RESUMEN
BACKGROUND: Investigating the functional interactions between different brain regions and revealing the transmission of information by computing brain connectivity have great potential and significance in the diagnosis of early Mild Cognitive Impairment (EMCI). METHODS: The Granger causality with Gate Recurrent Unit (GRU_GC) model is a suitable method that allows the detection of a nonlinear causal relationship and solves the limitation of fixed time lag, which cannot be detected by the classical Granger method. The model can transmit time series signals with any transmission delay length, and the time series can be screened and learned through the gate model. RESULTS: The classification experiment of 89 EMCI and 73 neurologically healthy controls (HC) shows that the accuracy reached 87.88%. Compared with multivariate variables GC (MVGC) and Long Short-Term Memory-based GC (LSTM_GC), the GRU_GC significantly improved the estimation of brain connectivity communication. Constructing a difference network to explore the brain effective connectivity between EMCI and HC. CONCLUSIONS: The GRU_GC can discover the abnormal brain regions, including the parahippocampal gyrus, the posterior cingulate gyrus. The method can be used in clinical applications as an effective brain connectivity analysis tool and provides auxiliary means for the medical diagnosis of EMCI.
RESUMEN
It is important to explore causal relationships in functional magnetic resonance imaging study. However, the traditional effective connectivity analysis method is easy to produce false causality, and the detection accuracy needs to be improved. In this paper, we introduce a novel functional magnetic resonance imaging effective connectivity method based on the asymmetry detection of transfer entropy, which quantifies the disparity in predictive information between forward and backward time, subsequently normalizing this disparity to establish a more precise criterion for detecting causal relationships while concurrently reducing computational complexity. Then, we evaluate the effectiveness of this method on the simulated data with different level of nonlinearity, and the results demonstrated that the proposed method outperforms others methods on the detection of both linear and nonlinear causal relationships, including Granger Causality, Partial Granger Causality, Kernel Granger Causality, Copula Granger Causality, and traditional transfer entropy. Furthermore, we applied it to study the effective connectivity of brain functional activities in seafarers. The results showed that there are significantly different causal relationships between different brain regions in seafarers compared with non-seafarers, such as Temporal lobe related to sound and auditory information processing, Hippocampus related to spatial navigation, Precuneus related to emotion processing as well as Supp_Motor_Area associated with motor control and coordination, which reflects the occupational specificity of brain function of seafarers.
Asunto(s)
Encéfalo , Imagen por Resonancia Magnética , Entropía , Encéfalo/diagnóstico por imagen , Emociones , CogniciónRESUMEN
Judging and identifying biological activities and biomarkers inside tissues from imaging features of diseases is challenging, so correlating pathological image data with genes inside organisms is of great significance for clinical diagnosis. This paper proposes a high-dimensional kernel non-negative matrix factorization (NMF) method based on muti-modal information fusion. This algorithm can project RNA gene expression data and pathological images (WSI) into a common feature space, where the heterogeneous variables with the largest coefficient in the same projection direction form a co-module. In addition, the miRNA-mRNA and miRNA-lncRNA interaction networks in the ceRNA network are added to the algorithm as a priori information to explore the relationship between the images and the internal activities of the gene. Furthermore, the radial basis kernel function is used to calculate the feature proportion between different kinds of genes mapped in the high-dimensional feature space and projected into the common feature space to explore the gene interaction in the high-dimensional situation. The original feature matrix is regularized to improve biological correlation, and the feature factors are sparse by orthogonal constraints to reduce redundancy. Experimental results show that the proposed NMF method is better than the traditional NMF method in stability, decomposition accuracy, and robustness. Through data analysis applied to lung cancer, genes related to tissue morphology are found, such as COL7A1, CENPF and BIRC5. In addition, gene pairs with a correlation degree exceeding 0.8 are found, and potential biomarkers of significant correlation with survival are obtained such as CAPN8. It has potential application value for the clinical diagnosis of lung cancer.
