shuai huang - Academia.edu (original) (raw)

Papers by shuai huang

Research paper thumbnail of Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation

Neuroimage, 2010

Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disea... more Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number of brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88% and 88%, respectively.

Research paper thumbnail of A Transfer Learning Approach for Network Modeling

Iie Transactions, 2012

Network models have been widely used in many subject areas to characterize the interactions betwe... more Network models have been widely used in many subject areas to characterize the interactions between physical entities. A typical problem is to identify the network for multiple related tasks that share some similarities. In this case, a transfer learning approach that can leverage the knowledge gained during the modeling of one task to help better model another task is highly desirable. This article proposes a transfer learning approach that adopts a Bayesian hierarchical model framework to characterize the relatedness between tasks and additionally uses L 1-regularization to ensure robust learning of the networks with limited sample sizes. A method based on the Expectation–Maximization (EM) algorithm is further developed to learn the networks from data. Simulation studies are performed that demonstrate the superiority of the proposed transfer learning approach over single-task learning that learns the network of each task in isolation. The proposed approach is also applied to identify brain connectivity networks associated with Alzheimer’s Disease (AD) from functional magnetic resonance image data. The findings are consistent with the AD literature.

Research paper thumbnail of Learning Brain Connectivity of Alzheimer's Disease from Neuroimaging Data

Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disea... more Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number of brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88% and 88%, respectively.

Research paper thumbnail of A Comparison of Elementary Schoolchildren's Exposure to Arsenic and Lead

Journal of Environmental Science and Health Part C-environmental Carcinogenesis & Ecotoxicology Reviews, 2008

One hundred fifty seven fifth-grade students (aged 10–12 years) from three elementary schools in ... more One hundred fifty seven fifth-grade students (aged 10–12 years) from three elementary schools in three different towns in Taichung County, Taiwan were chosen as study subjects for the present arsenic and lead exposure study. The three towns—Longgang, Shalach, and Shuntain—are known to be highly, moderately, and lightly (control) polluted by As and Pb, respectively. Spot morning urine samples of students were collected and analyzed for arsenic and lead. The levels of As in the urine of Longgang schoolchildren showed the highest value among the three schools, while those of the control group (Shuntain) had the lowest values. In addition, the levels of Pb in the urine of the schoolchildren in Shuntain were significantly lower than those in Longgang and Shalach, while the levels of Pb in the urine of the schoolchildren in Longgang and Shalach showed no significant difference. Results of daily intake of metals from the different exposure pathways (i.e., ingestion from drinking water, household dust and food, and inhalation from airborne particles) showed that the Longgang area had the highest daily intake of As and Pb among the three areas, while the lowest daily intake of As and Pb occurred in the control area (Shuntain). A significant correlation between the doses of daily intake and urinary concentrations of As (p = 0.002) and Pb (p = 0.020) was observed. This correlation suggests that the increase of unit dose of the daily intake for As resulted in an increase of 0.953 μ g g−1 creatinine of As, whereas the increase of unit dose of the daily intake for Pb led to an increase of 0.053 μ g g−1 creatinine of Pb. These data indicate that the level of As in urine increased about 18 times higher than that of Pb for the same amount of increase in daily intake.

Research paper thumbnail of Increased Levels of 8Hydroxy2′Deoxyguanosine Attributable to Carcinogenic Metal Exposure among Schoolchildren

Environmental Health Perspectives, 2005

Arsenic, chromium, and nickel are reported in several epidemiologic studies to be associated with... more Arsenic, chromium, and nickel are reported in several epidemiologic studies to be associated with lung cancer. However, the health effects of arsenic, chromium, and nickel exposures are equivocal for children. Therefore, we performed a cross-sectional study to investigate possible associations between the internal concentrations of arsenic, chromium, and nickel and the level of oxidative stress to DNA in children. We measured urinary levels of arsenic, chromium, and nickel for 142 nonsmoking children using atomic absorption spectrometry. As a biomarker for oxidative stress, urinary 8-hydroxy-2´-deoxyguanosine (8-OHdG) levels were analyzed with an enzymelinked immunosorbent assay kit. The median urinary 8-OHdG level for our subjects was 11.7 ng/mg creatinine. No obvious relationship between the levels of urinary nickel and 8-OHdG was found. Multiple linear regression analysis showed that children with higher urinary chromium had greater urinary 8-OHdG than did those with lower urinary chromium. Similarly, subjects with higher urinary arsenic had greater urinary 8-OHdG than did those with lower urinary arsenic. Furthermore, children with both high urinary arsenic and high urinary chromium had the highest 8-OHdG levels (mean ± SE, 16.0 ± 1.3; vs. low arsenic/low chromium, p < 0.01) in urine, followed by those with low arsenic/high chromium (13.7 ± 1.6; vs. low arsenic/low chromium, p = 0.25), high arsenic/low chromium (12.9 ± 1.6 vs. low arsenic/low chromium, p = 0.52), and low arsenic/low chromium (11.5 ± 1.3); the trend was significant (p < 0.001). Thus, environmental carcinogenic metal exposure to chromium and arsenic may play an important role in oxidative DNA damage to children.

