectively preserved lettuce seed health, and the efficacy of Johrei can be assessed by evaluating its effects on the physiological potential of stress-treated seeds. Using live plants allows the influence of Johrei to be evaluated without interference from the placebo effect. Future studies should evaluate the influence of Johrei in primary aging processes in plants, such as cellular respiration by mitochondrial analysis, to gain insight into Johrei's mechanisms of action and enable inferences regarding similar cellular-level processes that occur in humans.

A Weighted Voting Classifiers Ensemble for the Brain Tumors Classification in MR Images

The appropriate treatment of cancers critically depends on the precise identification of the tumor types. The Magnetic Resonance Imaging (MRI) provides a non-invasive approach for detecting the tumors and classifying their types. Computer-aided diagnosis (CAD) systems using image analysis and machine learning algorithms assist radiologists in making error-free decisions. This paper presents an integrated approach for the segmentation and classification of three types of brain tumors (i.e. meningioma, glioma, and pituitary tumor) in real MRI images. First, the images quality is enhanced by removing the noises using wiener and median filters. Then, the support vector machine (SVM) classifier is performed to segment the MR image. Following this, 42 features are extracted, among which eleven most practical features are chosen by the Differential Evolution (DE). Using the selected features, tumor slices are categorized using the k-nearest neighbors (KNN), the weighted kernel width SVM (WSVM) and the histogram intersection kernel SVM (HIK-SVM) classifiers. Afterwards, these classifiers are combined using a multi-objective differential evolution (MODE)-based ensemble technique to attain the classification accuracy of 92.46%, evaluated using the five-fold cross validation method. The results of the proposed approach are also compared with the experienced radiologist ground truth which shows that the proposed MODE-based ensemble technique is able to achieve high classification scores, measured also by sensitivity and specificity indices.

Proactive Critical Energy Infrastructure Protection via Deep Feature Learning

Autonomous fault detection plays a major role in the Critical Energy Infrastructure (CEI) domain, since sensor faults cause irreparable damage and lead to incorrect results on the condition monitoring of Cyber-Physical (CP) systems. This paper focuses on the challenging application of wind turbine (WT) monitoring. Specifically, we propose the two challenging architectures based on learning deep features, namely—Long Short Term Memory-Stacked Autoencoders (LSTM-SAE), and Convolutional Neural Network (CNN-SAE), for semi-supervised fault detection in wind CPs. The internal learnt features will facilitate the classification task by assigning each upcoming measurement into its corresponding faulty/normal operation status. To illustrate the quality of our schemes, their performance is evaluated against real-world’s wind turbine data. From the experimental section we are able to validate that both LSTM-SAE and CNN-SAE schemes provide high classification scores, indicating the high detection rate of the fault level of the wind turbines. Additionally, slight modification on our architectures are able to be applied on different fault/anomaly detection categories on variant Cyber-Physical systems.

Optimization for Greedy Non-maximum Suppression Based on Multi-task Convolutional Neural Network

Non-maximum suppression (NMS) is an essential part of the face detection pipeline based on a convolutional neural network (CNN). The typical approach for NMS used by face detection is a greedy, locally optimal strategy, which is to localize objects from a set of candidate locations. However, NMS still has some shortcomings, such as sometimes the detection box has no relationship with a high classification score, which leads to misjudging face localization during NMS. In this paper, we observed that the NMS implemented on the multi-task convolutional neural network (MTCNN) is a cascaded network and the enhanced NMS based on MTCNN to achieve high performance during face detection and alignment. We employ WIDER FACE as the test dataset to evaluate our proposal. The precision and recall curves are drawn when using three subsets at different thresholds. The result shows that the proposed approach can perform better performance than traditional NMS.

PENERAPAN METODE ROLE PLAYING UNTUK MENINGKATKAN HASIL BELAJAR PKN SISWA KELAS V SDN 004 TERATAK BULUH KECAMATAN SIAK HULU KABUPATEN KAMPAR

The role-playing method is a method of role-playing by providing real experience in learning. The research method used was classroom action research, this study was conducted in two cycles and each cycle was conducted three times. Data collection techniques in this study used teacher observation sheets, student activity sheets, and student learning outcomes. Learning tools consist of the syllabus, and plan for implementing learning. Learning outcomes in the first cycle, the number of students who get a high classification score amounted to 9 people with a percentage of 36%, and a moderate classification of 16 people with a percentage of 64%. In the second cycle, the results of high classification learning were 20 people with a percentage of 80%, and in the middle classification, there were 5 people with a percentage of 20%. So it was concluded that using the role-playing method can improve learning outcomes of PKn.

