Estrategia para la toma de decisiones en el reconocimiento automático de estados de sedación anestésica / Decision-making strategy for automatic recognition of sedation states / Estratégia de tomada de decisão para reconhecimento automático de estados de sedação

RESUMEN Introducción: La Anestesiología es la especialidad médica dedicada a la atención específica de los pacientes durante procedimientos quirúrgicos y en cuidados intensivos. Esta especialidad basada en los avances científicos y tecnológicos, ha incorporado el uso del monitoreo electroencefalográfico, facilitando el control continuo de estados de sedación anestésica durante las cirugías, con una adecuada concentración de fármacos. Objetivo: Proponer una estrategia de clasificación para el reconocimiento automático de tres estados de sedación anestésica en señales electroencefalográficas. Método: Se utilizaron con consentimiento informado escrito los registros electroencefalográficos de 27 pacientes sometidos a cirugía abdominal, excluyendo aquellos con antecedentes de epilepsia, enfermedades cerebrovasculares y otras afecciones neurológicas. Se aplicaron en total 12 fármacos anestésicos y dos relajantes musculares con montaje de 19 electrodos según el Sistema Internacional 10-20. Se eliminaron artefactos en los registros y se aplicaron técnicas de Inteligencia artificial para realizar el reconocimiento automático de los estados de sedación. Resultados: Se propuso una estrategia basada en el uso de máquinas de soporte vectorial con algoritmo multiclase Uno-Contra-Resto y la métrica Similitud Coseno, para realizar el reconocimiento automático de tres estados de sedación: profundo, moderado y ligero, en señales registradas por el canal frontal F4 y los occipitales O1 y O2. Se realizó una comparación de la propuesta con otros métodos de clasificación. Conclusiones: Se computa una exactitud balanceada del 92,67 % en el reconocimiento de los tres estados de sedación en las señales registradas por el canal electroencefalográfico F4, lo cual favorece el desarrollo de la monitorización anestésica.

ABSTRACT Introduction: Anesthesiology is the medical specialty concerned with the specific care of patients during surgical and intensive care procedures. This specialty, based on scientific and technological advances, has incorporated the use of electroencephalographic monitoring, facilitating the continuous control in the use of anesthesia for patient´s sedation states during surgeries, with an adequate concentration of drugs. Objective: Proposal for a classification strategy for automatic recognition of three sedation states in electroencephalographic signals. Methods: We used, with written informed consent, the electroencephalographic records of 27 patients undergoing abdominal surgery, excluding those with a history of epilepsy, cerebrovascular disease and other neurological conditions. A total of 12 drugs to produce anesthesia and two muscle relaxants with 19 electrodes, mounted according to the International System 10 -20, were applied. Artifacts in the records were eliminated and artificial intelligence techniques were applied to perform automatic recognition of sedation states. Results: A strategy based on the use of support vector machines with a multiclass algorithm One-against-Rest and the Cosine Similarity metric was proposed to perform the automatic recognition of three sedation states: deep, moderate and light, in signals recorded by the frontal channel F4 and the occipital channels O1 and O2. A comparison was carried out between the proposal showed and other classification methods. Conclusions: A balanced accuracy of 92.67% is computed about the recognition of the three states of sedation in the signals recorded by the electroencephalographic channel F4, which helps in a better anesthetic monitoring process.