Research paper thumbnail of In situ synthesis, crystal structures, and luminescence of two new tetrazole complexes

Journal of Coordination Chemistry, 2010

The synthesis, crystal structures, and luminescent properties of two new complexes containing tet... more The synthesis, crystal structures, and luminescent properties of two new complexes containing tetrazolyl ligands are described. Refluxing a mixture of fipronil (fipronil = (±)-5-amino-1-[2,6-dichloro-4-(trifluoromethyl)phenyl]-4-[(trifluoromethyl)sulfinyl]-1H-pyrazole-3-carbonitrile), sodium azide, and CuCl2 in ethanol and water gives complex 1, [M(L)2](H2O)2] ⋅ 2H2O (HL = (±)-5-amino-1-[2,6-dichloro-4-(trifluoromethyl)phenyl]-4-[(trifluoromethyl)sulfinyl]-1H-pyrazole-3-tetrazole, M = Cu). Hydrothermal reaction of fipronil, sodium azide, and Cd(ClO4)2 in the presence of water and ethanol (Demko–Sharpless

Research paper thumbnail of Identification of miRs-143 and -145 that Is Associated with Bone Metastasis of Prostate Cancer and Involved in the Regulation of EMT

PLOS One, 2011

The principal problem arising from prostate cancer (PCa) is its propensity to metastasize to bone... more The principal problem arising from prostate cancer (PCa) is its propensity to metastasize to bone. MicroRNAs (miRNAs) play a crucial role in many tumor metastases. The importance of miRNAs in bone metastasis of PCa has not been elucidated to date. We investigated whether the expression of certain miRNAs was associated with bone metastasis of PCa. We examined the miRNA expression profiles of 6 primary and 7 bone metastatic PCa samples by miRNA microarray analysis. The expression of 5 miRNAs significantly decreased in bone metastasis compared with primary PCa, including miRs-508-5p, -145, -143, -33a and -100. We further examined other samples of 16 primary PCa and 13 bone metastases using real-time PCR analysis. The expressions of miRs-143 and -145 were verified to down-regulate significantly in metastasis samples. By investigating relationship of the levels of miRs-143 and -145 with clinicopathological features of PCa patients, we found down-regulations of miRs-143 and -145 were negatively correlated to bone metastasis, the Gleason score and level of free PSA in primary PCa. Over-expression miR-143 and -145 by retrovirus transfection reduced the ability of migration and invasion in vitro, and tumor development and bone invasion in vivo of PC-3 cells, a human PCa cell line originated from a bone metastatic PCa specimen. Their upregulation also increased E-cadherin expression and reduced fibronectin expression of PC-3 cells which revealed a less invasive morphologic phenotype. These findings indicate that miRs-143 and -145 are associated with bone metastasis of PCa and suggest that they may play important roles in the bone metastasis and be involved in the regulation of EMT Both of them may also be clinically used as novel biomarkers in discriminating different stages of human PCa and predicting bone metastasis.

Research paper thumbnail of A novel three dimensional 3d–4f heterometallic coordination framework with 2, 2′-bipyridine-3-carboxylate and oxalate ligands

Inorganic Chemistry Communications, 2010

One new three-dimensional 3d-4f heterometallic coordination polymer, namely, {[EuZn 2 (imdc) 2 (C... more One new three-dimensional 3d-4f heterometallic coordination polymer, namely, {[EuZn 2 (imdc) 2 (C 2 O 4 ) 0.5 (H 2 O) 4 ]Á2H 2 O} n (1) (H 3 imdc = imidazole-4,5-dicarboxylic acid), has been successfully synthesized by the hydrothermal reactions of Eu 2 O 3 , Zn(NO 3 ) 2 Á6H 2 O, H 3 imdc, H 2 C 2 O 4 Á2H 2 O and H 2 O. Single-crystal X-ray diffraction analysis reveals that complex 1 possesses 3D heterometallic framework containing 2D layer based on L-Zn2-L (L = imdc) helical chains and L-EuZn1-L chains. Complex 1 exhibits unprecedented (3,4)-connected four-nodal topology with Schläfli symbol (6Á7Á8) (6Á7Á9) (6Á7 2 Á8 2 Á10) (7Á8 2 ). Moreover, the photoluminescence property of 1 was investigated in the solid-state at room temperature.

