Resting-state electroencephalogram relevance state recognition of Parkinson's disease based on dynamic weighted symbolic mutual information and k-means clustering / 生物医学工程学杂志

At present, the incidence of Parkinson's disease (PD) is gradually increasing. This seriously affects the quality of life of patients, and the burden of diagnosis and treatment is increasing. However, the disease is difficult to intervene in early stage as early monitoring means are limited. Aiming to find an effective biomarker of PD, this work extracted correlation between each pair of electroencephalogram (EEG) channels for each frequency band using weighted symbolic mutual information and k-means clustering. The results showed that State1 of Beta frequency band ( P = 0.034) and State5 of Gamma frequency band ( P = 0.010) could be used to differentiate health controls and off-medication Parkinson's disease patients. These findings indicated that there were significant differences in the resting channel-wise correlation states between PD patients and healthy subjects. However, no significant differences were found between PD-on and PD-off patients, and between PD-on patients and healthy controls. This may provide a clinical diagnosis reference for Parkinson's disease.

Câncer de colo de útero e componentes de saneamento nos municípios do estado do Mato Grosso do Sul: seria uma questão de má qualidade da água? / Cervical cancer and sanitation components in the municipalities of the state of Mato Grosso do Sul: would it be a question of poor water quality?

Resumo Este artigo investiga relações entre a incidência de câncer de colo de útero (ICC) e os componentes e indicadores de qualidade da água nos municípios do Mato Grosso do Sul, entre 2014 e 2017, por correlação estatística (Determinante de Pearson) e espacial (agrupamentos por k-médias). Houve maior resposta estatística de ICC em relação à tarifa média dos serviços de abastecimento praticado (-36,28%) e de água (-34,15%); à quantidade de suas interrupções sistemáticas (28,3%) e paralizações (22,28%); ao consumo médio per capita de água (20,74%) e à quantidade de serviços executados (-17,98%), todas as respostas sob p-valor ≤ 0,001. Em Costa Rica, cidade sob maior ICC média, os agrupamentos espaciais identificaram maior efeito daquelas interrupções (z-valor = 8,741) e das paralizações (z = 7,6097); enquanto em Rochedo, também sob alta ICC, houve maior efeito à incidência de análises com resultados fora do padrão para coliformes totais (z = 8,6803) e turbidez (z = 5,7427), sob correlação estatística de 12,05% (p-valor = 0,032) e 15,18% (p-valor = 0,007), respectivamente. Dados do SISAGUA revelaram a presença de coliformes e de altos níveis de turbidez, por exemplo, em Antônio João e Tacuru, cidades sob altas ICC médias. Recomenda-se maiores investigações sobre as relações aqui apresentadas entre ICC e água.

Abstratct This article investigates relationships between the incidence of cervical cancer (CCI) and the water components and quality indicators, in the municipalities of Mato Grosso do Sul, between 2014 and 2017, by statistical (Pearson's Determinant) and spatial (k-means Clustering) correlation. There was a greater statistical response of CCI in relation to the average tariff of the practiced supply (−36.28%) and water (−34.15%) services; the number of their systematic interruptions (28.3%) and outages (22.28%); the average per capita consumption of water (20.74%); and the number of services performed (−17.98%), all answers under p-value ≤ 0.001. In Costa Rica, city with the highest average CCI, the spatial clustering identified a greater effect of those interruptions (z-value = 8.741) and outages (z = 7.6097); whereas, in Rochedo, also under high CCI, the analyses showed greater effect with non-standard results for total coliforms (z = 8.6803) and turbidity (z = 5.7427), under a statistical correlation of 12.05% (p-value = 0.032) and 15.18% (p-value = 0.007), respectively. Data from SISAGUA revealed the presence of coliforms and high levels of turbidity, for example, in Antônio João and Tacuru, cities with high average ICC. We recommend further investigation into the relationships presented here between CCI and water.

