Search tips
Search criteria

Results 1-14 (14)

Clipboard (0)

Select a Filter Below

more »
Year of Publication
1.  Handedness and language learning disability differentially distribute in progressive aphasia variants 
Brain  2013;136(11):3461-3473.
Primary progressive aphasia is a neurodegenerative clinical syndrome that presents in adulthood with an isolated, progressive language disorder. Three main clinical/anatomical variants have been described, each associated with distinctive pathology. A high frequency of neurodevelopmental learning disability in primary progressive aphasia has been reported. Because the disorder is heterogeneous with different patterns of cognitive, anatomical and biological involvement, we sought to identify whether learning disability had a predilection for one or more of the primary progressive aphasia subtypes. We screened the University of California San Francisco Memory and Aging Center's primary progressive aphasia cohort (n = 198) for history of language-related learning disability as well as hand preference, which has associations with learning disability. The study included logopenic (n = 48), non-fluent (n = 54) and semantic (n = 96) variant primary progressive aphasias. We investigated whether the presence of learning disability or non-right-handedness was associated with differential effects on demographic, neuropsychological and neuroimaging features of primary progressive aphasia. We showed that a high frequency of learning disability was present only in the logopenic group (χ2 = 15.17, P < 0.001) and (χ2 = 11.51, P < 0.001) compared with semantic and non-fluent populations. In this group, learning disability was associated with earlier onset of disease, more isolated language symptoms, and more focal pattern of left posterior temporoparietal atrophy. Non-right-handedness was instead over-represented in the semantic group, at nearly twice the prevalence of the general population (χ2 = 6.34, P = 0.01). Within semantic variant primary progressive aphasia the right-handed and non-right-handed cohorts appeared homogeneous on imaging, cognitive profile, and structural analysis of brain symmetry. Lastly, the non-fluent group showed no increase in learning disability or non-right-handedness. Logopenic variant primary progressive aphasia and developmental dyslexia both manifest with phonological disturbances and posterior temporal involvement. Learning disability might confer vulnerability of this network to early-onset, focal Alzheimer’s pathology. Left-handedness has been described as a proxy for atypical brain hemispheric lateralization. As non-right-handedness was increased only in the semantic group, anomalous lateralization mechanisms might instead be related to frontotemporal lobar degeneration with abnormal TARDBP. Taken together, this study suggests that neurodevelopmental signatures impart differential trajectories towards neurodegenerative disease.
PMCID: PMC3808687  PMID: 24056533
Alzheimer’s disease; frontotemporal dementia; dementia aphasia; case control study; risk factors in epidemiology
2.  A Sparse Structure Learning Algorithm for Gaussian Bayesian Network Identification from High-Dimensional Data 
Structure learning of Bayesian Networks (BNs) is an important topic in machine learning. Driven by modern applications in genetics and brain sciences, accurate and efficient learning of large-scale BN structures from high-dimensional data becomes a challenging problem. To tackle this challenge, we propose a Sparse Bayesian Network (SBN) structure learning algorithm that employs a novel formulation involving one L1-norm penalty term to impose sparsity and another penalty term to ensure that the learned BN is a Directed Acyclic Graph (DAG)—a required property of BNs. Through both theoretical analysis and extensive experiments on 11 moderate and large benchmark networks with various sample sizes, we show that SBN leads to improved learning accuracy, scalability, and efficiency as compared with 10 existing popular BN learning algorithms. We apply SBN to a real-world application of brain connectivity modeling for Alzheimer’s disease (AD) and reveal findings that could lead to advancements in AD research.
PMCID: PMC3924722  PMID: 22665720
Bayesian network; machine learning; data mining
3.  A Transfer Learning Approach for Network Modeling 
Networks models have been widely used in many domains to characterize the interacting relationship between physical entities. A typical problem faced is to identify the networks of 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. In this paper, we propose a transfer learning approach, which adopts a Bayesian hierarchical model framework to characterize task relatedness and additionally uses the L1-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, which 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 identification of brain connectivity networks of Alzheimer’s disease (AD) from functional magnetic resonance image (fMRI) data. The findings are consistent with the AD literature.
PMCID: PMC3920601  PMID: 24526804
4.  A prior feature SVM – MRF based method for mouse brain segmentation 
NeuroImage  2011;59(3):2298-2306.