Influence of nerve cuff channel count and implantation site on the separability of afferent ENG

Objective. Recording of neural signals from intact peripheral nerves in patients with spinal cord injury or stroke survivors offers the possibility for the development of closed-loop sensorimotor prostheses. Nerve cuffs have been found to provide stable recordings from peripheral nerves for prolonged periods of time. However, questions remain over the design and positioning of nerve cuffs such that the separability of neural data recorded from the peripheral nerves is improved. Approach. Afferent electroneurographic (ENG) signals were recorded with nerve cuffs placed on the sciatic nerve of rats in response to various mechanical stimuli to the hindpaw. The mean absolute value of the signal was extracted and input to a classifier. The performance of the classifier was evaluated under two conditions: (1) when information from either a 3- or 16-channel cuff was used; (2) when information was available from a cuff placed either distally or proximally along the nerve. Main results. We show that both 3- and 16-channel cuffs were able to separate afferent ENG signals with an accuracy greater than chance. The highest classification scores were achieved when the classifier was fed with information obtained from a 16-channel cuff placed distally. While the 16-channel cuff always outperformed the 3-channel cuff, the difference in performance was increased when the 16-channel cuff was placed distally rather than proximally on the nerve. Significance. The results indicate that increasing the complexity of a nerve cuff may only be advantageous if the nerve cuff is to be implanted distally, where the nerve has begun to divide into individual fascicles.

Decoding spoken phonemes from sensorimotor cortex with high-density ECoG grids

For people who cannot communicate due to severe paralysis or involuntary movements, technology that decodes intended speech from the brain may offer an alternative means of communication. If decoding proves to be feasible, intracranial Brain-Computer Interface systems can be developed which are designed to translate decoded speech into computer generated speech or to instructions for controlling assistive devices. Recent advances suggest that such decoding may be feasible from sensorimotor cortex, but it is not clear how this challenge can be approached best. One approach is to identify and discriminate elements of spoken language, such as phonemes. We investigated feasibility of decoding four spoken phonemes from the sensorimotor face area, using electrocorticographic signals obtained with high-density electrode grids. Several decoding algorithms including spatiotemporal matched filters, spatial matched filters and support vector machines were compared. Phonemes could be classified correctly at a level of over 75% with spatiotemporal matched filters. Support Vector machine analysis reached a similar level, but spatial matched filters yielded significantly lower scores. The most informative electrodes were clustered along the central sulcus. Highest scores were achieved from time windows centered around voice onset time, but a 500 ms window before onset time could also be classified significantly. The results suggest that phoneme production involves a sequence of robust and reproducible activity patterns on the cortical surface. Importantly, decoding requires inclusion of temporal information to capture the rapid shifts of robust patterns associated with articulator muscle group contraction during production of a phoneme. The high classification scores are likely to be enabled by the use of high density grids, and by the use of discrete phonemes. Implications for use in Brain-Computer Interfaces are discussed.

Feature selection using a one dimensional naïve Bayes' classifier increases the accuracy of support vector machine classification of CDR3 repertoires.

MotivationSomatic DNA recombination, the hallmark of vertebrate adaptive immunity, has the potential to generate a vast diversity of antigen receptor sequences. How this diversity captures antigen specificity remains incompletely understood. In this study we use high throughput sequencing to compare the global changes in T cell receptor β chain complementarity determining region 3 (CDR3β) sequences following immunization with ovalbumin administered with complete Freund’s adjuvant (CFA) or CFA alone.ResultsThe CDR3β sequences were deconstructed into short stretches of overlapping contiguous amino acids. The motifs were ranked according to a one-dimensional Bayesian classifier score comparing their frequency in the repertoires of the two immunization classes. The top ranking motifs were selected and used to create feature vectors which were used to train a support vector machine. The support vector machine achieved high classification scores in a leave-one-out validation test reaching >90% in some cases.SummaryThe study describes a novel two-stage classification strategy combining a one-dimensional Bayesian classifier with a support vector machine. Using this approach we demonstrate that the frequency of a small number of linear motifs three amino acids in length can accurately identify a CD4 T cell response to ovalbumin against a background response to the complex mixture of antigens which characterize Complete Freund’s Adjuvant.Availability and implementationThe sequence data is available at www.ncbi.nlm.nih.gov/sra/?term¼SRP075893. The Decombinator package is available at github.com/innate2adaptive/Decombinator. The R package e1071 is available at the CRAN repository https://cran.r-project.org/web/packages/e1071/index.html.Supplementary information Supplementary data are available at Bioinformatics online.