RESUMO Introdução: A Anestesiologia é a especialidade médica dedicada ao atendimento específico de pacientes durante procedimentos cirúrgicos e em terapia intensiva. Essa especialidade, baseada nos avanços científicos e tecnológicos, incorporou o uso da monitorização eletroencefalográfica, facilitando o controle contínuo dos estados de sedação anestésica durante as cirurgias, com concentração adequada de fármacos. Objetivo: Propor uma estratégia de classificação para o reconhecimento automático de três estados de sedação anestésica em sinais eletroencefalográficos. Método: Foram utilizados registros eletroencefalográficos de 27 pacientes submetidos à cirurgia abdominal com consentimento informado por escrito, excluindo aqueles com histórico de epilepsia, doenças cerebrovasculares e outras condições neurológicas. Um total de 12 drogas anestésicas e dois relaxantes musculares foram aplicados com um conjunto de 19 eletrodos de acordo com o Sistema Internacional 10-20. Artefatos nos prontuários foram removidos e técnicas de inteligência artificial foram aplicadas para realizar o reconhecimento automático dos estados de sedação. Resultados: Foi proposta uma estratégia baseada no uso de máquinas de vetores de suporte com algoritmo One-Against-Rest multiclasse e a métrica Cosine Similarity para realizar o reconhecimento automático de três estados de sedação: profundo, moderado e leve, em sinais registrados pelo canal frontal F4 e os canais occipitais O1 e O2. Foi feita uma comparação da proposta com outros métodos de classificação. Conclusões: Uma acurácia equilibrada de 92,67% é computada no reconhecimento dos três estados de sedação nos sinais registrados pelo canal eletroencefalográfico F4, o que favorece o desenvolvimento da monitorização anestésica.

Machine Learning Models for Early Prediction of Vascular Cognitive Impairment / 中国康复理论与实践

Objective:To explore the predictive performance of machine learning model based on vascular risk factors in early prediction of vascular cognitive impairment. Methods:From April to September, 2020, 70 subjects were enrolled and collected information of the demographics and vascular risk factors. They were assessed with Montreal Cognitive Assessment (MoCA), and then divided into normal group, vascular mild cognitive impairment (VaMCI) group and dementia group. The differences of vascular risk factors among the three groups were detected with one-way ANOVA, and the significant factors were selected to establish predictive models with support vector machine (SVM) and extreme learning machine (ELM). The predictive performance of two models was compared with Receiver Operating Characteristic Curve. Results:There were 32 cases in the normal group, 23 in VaMCI group and 15 in dementia group. Systolic blood pressure, fasting blood glucose, total cholesterol, low density lipoprotein and blood uric acid were significantly different among the three groups (F > 3.318, P < 0.05). The area under curve was the most (0.911) in SVM model predicting for VaMCI (P < 0.01), and the predictive efficacy was better for SVM model. Conclusion:SVM predictive model based on vascular risk factors may be more effective for predicting VaMCI.

Applicability of SA-SVM classifier to auxiliary diagnosis of renal involvement in SLE / 军事医学

Objective To establish a computer-aided diagnosis (CAD) model for the classification and diagnosis of systemic lupus erythematosus (SLE) complicated with renal involvement,and to provide a new method for the timely detection and diagnosis of the disease.Methods Simulated annealing(SA) algorithm was used to optimize the penalty coefficient C and kernel function parameter g of the support vector machines(SVM) algorithm before an SA-SVM classifier model was established and was applied to the intelligent assistant diagnosis of SLE.Results Unlike the single SVM classifier,this method never fell into local optimum,and improved the classification accuracy of a classifier.The classification accuracy for SLE with renal involvement was as high as 98.72％.Conclusion The experimental results show that this classification model is well applicable to the intelligent diagnosis of SLE with renal involvement.

Semi-supervised prediction of gene regulatory networks using machine learning algorithms.

Use of computational methods to predict gene regulatory networks (GRNs) from gene expression data is a challenging task. Many studies have been conducted using unsupervised methods to fulfill the task; however, such methods usually yield low prediction accuracies due to the lack of training data. In this article, we propose semi-supervised methods for GRN prediction by utilizing two machine learning algorithms, namely, support vector machines (SVM) and random forests (RF). The semi-supervised methods make use of unlabelled data for training. We investigated inductive and transductive learning approaches, both of which adopt an iterative procedure to obtain reliable negative training data from the unlabelled data. We then applied our semi-supervised methods to gene expression data of Escherichia coli and Saccharomyces cerevisiae, and evaluated the performance of our methods using the expression data. Our analysis indicated that the transductive learning approach outperformed the inductive learning approach for both organisms. However, there was no conclusive difference identified in the performance of SVM and RF. Experimental results also showed that the proposed semi-supervised methods performed better than existing supervised methods for both organisms.