A new hybrid gkm segmentation for detecting pathological microscopic glomerulosclerosis image of diabetic nephropathy / 国际药学研究杂志

Image segmentation has become a very essential process in medical image processing. It is involved in the application fields of digitized diagnosis for investigating tumors or abnormal tissues in pathology. We have varieties of techniques in image segmentation. One of the unique segmentation technique is k-means; unfortunately it has limitations in segmenting pathological images with scattered populace (group) of tumor nodes like Glomerulosclerosis image of Diabetic Nephropathy, which is a main challenge faced by pathologists for recognizing these regions of abnormalities. Hence we have proposed a new hybrid technique called Genetic K-Means (GKM) Algorithm, which resolves limitations of k-means technique. We have introduced a new hybrid module by superimposing of k-means with evolutionary Genetic Algorithm (GA) for these scattered or distributed nodes of populace (glomerulos) by providing multiple region segmentation with respect to each scattered node. Numerous methods were connected to problems related to the analysis of k-means clustering. Utilization of GKM algorithm is considered in various streams concerning cluster analysis and segmentation performance. In this work, we have explored the GKM utilization to decide on initialization of most excellent clusters and the streamlining of essential parameters like best fitness and mean fitness. Our investigation outcomes the extraordinary capability of GKM enhancement in identifying clusters for segmentation. The segmentation results were encouraging. We have achieved a good success rate of 99% in detecting pathological microscopic glomerulosclerosis image of diabetic nephropathy.

Automatic clustering method of flow cytometry data based on -distributed stochastic neighbor embedding / 生物医学工程学杂志

The traditional method of multi-parameter flow data clustering in flow cytometry is to mainly use professional software to manually set the door and circle out the target cells for analysis. The analysis process is complex and professional. Based on this, a clustering algorithm, which is based on -distributed stochastic neighbor embedding ( -SNE) algorithm for multi-parameter stream data, is proposed in the paper. In this algorithm, the Euclidean distance of sample data in high dimensional space is transformed into conditional probability to represent similarity, and the data is reduced to low dimensional space. In this paper, the stained human peripheral blood cells were treated by flow cytometry, and the processed data were derived as experimental sample data. The -SNE algorithm is compared with the kernel principal component analysis (KPCA) dimensionality reduction algorithm, and the main component data obtained by the dimensionality reduction are classified using -means algorithm. The results show that the -SNE algorithm has a good clustering effect on the cell population with asymmetric and trailing distribution, and the clustering accuracy can reach 92.55%, which may be helpful for automatic analysis of multi-color multi-parameter flow data.

Performance quantification of clustering algorithms for false positive removal in fMRI by ROC curves

Abstract Introduction Functional magnetic resonance imaging (fMRI) is a non-invasive technique that allows the detection of specific cerebral functions in humans based on hemodynamic changes. The contrast changes are about 5%, making visual inspection impossible. Thus, statistic strategies are applied to infer which brain region is engaged in a task. However, the traditional methods like general linear model and cross-correlation utilize voxel-wise calculation, introducing a lot of false-positive data. So, in this work we tested post-processing cluster algorithms to diminish the false-positives. Methods In this study, three clustering algorithms (the hierarchical cluster, k-means and self-organizing maps) were tested and compared for false-positive removal in the post-processing of cross-correlation analyses. Results Our results showed that the hierarchical cluster presented the best performance to remove the false positives in fMRI, being 2.3 times more accurate than k-means, and 1.9 times more accurate than self-organizing maps. Conclusion The hierarchical cluster presented the best performance in false-positive removal because it uses the inconsistency coefficient threshold, while k-means and self-organizing maps utilize a priori cluster number (centroids and neurons number); thus, the hierarchical cluster avoids clustering scattered voxels, as the inconsistency coefficient threshold allows only the voxels to be clustered that are at a minimum distance to some cluster.

Exploration on the Medication Rules of Traditional Chinese Medicine for Hypertension Treatment through Parallel K-means Optimization Algorithm / 医学信息学杂志

The paper introduces relevant concepts of parallelization,including Mapreduce programming frame and distributed document system of Hadoop,optimizes the traditional K-means algorithm,establishes parallel K-means clustering model,explores the medication rules of Traditional Chinese Medicine (TCM) for hypertension treatment,digs into 8 groups of prescriptions for differentiation treatment of hypertension through TCM,verifies authenticity of the conclusion by combining the theories of TCM.

Rapid automated analysis method of flow cytometry data / 军事医学

Objective A major component of flow cytometry data analysis involves gating , which is the process of identifying homogeneous groups of cells .As manual gating is error-prone, non-reproducible, nonstandardized, and time-consuming , we propose a time-efficient and accurate approach to automated analysis of flow cytometry data .Methods Unlike manual analysis that successively gates the data projected onto a two-dimensional filed, this approach, using the K-means clustering results , directly analyzed multidimensional flow cytometry data via a similar subpopulations-merged algorithm.In order to apply the K-means to analysis of flow cytometric data , kernel density estimation for selecting the initial number of clustering and k-d tree for optimizing efficiency were proposed .After K-means clustering , results closest to the true populations could be achieved via a two-segment line regression algorithm .Results The misclassification rate (MR) was 0.0736 and time was 2 s in Experiment One, but was 0.0805 and 1 s respectively in Experiment Two. Conclusion The approach we proposed is capable of a rapid and direct analysis of the multidimensional flow cytometry data with a lower misclassification rate compared to both nonprobabilistic and probabilistic clustering methods .