We introduce an automated method, called prior feature Support Vector Machine- Markov Random Field (pSVMRF), to segment three-dimensional mouse brain Magnetic Resonance Microscopy (MRM) images. Our earlier work, extended MRF (eMRF) integrated Support Vector Machine (SVM) and Markov Random Field (MRF) approaches, leading to improved segmentation accuracy; however, the computation of eMRF is very expensive, which may limit its performance on segmentation and robustness. In this study pSVMRF reduces training and testing time for SVM, while boosting segmentation performance. Unlike the eMRF approach, where MR intensity information and location priors are linearly combined, pSVMRF combines this information in a nonlinear fashion, and enhances the discriminative ability of the algorithm. We validate the proposed method using MR imaging of unstained and actively stained mouse brain specimens, and compare segmentation accuracy with two existing methods: eMRF and MRF. C57BL/6 mice are used for training and testing, using cross validation. For formalin fixed C57BL/6 specimens, pSVMRF outperforms both eMRF and MRF. The segmentation accuracy for C57BL/6 brains, stained or not, was similar for larger structures like hippocampus and caudate putamen, (~87%), but increased substantially for smaller regions like susbtantia nigra (from 78.36% to 91.55%), and anterior commissure (from ~50% to ~80%). To test segmentation robustness against increased anatomical variability we add two strains, BXD29 and a transgenic mouse model of Alzheimer’s Disease. Segmentation accuracy for new strains is 80% for hippocampus, and caudate putamen, indicating that pSVMRF is a promising approach for phenotyping mouse models of human brain disorders.
PMCID: PMC3508710  PMID: 21988893
Automated segmentation; Magnetic resonance microscopy; Markov Random Field; Mouse brain; Support Vector Machine
5.  Automatic Monitoring of Localized Skin Dose with Fluoroscopic and Interventional Procedures 
Journal of Digital Imaging  2010;24(4):626-639.
This software tool locates and computes the intensity of radiation skin dose resulting from fluoroscopically guided interventional procedures. It is comprised of multiple modules. Using standardized body specific geometric values, a software module defines a set of male and female patients arbitarily positioned on a fluoroscopy table. Simulated X-ray angiographic (XA) equipment includes XRII and digital detectors with or without bi-plane configurations and left and right facing tables. Skin dose estimates are localized by computing the exposure to each 0.01 × 0.01 m2 on the surface of a patient irradiated by the X-ray beam. Digital Imaging and Communications in Medicine (DICOM) Structured Report Dose data sent to a modular dosimetry database automatically extracts the 11 XA tags necessary for peak skin dose computation. Skin dose calculation software uses these tags (gantry angles, air kerma at the patient entrance reference point, etc.) and applies appropriate corrections of exposure and beam location based on each irradiation event (fluoroscopy and acquistions). A physicist screen records the initial validation of the accuracy, patient and equipment geometry, DICOM compliance, exposure output calibration, backscatter factor, and table and pad attenuation once per system. A technologist screen specifies patient positioning, patient height and weight, and physician user. Peak skin dose is computed and localized; additionally, fluoroscopy duration and kerma area product values are electronically recorded and sent to the XA database. This approach fully addresses current limitations in meeting accreditation criteria, eliminates the need for paper logs at a XA console, and provides a method where automated ALARA montoring is possible including email and pager alerts.
PMCID: PMC3138926  PMID: 20706859
Peak skin dose; sentinal event; DICOM structured report dose; patient entrance reference point; fluoroscopy; interventional radiology; Joint Commission (JC); radiation dose; Digital Imaging and Communications in Medicine (DICOM)
6.  Atypical, slowly progressive behavioral variant frontotemporal dementia associated with C9ORF72 hexanucleotide expansion 
Some patients meeting behavioral variant frontotemporal dementia (bvFTD) diagnostic criteria progress slowly and plateau at mild symptom severity. Such patients have mild neuropsychological and functional impairments, lack characteristic bvFTD brain atrophy, and have thus been referred to as bvFTD “phenocopies” or slowly progressive (bvFTD-SP). The few patients with bvFTD-SP that have been studied at autopsy have found no evidence of FTD pathology, suggesting that bvFTD-SP is neuropathologically distinct from other forms of FTD. Here, we describe two patients with bvFTD-SP with chromosome 9 open reading frame 72 (C9ORF72) hexanucleotide expansions.
Three hundred and eighty-four patients with FTD clinical spectrum and Alzheimer’s disease diagnoses were screened for C9ORF72 expansion. Two bvFTD-SP mutation carriers were identified. Neuropsychological and functional data, as well as brain atrophy patterns assessed using voxel-based morphometry (VBM), were compared with 44 patients with sporadic bvFTD and 85 healthy controls.