Application of least squares support vector machines and partial least squares in quantitation on purification of Gardenia jasminoides intermediate using NIR spectroscopy / 中草药

Objective: To establish the quantitative models for analyzing the content of critical quality indicators in the purification process of Gardenia jasminoides intermediate in Reduning Injection using near-infrared (NIR) spectroscopy. Methods: The contents of shanzhiside, geniposidic acid, deacetyl asperulosidic acid methyl ester, genipin-1-β-D-gentiobioside, geniposide, chlorogenic acid, and total acid were determined by the reference method and NIR spectra were acquired. After removing the outliers, selecting the optimal spectral preprocessing method and selecting the best spectral wavelength, partial least squares (PLS) and the least squares support vector machines (LS-SVM) were used to build the models for predicting the contents of the above quality indicators in 18 unknown samples. Results: For shanzhiside, geniposidic acid, deacetyl asperulosidic acid methyl ester, genipin-1-β-D-gentiobioside, geniposide, chlorogenic acid, and total acid, the relative standard errors of prediction (RSEP) was lower than 3% for PLS models and LS-SVM models, indicating both methods could exhibit the satisfactory fitting results and predictive abilities. However, the LS-SVM models of shanzhiside and total acid showed lower predictive errors than PLS models. For geniposidic acid, deacetyl asperulosidic acid methyl ester, genipin-1-β-D-gentiobioside, geniposide, and chlorogenic acid, both models have the closer predictive errors. Conclusion: S-SVM shows better predictive performance than PLS. The established NIR quantitative models can be used for rapidly measuring the content of critical quality indicators in the purification process of G. jasminoides intermediate in Reduning Injection.

Computational Discrimination of Breast Cancer for Korean Women Based on Epidemiologic Data Only

Breast cancer is the second leading cancer for Korean women and its incidence rate has been increasing annually. If early diagnosis were implemented with epidemiologic data, the women could easily assess breast cancer risk using internet. National Cancer Institute in the United States has released a Web-based Breast Cancer Risk Assessment Tool based on Gail model. However, it is inapplicable directly to Korean women since breast cancer risk is dependent on race. Also, it shows low accuracy (58%-59%). In this study, breast cancer discrimination models for Korean women are developed using only epidemiological case-control data (n = 4,574). The models are configured by different classification techniques: support vector machine, artificial neural network, and Bayesian network. A 1,000-time repeated random sub-sampling validation is performed for diverse parameter conditions, respectively. The performance is evaluated and compared as an area under the receiver operating characteristic curve (AUC). According to age group and classification techniques, AUC, accuracy, sensitivity, specificity, and calculation time of all models were calculated and compared. Although the support vector machine took the longest calculation time, the highest classification performance has been achieved in the case of women older than 50 yr (AUC = 64%). The proposed model is dependent on demographic characteristics, reproductive factors, and lifestyle habits without using any clinical or genetic test. It is expected that the model could be implemented as a web-based discrimination tool for breast cancer. This tool can encourage potential breast cancer prone women to go the hospital for diagnostic tests.

Automated Classification to Predict the Progression of Alzheimer's Disease Using Whole-Brain Volumetry and DTI

OBJECTIVE: This study proposes an automated diagnostic method to classify patients with Alzheimer's disease (AD) of degenerative etiology using magnetic resonance imaging (MRI) markers. METHODS: Twenty-seven patients with subjective memory impairment (SMI), 18 patients with mild cognitive impairment (MCI), and 27 patients with AD participated. MRI protocols included three dimensional brain structural imaging and diffusion tensor imaging to assess the cortical thickness, subcortical volume and white matter integrity. Recursive feature elimination based on support vector machine (SVM) was conducted to determine the most relevant features for classifying abnormal regions and imaging parameters, and then a factor analysis for the top-ranked factors was performed. Subjects were classified using nonlinear SVM. RESULTS: Medial temporal regions in AD patients were dominantly detected with cortical thinning and volume atrophy compared with SMI and MCI patients. Damage to white matter integrity was also accredited with decreased fractional anisotropy and increased mean diffusivity (MD) across the three groups. The microscopic damage in the subcortical gray matter was reflected in increased MD. Classification accuracy between pairs of groups (SMI vs. MCI, MCI vs. AD, SMI vs. AD) and among all three groups were 84.4% (+/-13.8), 86.9% (+/-10.5), 96.3% (+/-4.6), and 70.5% (+/-11.5), respectively. CONCLUSION: This proposed method may be a potential tool to diagnose AD pathology with the current clinical criteria.