Thalamus segmentation based on resting-state functional magnetic resonance imaging / 国际生物医学工程杂志

Objective To obtain an accurate and effective method for thalamus segmentation based on resting-state functional magnetic resonance imaging (fMRI).Methods Based on the fact that resting-state fMRI technique examined spatial synchronization of spontaneous fluctuations in blood oxygen level-dependent (BOLD) signals indirectly reflect the neuronal and synaptic activity,the in-thalamus BOLD signal correlations were calculated,and then the k-means clustering algorithm was applied to obtain functional connectivity-based thalamus segmentation.Results The thalamus was divided into seven regions.Voxels within the same region were highly correlated with each other.The segmentation result was similar to that divided by functional connectivity between thalamus and the cerebral cortex.Conclusions Resting-state fMRI could provide not only the functional connectivity network between cortical and subcortical brain regions,but also the functional characteristics of thalamus.Segmentation algorithm using only internal information of thalamus shows lower computational complexity and higher processing speed than that based on the functional connectivity between thalamus and the cerebral cortex.

Componente web para el análisis de información clínica usando la técnica de Minería de Datos por agrupamiento / Web component for the analysis of clinical information using the technique of clustering data mining

La digitalización de los diferentes procesos y la automatización de los servicios generan grandes volúmenes de información. La Minería de Datos (MD) es una técnica de Inteligencia Artificial que permite encontrar la información no trivial que reside en los datos almacenados. La presente investigación pretende desarrollar una vista de análisis para el Sistema Integral para la Atención Primaria de Salud (SIAPS), usando la técnica de agrupamiento enmarcada en el algoritmo Simple K-Means, con el objetivo de realizar un análisis de la información clínica de los pacientes; para ello se plantea la extracción del conocimiento del almacén de datos alimentado del repositorio de historias clínicas electrónicas. La investigación se sustenta en la herramienta de libre distribución WEKA, esta funciona de forma aislada al SIAPS; la interfaz, así como las vistas, modelos e informes generados por WEKA en ocasiones resultan de difícil comprensión por los profesionales de la salud, los que no necesariamente tienen que poseer conocimientos avanzados de las nuevas tecnologías de la información. Para el desarrollo de la solución se empleó el lenguaje de programación Java 1.6, como servidor de aplicación JBoss 4.2 y Eclipse 3.4 como plataforma de desarrollo, como Sistema Gestor de Bases de Datos PostgreSQL 8.4 y SEAM como framework de integración. Durante todo el proceso se hizo uso de la plataforma Java Enterprise Edition 5.0. Como resultado se espera obtener una vista de análisis que facilite la comprensión de los modelos generados, apoyando de esta forma el proceso de toma de decisiones clínicas(AU)

The digitization of the different processes and automation services generate large volumes of information. Data mining (DM) is an artificial intelligence technique that allows finding non-trivial information residing in stored data. This research aims to develop a view of analysis for the Integral System for Primary Health Care (SIAPS), using grouping technique framed on Simple K-Means algorithm, with the goal of completing an analysis of the patients' clinical information, for it raises the extraction of knowledge from data warehouse powered by the repository of electronic medical records. The research is based on the free distribution tool WEKA, it works in isolation of SIAPS, the interface, as well as the views, models and reports generated by WEKA are sometimes difficult to understand by health professionals, who do not necessarily have to possess advanced knowledge of new information technologies. For the development of the solution was used Java 1.6 as a programming language, JBoss 4.2 as the application Server and Eclipse 3.4 as a development platform. PostgreSQL 8.4 was used as Database Management System and the integration framework SEAM. Java Enterprise Edition 5.0 platform was used during the whole process. An analysis view to facilitate the understanding of the generated models is expected as a result, to support the process of making clinical decisions(AU)

Selección de Electrodos Basada en k-means para la Clasificación de Actividad Motora en EEG / Eletrode Selection Based on k-means for Motor Activity Classification in EEG

Se presenta un algoritmo para la selección del grupo de electrodos relacionados con la imaginación de movimiento. El algoritmo utiliza la técnica de agrupamiento llamada k-means para formar grupos de sensores y selecciona el grupo que corresponde a la actividad correlacionada más alta. Para evaluar la selección de electrodos, se calcula el indice de clasificación aplicando la descomposición proyectiva llamada patrones espaciales comunes y un discriminante lineal en una prueba de una sola época para identificar la imaginación del movimiento de mano izquierda vs pie derecho. Esta propuesta reduce significativamente el número de electrodos de 118 a 35, además de mejorar el índice de clasificación.