Both patients were age 48 at baseline and met possible bvFTD criteria. In the first patient, VBM revealed thalamic and posterior insula atrophy. Over seven years, his neuropsychological performance and brain atrophy remained stable. In the second patient, VBM revealed cortical atrophy with subtle frontal and insular volume loss. Over two years, her neuropsychological and functional scores as well as brain atrophy remained stable.
C9ORF72 mutations can present with a bvFTD-SP phenotype. Some bvFTD-SP patients may have neurodegenerative pathology, and C9ORF72 mutations should be considered in patients with bvFTD-SP and a family history of dementia or motor neuron disease.
PMCID: PMC3388906  PMID: 22399793
C9ORF72; C9FTD/ALS; frontotemporal dementia; genetics; dementia
7.  An Automated DICOM Database Capable of Arbitrary Data Mining (Including Radiation Dose Indicators) for Quality Monitoring 
Journal of Digital Imaging  2010;24(2):223-233.
The U.S. National Press has brought to full public discussion concerns regarding the use of medical radiation, specifically x-ray computed tomography (CT), in diagnosis. A need exists for developing methods whereby assurance is given that all diagnostic medical radiation use is properly prescribed, and all patients’ radiation exposure is monitored. The “DICOM Index Tracker©” (DIT) transparently captures desired digital imaging and communications in medicine (DICOM) tags from CT, nuclear imaging equipment, and other DICOM devices across an enterprise. Its initial use is recording, monitoring, and providing automatic alerts to medical professionals of excursions beyond internally determined trigger action levels of radiation. A flexible knowledge base, aware of equipment in use, enables automatic alerts to system administrators of newly identified equipment models or software versions so that DIT can be adapted to the new equipment or software. A dosimetry module accepts mammography breast organ dose, skin air kerma values from XA modalities, exposure indices from computed radiography, etc. upon receipt. The American Association of Physicists in Medicine recommended a methodology for effective dose calculations which are performed with CT units having DICOM structured dose reports. Web interface reporting is provided for accessing the database in real-time. DIT is DICOM-compliant and, thus, is standardized for international comparisons. Automatic alerts currently in use include: email, cell phone text message, and internal pager text messaging. This system extends the utility of DICOM for standardizing the capturing and computing of radiation dose as well as other quality measures.
PMCID: PMC3056966  PMID: 20824303
Data extraction; medical informatics applications; radiation dose; database management systems; knowledge base
8.  Learning brain connectivity of Alzheimer’s disease by sparse inverse covariance estimation 
NeuroImage  2010;50(3):935-949.
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 o f 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.
PMCID: PMC3068623  PMID: 20079441
Brain connectivity; Sparse inverse covariance; Alzheimer’s; PET; Biomarker
9.  Tools for evaluating team performance in simulation-based training 
Teamwork training constitutes one of the core approaches for moving healthcare systems toward increased levels of quality and safety, and simulation provides a powerful method of delivering this training, especially for face-paced and dynamic specialty areas such as Emergency Medicine. Team performance measurement and evaluation plays an integral role in ensuring that simulation-based training for teams (SBTT) is systematic and effective. However, this component of SBTT systems is overlooked frequently. This article addresses this gap by providing a review and practical introduction to the process of developing and implementing evaluation systems in SBTT. First, an overview of team performance evaluation is provided. Second, best practices for measuring team performance in simulation are reviewed. Third, some of the prominent measurement tools in the literature are summarized and discussed relative to the best practices. Subsequently, implications of the review are discussed for the practice of training teamwork in Emergency Medicine.
PMCID: PMC2966568  PMID: 21063558
Simulation-based team training; simulation; team training; team performance measurement; team evaluation; team performance
10.  Automated Segmentation of Mouse Brain Images Using Extended MRF 
NeuroImage  2009;46(3):717-725.
We introduce an automated segmentation method, extended Markov Random Field (eMRF) to classify 21 neuroanatomical structures of mouse brain based on three dimensional (3D) magnetic resonance imaging (MRI). The image data are multispectral: T2-weighted, proton density-weighted, diffusion x, y and z weighted. Earlier research (Ali et al., 2005) successfully explored the use of MRF for mouse brain segmentation. In this research, we study the use of information generated from Support Vector Machine (SVM) to represent the probabilistic information. Since SVM in general has a stronger discriminative power than the Gaussian likelihood method and is able to handle nonlinear classification problems, integrating SVM into MRF improved the classification accuracy. The eMRF employs the posterior probability distribution obtained from SVM to generate a classification based on the MR intensity. Secondly eMRF introduces a new potential function based on location information. Third, to maximize the classification performance eMRF uses the contribution weights optimally determined for each of the three potential functions: observation, location and contextual functions, which are traditionally equally weighted. We use the voxel overlap percentage and volume difference percentage to evaluate the accuracy of eMRF segmentation and compare the algorithm with three other segmentation methods – mixed ratio sampling SVM (MRS-SVM), atlas-based segmentation and MRF. Validation using classification accuracy indices between automatically segmented and manually traced data shows that eMRF outperforms other methods.