Automatic pain quantification using autonomic parameters

The objective measurement of subjective, multi-dimensionally experienced pain is a problem for which there has not been an adequate solution. Although verbal methods (e.g., pain scales and questionnaires) are commonly used to measure clinical pain, they tend to lack objectivity, reliability, or validity when applied to mentally impaired individuals. Biopotential and behavioral parameters may represent a solution. Such coding systems already exist, but they are either very costly or time-consuming or have not been sufficiently evaluated. In this context, we collected a database of biopotentials to advance an automated pain recognition system, determine its theoretical testing quality, and optimize its performance. For this purpose, participants were subjected to painful heat stimuli under controlled conditions. One hundred thirty-five features were extracted from the mathematical groupings of amplitude, frequency, stationarity, entropy, linearity, and variability. The following features were chosen as the most selective: (1) electromyography corrugator peak to peak, (2) corrugator shannon entropy, and (3) heart rate variability slope RR. Individual-specific calibration allows the adjustment of feature patterns, resulting in significantly more accurate pain detection rates. The objective measurement of pain in patients will provide valuable information for the clinical team, which may aid the objective assessment of treatment (e.g., effectiveness of drugs for pain reduction, information on surgical indication, and quality of care provided to patients).

Osteoporosis Risk Prediction for Bone Mineral Density Assessment of Postmenopausal Women Using Machine Learning

PURPOSE: A number of clinical decision tools for osteoporosis risk assessment have been developed to select postmenopausal women for the measurement of bone mineral density. We developed and validated machine learning models with the aim of more accurately identifying the risk of osteoporosis in postmenopausal women compared to the ability of conventional clinical decision tools. MATERIALS AND METHODS: We collected medical records from Korean postmenopausal women based on the Korea National Health and Nutrition Examination Surveys. The training data set was used to construct models based on popular machine learning algorithms such as support vector machines (SVM), random forests, artificial neural networks (ANN), and logistic regression (LR) based on simple surveys. The machine learning models were compared to four conventional clinical decision tools: osteoporosis self-assessment tool (OST), osteoporosis risk assessment instrument (ORAI), simple calculated osteoporosis risk estimation (SCORE), and osteoporosis index of risk (OSIRIS). RESULTS: SVM had significantly better area under the curve (AUC) of the receiver operating characteristic than ANN, LR, OST, ORAI, SCORE, and OSIRIS for the training set. SVM predicted osteoporosis risk with an AUC of 0.827, accuracy of 76.7%, sensitivity of 77.8%, and specificity of 76.0% at total hip, femoral neck, or lumbar spine for the testing set. The significant factors selected by SVM were age, height, weight, body mass index, duration of menopause, duration of breast feeding, estrogen therapy, hyperlipidemia, hypertension, osteoarthritis, and diabetes mellitus. CONCLUSION: Considering various predictors associated with low bone density, the machine learning methods may be effective tools for identifying postmenopausal women at high risk for osteoporosis.

Predictors of Medication Adherence in Elderly Patients with Chronic Diseases Using Support Vector Machine Models / 대한의료정보학회지