We present an algorithm for electrodes selection associated with motor imagery activity. The algorithm uses a clustering technique called k-means to form groups of sensors and selects the group corresponding to the highest correlation activity. Then, we evaluate the selected electrodes computing the classification index using the projective decomposition called common spatial patterns and a linear discriminant method in a left hand vs right foot motor imagery classification task. This approach significantly reduces the number of electrodes from 118 to 35 while improving the classification accuracy index.

A study of the geographical clustering of districts in Uttar Pradesh using nutritional anthropometric data of preschool children.

Background & objectives: Worldwide variations in human growth and its genetic and environmental factors have been described. In this study, an attempt was made to assess the morphological differences and similarities among under 5 year children of rural areas of Uttar Pradesh State in India, and to determine differences or similarities of body size among children living in diverse regions. Methods: For this purpose, a cross-sectional district nutrition profile study conducted during 2002-2003 was used. The data on 10,096 children drawn from 1080 villages in 54 districts were part of the district level Diet and Nutrition Assessment survey. The mean values for height and weight for 54 districts were taken as the input data for subsequent analysis. The data were first normalized by means of principal component analysis (PCA) and then K-means clustering was performed. Results: The PCA and cluster analysis yielded four distinguishable clusters or patterns in the anthropometric data of children. These clusters were ordered according to the average body size (weight and height) of children. The mean stature and body weight of these children in cluster I were 3.2 cm and 1.4 kg higher than those of cluster IV indicating differences between clusters. Also, the variations between clusters in their social, demographic, health and nutrition parameters were compared. Interpretation & conclusions: The use of PCA and cluster analysis methods and their merits in studying the Uttar Pradesh preschool children growth variations are discussed. These results helped in identifying the districts with higher prevalence of undernutrition and the contributing factors.

Definição de zonas de manejo a partir de mapas de condutividade elétrica e matéria orgânica / Definition of management zones from maps of electrical conductivity and organic matter

Este trabalho teve por objetivo definir zonas de manejo com base na variabilidade espacial da condutividade elétrica aparente do solo e da matéria orgânica, em áreas de plantio direto de milho e soja. Para caracterizar a variabilidade espacial foram utilizados métodos geoestatísticos. Comprovada a dependência espacial foram elaborados os mapas temáticos, por meio da krigagem. Para delimitação das zonas de manejo a partir dos mapas de variabilidade interpolados foi utilizado o algoritmo fuzzi K-means e para definição do número ótimo de classes foi determinado o índice de perfomance fuzzi e entropia da partição modificada. As variáveis utilizadas para a definição das zonas de manejo foram a altitude, a condutividade elétrica a 20 cm e 40 cm de profundidade e a matéria orgânica. A partir destas variáveis foram gerados sete mapas de zonas de manejo, e posteriormente pelo teste de Kappa foi analisada a concordância entre os mapas gerados pelas zonas de manejo e os mapas das propriedades físico-químicas do solo. Como resultado verificou-se o valor ótimo de número de classes igual a dois. Os melhores resultados na classificação das zonas de manejo, para os atributos referentes a textura do solo são observados a partir de mapas de matéria orgânica ou de condutividade elétrica e, para os atributos químicos, a partir de mapas de matéria orgânica ou de altitude e matéria orgânica. As zonas de manejo definidas a partir da condutividade elétrica a 20 cm permitiram detectar diferenças significativas entre as médias de produtividade de soja.