Asunto(s)
Neoplasias Pulmonares , MicroARNs , Humanos , Neoplasias Pulmonares/diagnóstico por imagen , Neoplasias Pulmonares/genética , Algoritmos , MicroARNs/genética , Biomarcadores , Colágeno Tipo VIIRESUMEN
OBJECTIVE: Long non-coding RNAs (lncRNAs) have been shown to be associated with the pathogenesis of different kinds of diseases and play important roles in various biological processes. Although numerous lncRNAs have been found, the functions of most lncRNAs and physiological/pathological significance are still in its infancy. Meanwhile, their expression patterns and regulation mechanisms are also far from being fully understood. METHODS: In order to reveal functional lncRNAs and identify the key lncRNAs, we develop a new sparse independence component analysis (ICA) method to identify lncRNA-mRNA-miRNA expression co-modules based on the competitive endogenous RNA (ceRNA) theory using the sample-matched lncRNA, mRNA and miRNA expression profiles. The expression data of the three RNA combined together is approximated sparsely to obtain the corresponding sparsity coefficient, and then it is decomposed by using ICA constraint optimization to obtain the common basis and modules. Subsequently, affine propagation clustering is used to perform cluster analysis on the common basis under multiple running conditions to obtain the co-modules for the selection of different RNA elements. RESULTS: We applied sparse ICA to Liver Hepatocellular Carcinoma (LIHC) dataset and the experiment results demonstrate that the proposed sparse ICA method can effectively discover biologically functional expression common modules. CONCLUSION: It may provide insights into the function of lncRNAs and molecular mechanism of LIHC. Clinical and Translational Impact Statement-The results on LIHC dataset demonstrate that the proposed sparse ICA method can effectively discover biologically functional expression common modules, which may provide insights into the function of IncRNAs and molecular mechanism of LIHC.
Asunto(s)
Carcinoma Hepatocelular , Neoplasias Hepáticas , MicroARNs , ARN Largo no Codificante , Humanos , Carcinoma Hepatocelular/genética , ARN Largo no Codificante/genética , Neoplasias Hepáticas/genética , Redes Reguladoras de Genes , Regulación Neoplásica de la Expresión Génica/genética , MicroARNs/genética , ARN Mensajero/genéticaRESUMEN
At present, fMRI studies mainly focus on the entire low-frequency band (0. 01-0.08 Hz). However, the neuronal activity is dynamic, and different frequency bands may contain different information. Therefore, a novel multi-frequency-based dynamic functional connectivity (dFC) analysis method was proposed in this study, which was then applied to a schizophrenia study. First, three frequency bands (Conventional: 0.01-0.08 Hz, Slow-5: 0.0111-0.0302 Hz, and Slow-4: 0.0302-0.0820 Hz) were obtained using Fast Fourier Transform. Next, the fractional amplitude of low-frequency fluctuations was used to identify abnormal regions of interest (ROIs) of schizophrenia, and dFC among these abnormal ROIs was implemented by the sliding time window method at four window-widths. Finally, recursive feature elimination was employed to select features, and the support vector machine was applied for the classification of patients with schizophrenia and healthy controls. The experimental results showed that the proposed multi-frequency method (Combined: Slow-5 and Slow-4) had a better classification performance compared with the conventional method at shorter sliding window-widths. In conclusion, our results revealed that the dFCs among the abnormal ROIs varied at different frequency bands and the efficiency of combining multiple features from different frequency bands can improve classification performance. Therefore, it would be a promising approach for identifying brain alterations in schizophrenia.
RESUMEN
Migraine is a common, chronic dysfunctional disease with recurrent headaches. Its etiology and pathogenesis have not been fully understood and there is a lack of objective diagnostic criteria and biomarkers. Meanwhile, resting-state functional magnetic resonance imaging (RS-fMRI) is increasingly being used in migraine research to classify and diagnose brain disorders. However, the RS-fMRI data is characterized by a large amount of data information and the difficulty of extracting high-dimensional features, which brings great challenges to relevant studies. In this paper, we proposed an automatic recognition framework based on static functional connectivity (sFC) strength features and dynamic functional connectome pattern (DFCP) features of migraine sufferers and normal control subjects, in which we firstly extracted sFC strength and DFCP features and then selected the optimal features using the recursive feature elimination based on the support vector machine (SVM-RFE) algorithm and, finally, trained and tested a classifier with the support vector machine (SVM) algorithm. In addition, we compared the classification performance of only using sFC strength features and DFCP features, respectively. The results showed that the DFCP features significantly outperformed sFC strength features in performance, which indicated that DFCP features had a significant advantage over sFC strength features in classification. In addition, the combination of sFC strength and DFCP features had the optimal performance, which demonstrated that the combination of both features could make full use of their advantage. The experimental results suggested the method had good performance in differentiating migraineurs and our proposed classification framework might be applicable for other mental disorders.