Research paper thumbnail of Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation

Neuroimage, 2010

Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disea... more Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number of brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88% and 88%, respectively.

Research paper thumbnail of A Transfer Learning Approach for Network Modeling

Iie Transactions, 2012

Network models have been widely used in many subject areas to characterize the interactions betwe... more Network models have been widely used in many subject areas to characterize the interactions between physical entities. A typical problem is to identify the network for multiple related tasks that share some similarities. In this case, a transfer learning approach that can leverage the knowledge gained during the modeling of one task to help better model another task is highly desirable. This article proposes a transfer learning approach that adopts a Bayesian hierarchical model framework to characterize the relatedness between tasks and additionally uses L 1-regularization to ensure robust learning of the networks with limited sample sizes. A method based on the Expectation–Maximization (EM) algorithm is further developed to learn the networks from data. Simulation studies are performed that demonstrate the superiority of the proposed transfer learning approach over single-task learning that learns the network of each task in isolation. The proposed approach is also applied to identify brain connectivity networks associated with Alzheimer’s Disease (AD) from functional magnetic resonance image data. The findings are consistent with the AD literature.

Research paper thumbnail of Learning Brain Connectivity of Alzheimer's Disease from Neuroimaging Data

Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disea... more Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number of brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88% and 88%, respectively.

Research paper thumbnail of A Comparison of Elementary Schoolchildren's Exposure to Arsenic and Lead

Journal of Environmental Science and Health Part C-environmental Carcinogenesis & Ecotoxicology Reviews, 2008

One hundred fifty seven fifth-grade students (aged 10–12 years) from three elementary schools in ... more One hundred fifty seven fifth-grade students (aged 10–12 years) from three elementary schools in three different towns in Taichung County, Taiwan were chosen as study subjects for the present arsenic and lead exposure study. The three towns—Longgang, Shalach, and Shuntain—are known to be highly, moderately, and lightly (control) polluted by As and Pb, respectively. Spot morning urine samples of students were collected and analyzed for arsenic and lead. The levels of As in the urine of Longgang schoolchildren showed the highest value among the three schools, while those of the control group (Shuntain) had the lowest values. In addition, the levels of Pb in the urine of the schoolchildren in Shuntain were significantly lower than those in Longgang and Shalach, while the levels of Pb in the urine of the schoolchildren in Longgang and Shalach showed no significant difference. Results of daily intake of metals from the different exposure pathways (i.e., ingestion from drinking water, household dust and food, and inhalation from airborne particles) showed that the Longgang area had the highest daily intake of As and Pb among the three areas, while the lowest daily intake of As and Pb occurred in the control area (Shuntain). A significant correlation between the doses of daily intake and urinary concentrations of As (p = 0.002) and Pb (p = 0.020) was observed. This correlation suggests that the increase of unit dose of the daily intake for As resulted in an increase of 0.953 μ g g−1 creatinine of As, whereas the increase of unit dose of the daily intake for Pb led to an increase of 0.053 μ g g−1 creatinine of Pb. These data indicate that the level of As in urine increased about 18 times higher than that of Pb for the same amount of increase in daily intake.

Research paper thumbnail of Increased Levels of 8Hydroxy2′Deoxyguanosine Attributable to Carcinogenic Metal Exposure among Schoolchildren

Environmental Health Perspectives, 2005

Arsenic, chromium, and nickel are reported in several epidemiologic studies to be associated with... more Arsenic, chromium, and nickel are reported in several epidemiologic studies to be associated with lung cancer. However, the health effects of arsenic, chromium, and nickel exposures are equivocal for children. Therefore, we performed a cross-sectional study to investigate possible associations between the internal concentrations of arsenic, chromium, and nickel and the level of oxidative stress to DNA in children. We measured urinary levels of arsenic, chromium, and nickel for 142 nonsmoking children using atomic absorption spectrometry. As a biomarker for oxidative stress, urinary 8-hydroxy-2´-deoxyguanosine (8-OHdG) levels were analyzed with an enzymelinked immunosorbent assay kit. The median urinary 8-OHdG level for our subjects was 11.7 ng/mg creatinine. No obvious relationship between the levels of urinary nickel and 8-OHdG was found. Multiple linear regression analysis showed that children with higher urinary chromium had greater urinary 8-OHdG than did those with lower urinary chromium. Similarly, subjects with higher urinary arsenic had greater urinary 8-OHdG than did those with lower urinary arsenic. Furthermore, children with both high urinary arsenic and high urinary chromium had the highest 8-OHdG levels (mean ± SE, 16.0 ± 1.3; vs. low arsenic/low chromium, p < 0.01) in urine, followed by those with low arsenic/high chromium (13.7 ± 1.6; vs. low arsenic/low chromium, p = 0.25), high arsenic/low chromium (12.9 ± 1.6 vs. low arsenic/low chromium, p = 0.52), and low arsenic/low chromium (11.5 ± 1.3); the trend was significant (p < 0.001). Thus, environmental carcinogenic metal exposure to chromium and arsenic may play an important role in oxidative DNA damage to children.