This study aimed to define management zones based on spatial variability of soil apparent electrical conductivity and organic matter in areas of tillage. To characterize the spatial geostatistical methods were used. Proven spatial dependence was prepared thematic maps through kriging. For delineation of management zones based on maps of variability was interpolated using the Fuzzy K-means algorithm and to define the optimal number of classes was determined Fuzzy performance index and entropy of the partition changed. The variables used for defining management zones were altitude, the electrical conductivity at 20 cm and 40 cm depth and organic matter. From these seven variables were generated maps of management zones, and later by the Kappa test was analyzed the correlation between the maps generated by the management zones and maps of the physical and chemical properties of soil. As a result there was an optimum number of classes equal to two, with the attributes related to soil texture management zone maps from organic matter or electrical conductivity and the chemical zone management from maps of organic matter or organic matter and altitude showed better results in their classification. The management zones defined from the electrical conductivity at 20 cm allowed us to detect significant differences between the average yield of soybean.

Segmentación deiimágenes de resonancia magnétic en contraste de fase para el eestudio de la dinámica del líquido cefalorraquídeo perimedular / Segmentation of Phase Contrast Magnetic Resonance Imaging to Study the Dynamic of Perimedullary Cerebrospinal Fluid

La imagen de resonancia magnética en contraste de fase permite estudiar la dinámica del líquido cefalorraquídeo (LCR) perimedular de manera cuantitativa. Sin embargo la anatomía propia del espacio subaracnoideo dificulta la segmentación del LCR debido a la presencia de estructuras vasculares y nervios raquídeos. El objetivo de este trabajo es describir un método de segmentación semiautomático para el estudio de la dinámica del LCR perimedular. El proceso se inicializa con un punto semilla dentro de la región a analizar. El algoritmo crea un mapa de correlación, calcula un valor de umbral y clasifica píxeles de LCR combinando diversas características temporales del comportamiento del flujo como atributos de entrada a un algoritmo k-medias. Un observador llevó a cabo diez veces la segmentación en cinco sujetos sanos y se calculó el volumen por ciclo y el área en el espacio perimedular C2C3. Las variaciones de las medidas fueron evaluadas como una estimación de la reproducibilidad del método. Para esto se calculó el coeficiente de variación. La variabilidad de las medidas fue menor del 5%. El método facilita la cuantificación del LCR perimedular. En 16 sujetos sanos se cuantificó el volumen por ciclo de LCR y el área en el espacio C2C3 y cisterna prepontina.

Phase contrast magnetic resonance imaging allows studying quantitatively the perimedullary cerebrospinal fluid (CSF) dynamics. However, the anatomy of the subarachnoid space difficults the segmentation of CSF due to the presence of vascular structures and spinal nerves. The aim of this paper is to describe a semiautomatic segmentation method for the study of the perimedullary CSF dynamics. The process is started with a seed point within the region to analyze. The algorithm creates a correlation map, calculates a threshold value and classifies pixels of CSF combining different temporal characteristics of flow behavior as input attributes to a k-means algorithm. One observer carried out ten times the segmentation of the cervical images in 5 healthy subjects; stroke volume and area were calculated. The variability of the obtained measurements was evaluated as an estimation of the reproducibility of the method. For this the coefficient of variation was calculated. The variability of the measurements was less than 5%. The method facilitates the quantification of perimedullary CSF. Stroke volume and the area at C2C3 space and prepontine cistern were measured in 16 healthy subjects.

Harris Operator and K-means Clustering-based Facial Features Localization on Infrared Images / 航天医学与医学工程

Objective To develop an image analyzing procedure for automatic localization of facial features on infrared images.Methods An unsupervised local and global features extraction method was adopted for the localization of facial features of frontal view face image.First,a threshold was used to segment the image into foreground and background,and the nose was localized by considering the symmetry of the face.Second,Harris operator was adopted to detect interest points in a rectangular area covering all the facial features,and then local maximum of the interest points were detected.And finally,K-means clustering method was used to cluster the points and obtain the facial features localization.Results The experimental result of 100 images demonstrated that the procedure could automatically localize eyes,nose,mouth,and distinguish the feature areas.Conclusion The proposed infrared image analyzing procedure based on Harris operator and K-means clustering can be used to locate facial features on infrared image more rapidly and reliablely.

The research of calculation of the injury area in the medical jurisprudence image / 重庆医科大学学报

Objective:The legal medical experts usuallycalculate the area ofinjured organs in bodyor other injured items bycountingthe grids in coordination paper.The experiences of the legal medical experts determine the result of this method.The result is subjective and has serious errors.So it is not conductive to the court and case detection.Methods:This paper proposes a method to calculate the injured area by hybrid of K-means algorithm and region growing algorithm.Results:The experiment shows this method is effective to calculate the area of irregular region.Conclusion:The result of this method is more accurate than the method of counting the grids in coordination paper.