RESUMEN
The particularity of seafarers' occupation makes their brain functional activities vulnerable to the influence of working environments, which leads to abnormal functional connectivities (FCs) between brain networks. To further investigate the influences of maritime environments on the seafarers' functional brain networks, the functional magnetic resonance imaging (fMRI) datasets of 33 seafarers before and after sailing were used to study FCs among the functional brain networks in this paper. On the basis of making full use of the intrinsic prior information from fMRI data, six resting-state brain functional networks of seafarers before and after sailing were obtained by using group independent component analysis with intrinsic reference, and then the differences between the static and dynamic FCs among these six brain networks of seafarers before and after sailing were, respectively, analyzed from both group and individual levels. Subsequently, the potential dynamic functional connectivity states of seafarers before and after sailing were extracted by using the affine propagation clustering algorithm and the probabilities of state transition between them were obtained simultaneously. The results show that the dynamic FCs among large-scale brain networks have significant difference seafarers before and after sailing both at the group level and individual level, while the static FCs between them varies only at the individual level. This suggests that the maritime environments can indeed affect the brain functional activity of seafarers in real time, and the degree of influence is different for different subjects, which is of a great significance to explore the neural changes of seafarer's brain functional network.
RESUMEN
The Extreme Learning Machine (ELM) is a simple and efficient Single Hidden Layer Feedforward Neural Network(SLFN) algorithm. In recent years, it has been gradually used in the study of Alzheimer's disease (AD). When using ELM to diagnose AD based on high-dimensional features, there are often some features that have no positive impact on the diagnosis, while others have a significant impact on the diagnosis. In this paper, a novel Key Features Screening Method based on Extreme Learning Machine (KFS-ELM) is proposed. It can screen for key features that are relevant to the classification (diagnosis). It can also assign weights to key features based on their importance. We designed an experiment to screen for key features of AD. A total of 920 key functional connections screened from 4005 functional connections. Their weights were also obtained. The results of the experiment showed that: (1) Using all (4,005) features to diagnose AD, the accuracy is 95.33%. Using 920 key features to diagnose AD, the accuracy is 99.20%. The 3,085 (4,005 - 920) features that were screened out had a negative effect on the diagnosis of AD. This indicates the KFS-ELM is effective in screening key features. (2) The higher the weight of the key features and the smaller their number, the greater their impact on AD diagnosis. This indicates that the KFS-ELM is rational in assigning weights to the key features for their importance. Therefore, KFS-ELM can be used as a tool for studying features and also for improving classification accuracy.
RESUMEN
Attention-deficit/hyperactivity disorder (ADHD) is a common childhood psychiatric disorder that often persists into adulthood. Extracting brain networks from functional magnetic resonance imaging (fMRI) data can help explore neurocognitive disorders in adult ADHD. However, there is still a lack of effective methods to extract large-scale brain networks to identify disease-related brain network changes. Hence, this study proposed a spatial constrained non-negative matrix factorization (SCNMF) method based on the fMRI real reference signal. First, non-negative matrix factorization analysis was carried out on each subject to select the brain network components of interest. Subsequently, the available spatial prior information was mined by integrating the interested components of all subjects. This prior constraint was then incorporated into the NMF objective function to improve its efficiency. For the sake of verifying the effectiveness and feasibility of the proposed method, we quantitatively compared the SCNMF method with other classical algorithms and applied it to the dynamic functional connectivity analysis framework. The algorithm successfully extracted ten resting-state brain functional networks from fMRI data of adult ADHD and healthy controls and found large-scale brain network changes in adult ADHD patients, such as enhanced connectivity between executive control network and right frontoparietal network. In addition, we found that older ADHD spent more time in the pattern of relatively weak connectivity. These findings indicate that the method can effectively extract large-scale functional networks and provide new insights into understanding the neurobiological mechanisms of adult ADHD from the perspective of brain networks.