Solving petrological problems through machine learning: the study case of tectonic discrimination using geochemical and isotopic data

Machine learning methods are evaluated to study the intriguing and debated topic of discrimination among different tectonic environments using geochemical and isotopic data. Volcanic rocks characterized by a whole geochemical signature of major elements (SiO2, TiO2, Al2O3, Fe2O3T, CaO, MgO, Na2O, K2O), selected trace elements (Sr, Ba, Rb, Zr, Nb, La, Ce, Nd, Hf, Sm, Gd, Y, Yb, Lu, Ta, Th) and isotopes (206Pb/204Pb, 207Pb/204Pb, 208Pb/204Pb, 87Sr/86Sr and 143Nd/144Nd) have been extracted from open-access and comprehensive petrological databases (i.e. PetDB and GEOROC). The obtained dataset has been analyzed using support vector machines, a set of supervised machine learning methods, which are considered particularly powerful in classification problems. Results from the application of the machine learning methods show that the combined use of major, trace elements and isotopes allow associating the geochemical composition of rocks to the relative tectonic setting with high classification scores (93%, on average). The lowest scores are recorded from volcanic rocks deriving from back-arc basins (65%). All the other tectonic settings display higher classification scores, with oceanic islands reaching values up to 99%. Results of this study could have a significant impact in other petrological studies potentially opening new perspectives for petrologists and geochemists. Other examples of applications include the development of more robust geo-thermometers and geo-barometers and the recognition of volcanic sources for tephra layers in tephro-chronological studies.

Discriminant analysis for the prediction of sand mass distribution in a holding pond using deposition thickness model of a single grain-sized particle

Discriminant analysis (DA) was applied to derive the classification function to spatially determine the sand mass distribution in an urban stormwater holding pond, Malaysia, using simulated deposition thickness results of a single grain-sized particle (d50 = 0.375 mm). The upstream boundary lines at the inflow gates of the hydrodynamic model were fixed to three inflow parameter values, 16, 40 and 80 m3 s−1, representing the inflow variations of diverted floodwater into the holding pond during major storm events. The annual mean suspended sediment concentration of approximately 373 mg l−1 at the confluence of the Klang and Ampang Rivers was fixed as input parameter value for the sand deposition model. The simulated deposition thickness of the single grain-sized particle displays a significantly strong correlation with the measured sand mass content of the surface sediment samples at corresponding locations. The surface sediment sampling locations were hierarchically clustered into two groups based on the sand mass composition of the sediment samples. Group one comprises nine sampling locations representing areas of relatively higher sand mass composition ranging from 16.9 to 45.5 %. Group two comprises twenty sampling locations representing areas of relatively lower sand mass composition ranging from 3.7 to 12.2 %. DA was conducted to generate linear combination of the independent variables that will discriminate best between the objects in the groups. Two classification functions also known as Fisher’s linear discriminant functions, F 1 = −6.441 − 125.438 × A + 57.519 × B and F 2 = −1.248 − 53.442 × A + 24.424 × B, were produced to predict group memberships of relatively high and low sand mass composition areas, where A and B denote the simulated deposition thickness results of a single grain-sized particle computed at nodal points of the model domain, generated by an inflow magnitude of 16 and 40 m3 s−1, respectively. The classification of areas which comprises of relatively high and low sand mass composition is based on whichever group (F 1 or F 2) is related to the highest classification score. Classification result indicates that the model correctly predicts 66.7 and 100.0 % of the surface sediment sampling locations consisting of relatively high (group one) and low (group two) sand mass percentage, respectively. The predicted results of the sand mass distribution show that 23.6 % of the model domain area (25.20 m2) at the western bank region consists of a relatively higher sand mass content compared to the remaining 76.4 % comprising areas of lower sand mass content mostly situated at the central and south-eastern regions of the holding pond. The predicted sand mass distribution results can be used implicitly to spatially distinguish the most likely contaminated sediment areas and identify accurately sampling locations of polluted sediment for environmental impact assessment study prior conducting any dredging activities for future maintenance.