Research paper thumbnail of In situ synthesis, crystal structures, and luminescence of two new tetrazole complexes

Journal of Coordination Chemistry, 2010

The synthesis, crystal structures, and luminescent properties of two new complexes containing tet... more The synthesis, crystal structures, and luminescent properties of two new complexes containing tetrazolyl ligands are described. Refluxing a mixture of fipronil (fipronil = (±)-5-amino-1-[2,6-dichloro-4-(trifluoromethyl)phenyl]-4-[(trifluoromethyl)sulfinyl]-1H-pyrazole-3-carbonitrile), sodium azide, and CuCl2 in ethanol and water gives complex 1, [M(L)2](H2O)2] ⋅ 2H2O (HL = (±)-5-amino-1-[2,6-dichloro-4-(trifluoromethyl)phenyl]-4-[(trifluoromethyl)sulfinyl]-1H-pyrazole-3-tetrazole, M = Cu). Hydrothermal reaction of fipronil, sodium azide, and Cd(ClO4)2 in the presence of water and ethanol (Demko–Sharpless

Research paper thumbnail of Identification of miRs-143 and -145 that Is Associated with Bone Metastasis of Prostate Cancer and Involved in the Regulation of EMT

PLOS One, 2011

The principal problem arising from prostate cancer (PCa) is its propensity to metastasize to bone... more The principal problem arising from prostate cancer (PCa) is its propensity to metastasize to bone. MicroRNAs (miRNAs) play a crucial role in many tumor metastases. The importance of miRNAs in bone metastasis of PCa has not been elucidated to date. We investigated whether the expression of certain miRNAs was associated with bone metastasis of PCa. We examined the miRNA expression profiles of 6 primary and 7 bone metastatic PCa samples by miRNA microarray analysis. The expression of 5 miRNAs significantly decreased in bone metastasis compared with primary PCa, including miRs-508-5p, -145, -143, -33a and -100. We further examined other samples of 16 primary PCa and 13 bone metastases using real-time PCR analysis. The expressions of miRs-143 and -145 were verified to down-regulate significantly in metastasis samples. By investigating relationship of the levels of miRs-143 and -145 with clinicopathological features of PCa patients, we found down-regulations of miRs-143 and -145 were negatively correlated to bone metastasis, the Gleason score and level of free PSA in primary PCa. Over-expression miR-143 and -145 by retrovirus transfection reduced the ability of migration and invasion in vitro, and tumor development and bone invasion in vivo of PC-3 cells, a human PCa cell line originated from a bone metastatic PCa specimen. Their upregulation also increased E-cadherin expression and reduced fibronectin expression of PC-3 cells which revealed a less invasive morphologic phenotype. These findings indicate that miRs-143 and -145 are associated with bone metastasis of PCa and suggest that they may play important roles in the bone metastasis and be involved in the regulation of EMT Both of them may also be clinically used as novel biomarkers in discriminating different stages of human PCa and predicting bone metastasis.

Research paper thumbnail of A novel three dimensional 3d–4f heterometallic coordination framework with 2, 2′-bipyridine-3-carboxylate and oxalate ligands

Inorganic Chemistry Communications, 2010

One new three-dimensional 3d-4f heterometallic coordination polymer, namely, {[EuZn 2 (imdc) 2 (C... more One new three-dimensional 3d-4f heterometallic coordination polymer, namely, {[EuZn 2 (imdc) 2 (C 2 O 4 ) 0.5 (H 2 O) 4 ]Á2H 2 O} n (1) (H 3 imdc = imidazole-4,5-dicarboxylic acid), has been successfully synthesized by the hydrothermal reactions of Eu 2 O 3 , Zn(NO 3 ) 2 Á6H 2 O, H 3 imdc, H 2 C 2 O 4 Á2H 2 O and H 2 O. Single-crystal X-ray diffraction analysis reveals that complex 1 possesses 3D heterometallic framework containing 2D layer based on L-Zn2-L (L = imdc) helical chains and L-EuZn1-L chains. Complex 1 exhibits unprecedented (3,4)-connected four-nodal topology with Schläfli symbol (6Á7Á8) (6Á7Á9) (6Á7 2 Á8 2 Á10) (7Á8 2 ). Moreover, the photoluminescence property of 1 was investigated in the solid-state at room temperature.