RESUMEN
Psychiatric disorders usually have similar clinical and neurobiological features. Nevertheless, previous research on functional dysconnectivity has mainly focused on a single disorder and the transdiagnostic alterations in brain networks remain poorly understood. Hence, this study proposed a spatiotemporal constrained nonnegative matrix factorization (STCNMF) method based on real reference signals to extract large-scale brain networks to identify transdiagnostic changes in neurocognitive networks associated with multiple diseases. Available temporal prior information and spatial prior information were first mined from the functional magnetic resonance imaging (fMRI) data of group participants, and then these prior constraints were incorporated into the nonnegative matrix factorization objective functions to improve their efficiency. The algorithm successfully obtained 10 resting-state functional brain networks in fMRI data of schizophrenia, bipolar disorder, attention deficit/hyperactivity disorder, and healthy controls, and further found transdiagnostic changes in these large-scale networks, including enhanced connectivity between right frontoparietal network and default mode network, reduced connectivity between medial visual network and default mode network, and the presence of a few hyper-integrated network nodes. Besides, each type of psychiatric disorder had its specific connectivity characteristics. These findings provide new insights into transdiagnostic and diagnosis-specific neurobiological mechanisms for understanding multiple psychiatric disorders from the perspective of brain networks.
Asunto(s)
Trastorno por Déficit de Atención con Hiperactividad , Encéfalo , Algoritmos , Trastorno por Déficit de Atención con Hiperactividad/diagnóstico por imagen , Encéfalo/diagnóstico por imagen , Mapeo Encefálico/métodos , Humanos , Imagen por Resonancia Magnética , Red Nerviosa/diagnóstico por imagenRESUMEN
Resting-state functional MRI (rs-fMRI) has been increasingly applied in the research of brain cognitive science and psychiatric diseases. However, previous studies only focused on specific activation areas of the brain, and there are few studies on the inactivation areas. This may overlook much information that explains the brain's cognitive function. In this paper, we propose a relatively inert network (RIN) and try to explore its important role in understanding the cognitive mechanism of the brain and the study of mental diseases, using adult attention deficit hyperactivity disorder (ADHD) as an example. Here, we utilize methods based on group independent component analysis (GICA) and t-test to identify RIN and calculate its corresponding time series. Through experiments, alterations in the RIN and the corresponding activation network (AN) in adult ADHD patients are observed. And compared with those in the left brain, the activation changes in the right brain are greater. Further, when the RIN functional connectivity is introduced as a feature to classify adult ADHD patients from healthy controls (HCs), the classification accuracy rate is 12% higher than that of the original functional connectivity feature. This was also verified by testing on an independent public dataset. These findings confirm that the RIN of the brain contains much information that will probably be neglected. Moreover, this research provides an effective new means of exploring the information integration between brain regions and the diagnosis of mental illness.
RESUMEN
BACKGROUND AND OBJECTIVE: The modular structure and hierarchy are important topological characteristics in real complex networks such as brain networks on temporal scale. However, there are few studies investigating the hierarchical structure at the spatial scale of brain networks, the application of which still remains to be further studied. METHODS: In this study, a novel model of brain hierarchical network based on the hierarchical characteristic of Internet topology is proposed for the first time, which is called Internet-like brain hierarchical network (IBHN). In this model, the whole brain network is partitioned into multiple levels: brain wide area network (Brain-WAN), brain metropolitan network (Brain-MAN), and brain local area network (Brain-LAN). A Brain-MAN is formed by the interconnection of multiple Brain-LANs, and the interconnection of multiple Brain-MANs forms a Brain-WAN. A multivariate analysis method is employed to measure overall functional connectivity between two brain networks at the same network level rather than detecting the change of each node pair's functional connection. Furthermore, we demonstrate the utility of IBHN model with application to a practical case-control study involving 64 patients with Alzheimer's disease and 75 healthy controls. RESULTS: The proposed model identified enhanced functional connectivity (P-value<0.05) at Brain-WAN level and reduced functional connectivity (P-value=0.004) at Brain-LAN level of Alzheimer's disease patients, which can be used as a multi-dimension functional reference for AD's diagnosis. CONCLUSIONS: This study not only provides insight into AD pathophysiology, but also further proves the effectiveness of the proposed IBHN model. In addition, the IBHN model makes it possible to explore the brain's functional organization from multiple dimensions and offers a multi-level perspective for the research of complex brain network.