Geometric morphometrics for the taxonomy of 11 species of Anopheles (Nyssorhynchus) mosquitoes.

The subgenus Anopheles (Nyssorhynchus) (Diptera: Culicidae) includes the primary vectors of Plasmodium spp. in Colombia. Most adult females of this subgenus are difficult to identify in the field using the available keys. With the objective of further investigating the discriminatory power of modern morphometrics, both landmark-based and outline-based approaches were explored using the wing venation geometry of 11 Anopheles (Nyssorhynchus) species. Wing shape was able to separate the closest species of the subgenus. When the 11 species were analysed together, validated classification scores on average 5.3-8.6 times higher than those expected by chance were observed. These scores computed from the total sample of 11 species were not satisfactory for the recognition of Anopheles benarrochi B, Anopheles oswaldoi s.l. and Anopheles strodei. These sibling species were captured in sympatry. To improve the identification power of the morphometric tool, it was necessary to analyse these species separately from the remaining species. The best classification scores were obtained using a combination of 12 landmarks collected not only on the intersections of wing veins, but also on spots. An outline approach also gave excellent reclassification scores. Another pair of sibling species, collected in allopatry, Anopheles nuneztovari and Anopheles rangeli, also showed high classification scores.

Evaluation of muscle force classification using shape analysis of the sEMG probability density function: a simulation study.

In this work, we propose to classify, by simulation, the shape variability (or non-Gaussianity) of the surface electromyogram (sEMG) amplitude probability density function (PDF), according to contraction level, using high-order statistics (HOS) and a recent functional formalism, the core shape modeling (CSM). According to recent studies, based on simulated and/or experimental conditions, the sEMG PDF shape seems to be modified by many factors as: contraction level, fatigue state, muscle anatomy, used instrumentation, and also motor control parameters. For sensitivity evaluation against these several sources (physiological, instrumental, and neural control) of variability, a large-scale simulation (25 muscle anatomies, ten parameter configurations, three electrode arrangements) is performed, by using a recent sEMG-force model and parallel computing, to classify sEMG data from three contraction levels (20, 50, and 80% MVC). A shape clustering algorithm is then launched using five combinations of HOS parameters, the CSM method and compared to amplitude clustering with classical indicators [average rectified value (ARV) and root mean square (RMS)]. From the results screening, it appears that the CSM method obtains, using Laplacian electrode arrangement, the highest classification scores, after ARV and RMS approaches, and followed by one HOS combination. However, when some critical confounding parameters are changed, these scores decrease. These simulation results demonstrate that the shape screening of the sEMG amplitude PDF is a complex task which needs both efficient shape analysis methods and specific signal recording protocol to be properly used for tracking neural drive and muscle activation strategies with varying force contraction in complement to classical amplitude estimators.

Concurrent tumor segmentation and registration with uncertainty-based sparse non-uniform graphs

In this paper, we present a graph-based concurrent brain tumor segmentation and atlas to diseased patient registration framework. Both segmentation and registration problems are modeled using a unified pairwise discrete Markov Random Field model on a sparse grid superimposed to the image domain. Segmentation is addressed based on pattern classification techniques, while registration is performed by maximizing the similarity between volumes and is modular with respect to the matching criterion. The two problems are coupled by relaxing the registration term in the tumor area, corresponding to areas of high classification score and high dissimilarity between volumes. In order to overcome the main shortcomings of discrete approaches regarding appropriate sampling of the solution space as well as important memory requirements, content driven samplings of the discrete displacement set and the sparse grid are considered, based on the local segmentation and registration uncertainties recovered by the min marginal energies. State of the art results on a substantial low-grade glioma database demonstrate the potential of our method, while our proposed approach shows maintained performance and strongly reduced complexity of the model.