OBJECTIVES: The aim of this study was to establish a prediction model of medication adherence in elderly patients with chronic diseases and to identify variables showing the highest classification accuracy of medication adherence in elderly patients with chronic diseases using support vector machine (SVM) and conventional statistical methods, such as logistic regression (LR). METHODS: We included 293 chronic disease patients older than 65 years treated at one tertiary hospital. For the medication adherence, Morisky's self-report was used. Data were collected through face-to-face interviews. The mean age of the patients was 73.8 years. The classification process was performed with LR (SPSS ver. 20.0) and SVM (MATLAB ver. 7.12) method. RESULTS: Taking into account 16 variables as predictors, the result of applying LR and SVM classification accuracy was 71.1% and 97.3%, respectively. We listed the top nine variables selected by SVM, and the accuracy using a single variable, self-efficacy, was 72.4%. The results suggest that self-efficacy is a key factor to medication adherence among a Korean elderly population both in LR and SVM. CONCLUSIONS: Medication non-adherence was strongly associated with self-efficacy. Also, modifiable factors such as depression, health literacy, and medication knowledge associated with medication non-adherence were identified. Since SVM builds an optimal classifier to minimize empirical classification errors in discriminating between patient samples, it could achieve a higher accuracy with the smaller number of variables than the number of variables used in LR. Further applications of our approach in areas of complex diseases, treatment will provide uncharted potentials to researchers in the domains.

Predicción de la estructura secundaria de proteínas usando Máquinas de soporte Vectorial / Protein secondary structure prediction using support vector machines

Entre los métodos computacionales utilizados para la predicción de la estructura secundaria de proteí­nas, se destaca el uso de máquinas de soporte vectorial. Este trabajo de investigación presenta la predicción de la estructura secundaria de proteínas desde su secuencia primaria de aminoácidos usando Máquinas de Soporte Vectorial. Como entradas, en la metodologí­a propuesta, se utilizan características de los diferentes motivos estructurales o cadenas de texto asociadas a la estructura primaria que representa la estructura secundaria, tales como el R-grupo y la probabilidad de que el aminoácido en la posición central adopte una determinada estructura secundaria. Para la extracción de características se utiliza un método de codificación de secuencias en el que cada símbolo en la estructura primaria se relaciona con cada sí­mbolo en la estructura secundaria. El uso de este método de codificación permite reducir la dimensionalidad de los datos de miles de características a sólo 220 de estas. Los resultados obtenidos son comparables a los registrados en la literatura, teniendo cerca de un 70% de precisión. Además, se logra reducir los costos computacionales en la construcción de los clasificadores debido a que este trabajo modela el problema de multi-clasificación como un grupo de clasificadores binarios.

Among the computational methods used for predicting secondary structure proteins highlights the use of support vector machines. This research shows the predicted secondary structure of protein from its primary amino acid sequence using Support Vector Machines. As inputs, in the proposed methodology, features are used from different structural motifs or text strings associated with the primary structure which represents the secondary structure, such as R-group and the probability that the amino acid at position adopts a central particular secondary structure. For feature extraction method is used coding of sequences in which each symbol in the primary structure is associated with each symbol in the secondary structure. The use of this encoding method reduces the dimensionality of the data of thousands of characteristics only 220 of these. The results obtained are comparable to those reported in the literature, taking about 70% accuracy. Furthermore, it is possible to reduce computational cost in the construction of classifiers because this work models the problem of multi classification as a group of binary classifiers.

Identification and analysis of anti-diabetic drugs based on terahertz time-domain spectroscopy / 医用生物力学

Objective To study the vibration frequencies and resonant peaks of different molecular groups in anti-diabetic drugs and to investigate the absorption spectra of these drugs in the range of terahertz, so as to accurately and efficiently identify similar drugs and provide foundation for understanding the contribution of such vibration in different molecular groups to pharmacology. Methods Using terahertz time domain spectroscopy (THz-TDS), six kinds of similar diabetes pills for treating diabetes mellitus (DM), including gliquidone, glipizide, gliclazide, glimepiride, repaglinide and metformin were tested to obtain their absorption spectra from 0.3~3.0 THz. Support vector machines (SVM) method was employed to separate these anti-diabetic drugs by selecting the experimental data from 1.5~2.0 THz. Results For gliquidone, glipizide, gliclazide, and glimepiride, an obvious resonance peak was found at 1.37 THz. According to the THz spectra, repaglinide and metformin could be easily separated from sulfonylurea drugs. Furthermore, with the help of SVM, the accuracy of discrimination for four kinds of sulfonylurea could reach 100%. Conclusions THz technology, as a new testing method, shall play a positive role in the drugs for their identification, quality control and distinguishing their chemical bonds/functional group.