Asunto(s)
Enfermedad de Alzheimer , Enfermedad de Alzheimer/diagnóstico por imagen , Encéfalo/diagnóstico por imagen , Mapeo Encefálico , Estudios de Casos y Controles , Humanos , Internet , Imagen por Resonancia Magnética , Red Nerviosa/diagnóstico por imagenRESUMEN
As a special occupational group, the working and living environments faced by seafarers are greatly different from those of land. It is easy to affect the psychological and physiological activities of seafarers, which inevitably lead to changes in the brain functional activities of seafarers. Therefore, it is of great significance to study the neural activity rules of seafarers' brain. In view of this, this paper studied the seafarers' brain alteration at the activated voxel level based on functional magnetic resonance imaging technology by comparing the differences in functional connectivities (FCs) between seafarers and non-seafarers. Firstly, the activated voxels of each group were obtained by independence component analysis, and then the distribution of these voxels in the brain and the common activated voxels between the two groups were statistically analyzed. Next, the FCs between the common activated voxels of the two groups were calculated and obtained the FCs that had significant differences between them through two-sample T-test. Finally, all FCs and FCs with significant differences (DFCs) between the common activated voxels were used as the features for the support vector machine to classify seafarers and non-seafarers. The results showed that DFCs between the activated voxels had better recognition ability for seafarers, especially for Precuneus_L and Precuneus_R, which may play an important role in the classification prediction of seafarers and non-seafarers, so that provided a new perspective for studying the specificity of neurological activities of seafarers.
RESUMEN
Migraine seriously affects the physical and mental health of patients because of its recurrence and the hypersensitivity to the environment that it causes. However, the pathogenesis and pathophysiology of migraine are not fully understood. We addressed this issue in the present study using an autodynamic functional connectome model (A-DFCM) with twice-clustering to compare dynamic functional connectome patterns (DFCPs) from resting-state functional magnetic resonance imaging data from migraine patients and normal control subjects. We used automatic localization of segment points to improve the efficiency of the model, and intergroup differences and network metrics were analyzed to identify the neural mechanisms of migraine. Using the A-DFCM model, we identified 17 DFCPs-including 1 that was specific and 16 that were general-based on intergroup differences. The specific DFCP was closely associated with neuronal dysfunction in migraine, whereas the general DFCPs showed that the 2 groups had similar functional topology as well as differences in the brain resting state. An analysis of network metrics revealed the critical brain regions in the specific DFCP; these were not only distributed in brain areas related to pain such as Brodmann area 1/2/3, basal ganglia, and thalamus but also located in regions that have been implicated in migraine symptoms such as the occipital lobe. An analysis of the dissimilarities in general DFCPs between the 2 groups identified 6 brain areas belonging to the so-called pain matrix. Our findings provide insight into the neural mechanisms of migraine while also identifying neuroimaging biomarkers that can aid in the diagnosis or monitoring of migraine patients.