Evidence for discrete subpopulations of sea perch (Helicolenus ercoides) across four fjords in Fiordland, New Zealand

In coastal populations of invertebrates and fishes, the distribution of discrete subpopulations is influenced by adult and larval dispersal, as well as by the effects of habitat heterogeneity on site fidelity or connectivity. Here, we examine evidence for spatial structure of sea perch, Helicolenus percoides, populations among four fjords in the Fiordland region of southwestern New Zealand. We examine patterns in adult morphology, length-at-age, δ13C and δ15N of muscle tissue, and trace elemental composition of whole otoliths as proxies for population isolation among the four inner fjord regions. A multivariate analysis of morphometrics reveals significant differences among populations from each of the four sites, suggesting existence of four distinct subpopulations. These patterns are consistent with observed differences in δ13C and δ15N, and length-at-age estimates among the four subpopulations. Differences in whole otolith concentrations of Sr, Ba, Mg and Li, and high classification scores based on the whole otolith elemental fingerprint are also consistent with significant subdivision among areas. Patterns across all four markers are consistent with discrete subpopulation structure of adult sea perch among the four study sites. These data indicate that the newly implemented network of marine protected areas in Fiordland is likely to contain discrete populations of sea perch.

Pattern recognition via projection-based [formula omitted]NN rules

A new procedure for pattern recognition is introduced based on the concepts of random projections and nearest neighbors. It can be considered as an improvement of the classical nearest neighbor classification rules. Besides the concept of neighbors, the notion of district, a larger set into which the data will be projected, is introduced. Then a one-dimensional k NN method is applied to the projected data on randomly selected directions. This method, which is more accurate to handle high-dimensional data, has some robustness properties. The procedure is also universally consistent. Moreover, the method is challenged with the Isolet data set where a very high classification score is obtained.

Significance of classification scores subsequent to feature selection

Feature selection prior to classification is shown to inflate performance. Empirical distributions are used to estimate statistical significance of classification scores. In an example study, eleven high classification scores are obtained but only three are found to be significant at p = 0.05.

Pig carcass quality in relation to carcass lean meat proportion

The objective of the study was to quantify differences in the carcass composition of pig carcasses with different lean meat proportions. Totally 132 carcasses with lean meat proportions ranging from 45 to 65% were analysed. The average lean meat proportion in the analysed set of carcasses was 55.38% (s = 4.319). The carcasses were classified into the groups according to their lean meat proportion (R, 45.0 to 49.9%; U, 50.0 to 54.9%; E, 55.0 to 59.9%; S, more than 60.0%). Carcasses from each group were cut into primal cuts and the main cuts (leg, loin, shoulder and belly with bones) were further separated into different tissues. Carcass value was markedly enhanced by the increasing carcass lean meat proportion. Average leg percentages in the groups R, U, E and S were 17.88% (s = 0.918), 19.32% (s = 0.889), 20.88% (s = 0.817) and 21.88% (s = 0.827), respectively. Average proportions of fat over leg expressed as percentages of leg weight in these groups were 5.82% (s = 0.402), 4.87% (s = 0.556), 4.05% (s = 0.479) and 3.21% (s = 0.321), respectively. Lean meat proportions of leg weight were improved from 84.93% (s = 1.136) in the group R to 88.12% (s = 0.986) in the group S. The most distinct differences were found in lean proportions of belly weight. The observed averages in the group R and S were 43.90% (s = 2.729) and 62.10% (s = 2.219), respectively. Actually, in carcasses with a high classification score the belly can be considered as a cut with the predominant muscle tissue. The results of the study proved the efficiency of carcass evaluation based on the lean meat proportion. Such a method can significantly contribute to the overall improvement of pig production. &amp;nbsp;

The Place of Articulation Distinction in Voiced Oral Stops: Evidence from Burst Spectra and Formant Transitions

Abstract This study concerns the extent to which place of articulation in the voiced obstruents /b d dz g/ can be separated from spectral parameters taken in the burst, formant transitions, and a combination of the two. Classifications were obtained by training on citation-form data produced by male speakers and testing on (i) citation-form data produced by female speakers and (ii) continuous speech data produced by the same male speakers. The results show that there is more information for the place distinction in the burst than in formant transitions; when the parameters are combined into a single model, classification scores are improved for the citation-form data, but not for the continuous speech data. The highest classification scores were in the vicinity of 90% correct for both types of data on the combined parameters. The results are seen as supporting a model of sufficient discriminability rather than one in which phonetic categories are characterised by invariant acoustic cues.