A Comparison of Intensive Care Unit Mortality Prediction Models through the Use of Data Mining Techniques / 대한의료정보학회지

OBJECTIVES: The intensive care environment generates a wealth of critical care data suited to developing a well-calibrated prediction tool. This study was done to develop an intensive care unit (ICU) mortality prediction model built on University of Kentucky Hospital (UKH)'s data and to assess whether the performance of various data mining techniques, such as the artificial neural network (ANN), support vector machine (SVM) and decision trees (DT), outperform the conventional logistic regression (LR) statistical model. METHODS: The models were built on ICU data collected regarding 38,474 admissions to the UKH between January 1998 and September 2007. The first 24 hours of the ICU admission data were used, including patient demographics, admission information, physiology data, chronic health items, and outcome information. RESULTS: Only 15 study variables were identified as significant for inclusion in the model development. The DT algorithm slightly outperformed (AUC, 0.892) the other data mining techniques, followed by the ANN (AUC, 0.874), and SVM (AUC, 0.876), compared to that of the APACHE III performance (AUC, 0.871). CONCLUSIONS: With fewer variables needed, the machine learning algorithms that we developed were proven to be as good as the conventional APACHE III prediction.

Automatic classffication of structural mri for diagnosis of neurodegenerative diseaess / Clasificación automática de IRM estructural para el diagnóstico de enfermedades neurodegenerativas

This paper presents an automatic approach which classifies structural Magnetic Resonance images into pathological or healthy controls. A classification model was trained to find the boundaries that allow to separate the study groups. The method uses the deformation values from a set of regions, automatically identified as relevant, in a process that selects the statistically significant regions of a t-test under the restriction that this significance must be spatially coherent within a neighborhood of 5 voxels. The proposed method was assessed to distinguish healthy controls from schizophrenia patients. Classification results showed accuracy between 74% and 89%, depending on the stage of the disease and number of training samples.

Este artículo presenta un método automático para la clasificación de individuos en grupos patológicos o controles sanos haciendo uso de imágenes de resonancia magnética. El método propuesto usa los valores de deformación del sujeto analizado a un cerebro plantilla, para entrenar un modelo de clasificación capaz de identificar las fronteras que separan los grupos de estudio en un espacio de características dado. Con el fin de reducir la dimensionalidad del problema, un conjunto de regiones relevantes es automáticamente extraído en un proceso que selecciona las regiones estadísticamente significativas en una prueba t-student, con la restricción de mantener coherencia en dicha significancia en una vecindad de 5 voxeles. El método propuesto fue evaluado en la clasificación de pacientes con esquizofrenia y sujetos sanos. Los resultados mostraron un desempeño entre el 74 y el 89%, el cual depende principalmente del número de muestras empleadas para el entrenamiento del modelo.

Rough Set and Support Vector Machines for Assistant Detection of Parkinson Disease / 中国康复理论与实践

@# Objective To study the feasibility of using rough set and support vector machines to detect Parkinson disease. Methods The reduction algorithm based on the importance of the attributes in the rough set theory was used to reduce the common diagnosis features in the clinical practice. The support vector machines were applied for classification with the linear, polynomial and RBF kernel, and the Results were compare with that of BP neural network. Results The algorism combined attributes reduction and support vector machines appeared the highest accuracy of 92.71% in the classification, which seemed greater advantage in accuracy and stability than BP neural network. Conclusion Improving the accuracy of the classification as well as saving the resources, rough set and support vector machines are proved to be an effective method to assist the clinical diagnosis of Parkinson disease.