Asunto(s)
Encéfalo/diagnóstico por imagen , Encéfalo/fisiopatología , Conectoma/métodos , Imagen por Resonancia Magnética/métodos , Trastornos Migrañosos/diagnóstico por imagen , Trastornos Migrañosos/fisiopatología , Adolescente , Adulto , Algoritmos , Análisis por Conglomerados , Biología Computacional , Conectoma/estadística & datos numéricos , Femenino , Humanos , Imagen por Resonancia Magnética/estadística & datos numéricos , Masculino , Persona de Mediana Edad , Modelos Neurológicos , Redes Neurales de la Computación , Descanso/fisiología , Adulto JovenRESUMEN
Microenvironment-driven tumor heterogeneity causes the limitation of immunotherapy of sarcomas. Nonetheless, systematical studies of various molecular levels can enhance the understanding of tumor microenvironment (TME) related to prognosis and provide novel insights of precision immunotherapy. Three prognostic-related TME phenotypes were identified by consensus clustering of the relative infiltration of 22 immune cells from 869 samples of sarcomas. Additionally, integrative immunogenomic analysis is applied to explore the characteristics of different TME groups. The results revealed that most of the immune cell infiltration is higher in the better prognostic group, which are more affected by lower DNA methylation levels and fewer copy number variations in the worse prognostic group. The signaling pathway crosstalk analysis suggested that the changes in the TME will cause considerable variation in the flow of information between pathways, especially when the degree of relative infiltration of immune cells is low, patient's endocrine system may also be significantly affected. Also, the endogenous competitive network analysis indicated that several differentially expressed long non-coding RNAs (lncRNAs) associated with the prognosis or tumor recurrence of sarcoma patients affected the regulatory relationship between miRNAs and different genes when the sarcoma microenvironment changes. In summary, the significant relationship between genetic alterations and prognostic-related TME characteristics in sarcomas were determined in this study. These findings may provide new clues for the treatment of sarcomas.
RESUMEN
Many studies have revealed the structural or functional brain changes induced by occupational factors. However, it remains largely unknown how occupation-related connectivity shapes the brain. In this paper, we denote occupational neuroplasticity as the neuroplasticity that takes place to satisfy the occupational requirements by extensively professional training and to accommodate the long-term, professional work of daily life, and a critical review of occupational neuroplasticity related to the changes in brain structure and functional networks has been primarily presented. Furthermore, meta-analysis revealed a neurophysiological mechanism of occupational neuroplasticity caused by professional experience. This meta-analysis of functional neuroimaging studies showed that experts displayed stronger activation in the left precentral gyrus [Brodmann area (BA)6], left middle frontal gyrus (BA6), and right inferior frontal gyrus (BA9) than novices, while meta-analysis of structural studies suggested that experts had a greater gray matter volume in the bilateral superior temporal gyrus (BA22) and right putamen than novices. Together, these findings not only expand the current understanding of the common neurophysiological basis of occupational neuroplasticity across different occupations and highlight some possible targets for neural modulation of occupational neuroplasticity but also provide a new perspective for occupational science research.
RESUMEN
BACKGROUND: Studies of brain functional connectivity (FC) and effective connectivity (EC) using the functional magnetic resonance imaging (fMRI) have advanced our understanding of functional organization on visual cortex of human brain. The current studies mainly focus on static or dynamic connectivity, while the relationships between them have not been well characterized especially for static EC (sEC) and dynamic EC (dEC), as well as the consistency characteristics of changing trend of dFCs and dECs, which is of great importance to reveal the neural information processing mechanism in visual cortex region. METHOD: In this study, we explore these relationships among several subareas of human visual cortex (V1-V5) by calculating the connection intensity and information flow among them over time by sliding window method, which are defined by Pearson correlation coefficient and Granger causality analysis, respectively, in each window. RESULTS: The results demonstrate that there are extensive connections existing in human visual network, which are time-varying both in resting and task-related states. sFC intensity is negatively correlated with the variance of dFC, while sEC intensity is positively correlated with the variance of dEC. Furthermore, we also find that dFC within visual cortex at rest shows more consistency, while dEC shows less compared with task state in changing trend. CONCLUSION: Therefore, this study provides novel findings about dynamics of connectivity in human visual cortex from the perspective of functional and effective connectivity.