Establishment of a diagnostic model from plasma proteomic patterns of ovarian serous cystadenocarcinoma patients using SELDI-TOF-MS technology / 肿瘤

Objective: To screen differentiated expressed proteins in plasma of ovarian serous cystadenocarcinoma patients by using surface enhanced laser desorption/ionization time of flight mass spectrometry (SELDI-TOF-MS) associated with bioinformatic support vector machines (SVM) and discuss how to establish algorithmic logical model for diagnosis of ovarian serous cys-tadenocarcinoma and its significance. Methods: SELDI-TOF-MS and CM10 chip were used to analyze the plasma samples from 26 ovarian serous cystadenocarcinoma women and 51 control women including 12 cases of ovarian cyst, 31 cases of uterous leiomyoma, 8 cases of ovarian benign cystadenoma. The data was analyzed by Biomarker Wizard software. The plasma proteomic diagnostic model for ovarian serous cys-tadenocarcinoma patients and control subjects were established by using SVM (a bioinformatic method). Results: Seventy-one differentiated protein peaks were screened by Biomarker Wizard software which were captured by SELDI-TOF-MS from CM10 chip (P <0.01). The proteomic profiling for ovarian serous cystadenocarcinoma was optimized by SVM re-screening. The key m/z value of these 7 proteins was 4 099, 4 477, 4 123, 4 081 and 3 938 (up-regulated), 8 785 and 13 783 (down-regulated). Three-fold cross validation followed by blinded determination demonstrated that the sensitivity and specificity of the established model were 84.62% and 96.08% separately, and the positive predictive value was 92.21% for differential diagnosis of ovarian serous cystadenocarcinoma patients. Conclusion: ProteinChip-mass spectrometry technology can rapidly and effectively screen differentiated proteins from the plasma of ovarian serous cystadenocarcinoma patients. Combined with SVM, a diagnostic model was generated from proteomic patterns of ovarian serous cystadenocarcinoma, which had potential significance for establishing diagnostic methods for ovarian cancer.

Personalized diagnosis by cached solutions with hypertension as a study model

Statistical modeling of links between genetic profiles with environmental and clinical data to aid in medical diagnosis is a challenge. Here, we present a computational approach for rapidly selecting important clinical data to assist in medical decisions based on personalized genetic profiles. What could take hours or days of computing is available on-the-fly, making this strategy feasible to implement as a routine without demanding great computing power. The key to rapidly obtaining an optimal/nearly optimal mathematical function that can evaluate the [quot ]disease stage[quot ] by combining information of genetic profiles with personal clinical data is done by querying a precomputed solution database. The database is previously generated by a new hybrid feature selection method that makes use of support vector machines, recursive feature elimination and random sub-space search. Here, to evaluate the method, data from polymorphisms in the renin-angiotensin-aldosterone system genes together with clinical data were obtained from patients with hypertension and control subjects. The disease [quot ]risk[quot ] was determined by classifying the patients' data with a support vector machine model based on the optimized feature; then measuring the Euclidean distance to the hyperplane decision function. Our results showed the association of renin-angiotensin-aldosterone system gene haplotypes with hypertension. The association of polymorphism patterns with different ethnic groups was also tracked by the feature selection process. A demonstration of this method is also available online on the project's web site.

Protein Fold Recognition With Support Vector Machines Fusion Network / 生物化学与生物物理进展

One of the important approaches to structure analysis is protein fold recognition, which is oftenapplied when there is no significant sequence similarity between structurally similar proteins. A framework with athree-layer support vector machines fusion network (SFN) is presented. The framework is applied to 27-classprotein fold recognition from primary structure of proteins. SFN uses support vector machines as memberclassifiers, and adopts All-Versus-All as multi-class categorization. Six groups of features are divided into majorand minor ones by SFN, and several diversity fusion schemes are correspondingly built. The final decision is madeby dynamic selection of the results of all fusion schemes. When it is still difficult to know what kind of fusion offeature groups can achieve good prediction,SFN is a dependable solution by selecting the optimal fusion offeature groups automatically, which can ensure the best recognition. Overall recognition system achieves 61.04%fold prediction accuracy on the independent test dataset. The results and the comparison with other approachesdemonstrate the effectiveness of SFN, and thus encourage its further exploration.

Support vector machines for novel class detection in Bioinformatics

Novelty detection techniques might be a promising way of dealing with high-dimensional classification problems in Bioinformatics. We present preliminary results of the use of a one-class support vector machine approach to detect novel classes in two Bioinformatics databases. The results are compatible with theory and inspire further investigation.