Asunto(s)
Corteza Visual , Encéfalo , Mapeo Encefálico , Humanos , Imagen por Resonancia Magnética , Vías Nerviosas/diagnóstico por imagen , Neuroimagen , Corteza Visual/diagnóstico por imagenRESUMEN
BACKGROUND: Alzheimer's disease (AD) is a common neurodegenerative disease occurring in the elderly population. The effective and accurate classification of AD symptoms by using functional magnetic resonance imaging (fMRI) has a great significance for the clinical diagnosis and prediction of AD patients. METHODS: Therefore, this paper proposes a new method for identifying AD patients from healthy subjects by using functional connectivities (FCs) between the activity voxels in the brain based on fMRI data analysis. Firstly, independent component analysis is used to detect the activity voxels in the fMRI signals of AD patients and healthy subjects; Secondly, the FCs between the common activity voxels of the two groups are calculated, and then the FCs with significant differences are further identified by statistical analysis between them; Finally, the classification of AD patients from healthy subjects is realized by using FCs with significant differences as the feature samples in support vector machine. RESULTS: The results show that the proposed identification method can obtain higher classification accuracy, and the FCs between activity voxels within prefrontal lobe as well as those between prefrontal and parietal lobes play an important role in the prediction of AD patients. Furthermore, we also find that more brain regions and much more voxels in some regions are activity in AD group compared with health control group. CONCLUSION: It has a great potential value for the AD pathogenesis mechanism study.
RESUMEN
Extracting massive features from images to quantify tumors provides a new insight to solve the problem that tumor heterogeneity is difficult to assess quantitatively. However, quantification of tumors by single-mode methods often has defects such as difficulty in features extraction and high computational complexity. The multimodal approach has shown effective application prospects in solving these problems. In this paper, we propose a feature fusion method based on positron emission tomography (PET) images and clinical information, which is used to obtain features for lung metastasis prediction of soft tissue sarcomas (STSs). Random forest method was adopted to select effective features by eliminating irrelevant or redundant features, and then they were used for the prediction of the lung metastasis combined with back propagation (BP) neural network. The results show that the prediction ability of the proposed model using fusion features is better than that of the model using an image or clinical feature alone. Furthermore, a good performance can be obtained using 3 standard uptake value (SUV) features of PET image and 7 clinical features, and its average accuracy, sensitivity, and specificity on all the sets can reach 92%, 91%, and 92%, respectively. Therefore, the fusing features have the potential to predict lung metastasis for STSs.
Asunto(s)
Interpretación de Imagen Asistida por Computador/métodos , Neoplasias Pulmonares/diagnóstico por imagen , Neoplasias Pulmonares/secundario , Tomografía de Emisión de Positrones/métodos , Sarcoma/diagnóstico por imagen , Sarcoma/secundario , Neoplasias de los Tejidos Blandos/diagnóstico por imagen , Neoplasias de los Tejidos Blandos/patología , Adulto , Anciano , Biología Computacional , Femenino , Fluorodesoxiglucosa F18 , Humanos , Interpretación de Imagen Asistida por Computador/estadística & datos numéricos , Masculino , Persona de Mediana Edad , Redes Neurales de la Computación , Tomografía de Emisión de Positrones/estadística & datos numéricos , RadiofármacosRESUMEN
Granger causality (GC) is one of the most popular measures to investigate causality influence among brain regions and has been achieved significant results for exploring brain networks based on functional magnetic resonance imaging (fMRI). However, the predictors and order selection of conventional GC are based on linear models which result in such restrictions as poorly detection of nonlinearity and so on, in the application. This paper proposes a novel GC model called back propagation (BP) based kernel function Granger causality (BP_KFGC), in which symplectic geometry is used for embedding dimension and fuzzy inference system for predicting time series. The proposed method doesn't depend on the prediction of the vector auto-regression model, so that time series don't need to be wide-sense stationary as linear GC and kernel GC. In addition, it is a multivariate approach which is applicable to both linear and nonlinear systems and eliminates the effects of latent variables. The performance of the new method is evaluated and compared with linear GC, partial GC, neural network GC and kernel GC by simulated data with multiple adjustments to the nonlinearity. The results show that BP_KFGC outperforms the other four methods in detecting both linear and nonlinear causalities. Furthermore, we applied BP_KFGC to construct directed weight network (DWN) of Alzheimer's disease (AD) patients and health controls (HCs), and then nine graph-based features of DWN were used for classification by the classifier of support vector machine with radial basis kernel function. The accuracy of 95.89%, sensitivity of 93.31%, and specificity of 94.97% were achieved which may provide an auxiliary mean for the clinical diagnosis of AD.