PMCID: PMC2748869  PMID: 19236923
Automated segmentation; Data mining; Magnetic resonance microscopy; Markov Random Field; Mouse brain; Support Vector Machine
11.  Human Immunodeficiency Virus Type 1 Envelope gp120 Induces a Stop Signal and Virological Synapse Formation in Noninfected CD4+ T Cells▿ †  
Journal of Virology  2008;82(19):9445-9457.
Human immunodeficiency virus type 1 (HIV-1)-infected T cells form a virological synapse with noninfected CD4+ T cells in order to efficiently transfer HIV-1 virions from cell to cell. The virological synapse is a specialized cellular junction that is similar in some respects to the immunological synapse involved in T-cell activation and effector functions mediated by the T-cell antigen receptor. The immunological synapse stops T-cell migration to allow a sustained interaction between T-cells and antigen-presenting cells. Here, we have asked whether HIV-1 envelope gp120 presented on a surface to mimic an HIV-1-infected cell also delivers a stop signal and if this is sufficient to induce a virological synapse. We demonstrate that HIV-1 gp120-presenting surfaces arrested the migration of primary activated CD4 T cells that occurs spontaneously in the presence of ICAM-1 and induced the formation of a virological synapse, which was characterized by segregated supramolecular structures with a central cluster of envelope surrounded by a ring of ICAM-1. The virological synapse was formed transiently, with the initiation of migration within 30 min. Thus, HIV-1 gp120-presenting surfaces induce a transient stop signal and supramolecular segregation in noninfected CD4+ T cells.
PMCID: PMC2546991  PMID: 18632854
12.  Nuclear Receptor Interaction Protein (NRIP) expression assay using human tissue microarray and immunohistochemistry technology confirming nuclear localization 
A novel human nuclear receptor interaction protein (NRIP) has recently been discovered by Chen SL et al, which may play a role in enhancing the transcriptional activity of steroid nuclear receptors in prostate (LNCaP) and cervical (C33A) cancer cell lines. However, knowledge about the biological functions and clinical implications of NRIP, is still incomplete. Our aim was to determine the distribution of NRIP expression and to delineate the cell types that express NRIP in various malignant tumors and healthy non-pathological tissues. This information will significantly affect the exploration of its physiological roles in healthy and tumor cells.
By using tissue microarray (TMA) technology and an anti-NRIP monoclonal antibody immunohistochemical (IHC) survey, NRIP expression was examined in 48 types of tumors and in a control group of 48 matched or unmatched healthy non-neoplastic tissues.
Our survey results showed that ten cases were revealed to express the NRIP in six malignancies (esophageal, colon, breast, ovarian, skin, and pancreatic cancers), but not all of these specific tumor types consistently showed positive NRIP expression. Moreover, malignant tumors of the stomach, prostate, liver, lung, kidney, uterine cervix, urinary bladder, lymph node, testis, and tongue revealed no NRIP expression. Among the control group of 48 matched and unmatched non-neoplastic tissues, all of them demonstrated IHC scores less than the cut-off threshold of 3. In addition, ten cores out of thirty-six carcinomatous tissues revealed positive NRIP expression, which indicated that NRIP expression increases significantly in carcinoma tissue cores, comparing to the matched controlled healthy tissues.
This is the first study to use a human TMA and IHC to validate the nuclear localization for this newly identified NRIP expression. In considering the use of NRIP as a potential diagnostic tool for human malignancies survey, it is important to note that NRIP expression carries a sensitivity of only 23%, but has a specificity of 100%. There is also a significant difference in positive NRIP expression between primary carcinomatous tissues and matched controlled healthy tissues. Although further large-scale studies will merit to be conducted to evaluate its role as a potential adjunct for cancer diagnosis, data from this study provides valuable references for the future investigation of the biological functions of NRIP in humans.
PMCID: PMC2683569  PMID: 18673574
13.  Cyclosporin Analogs Inhibit In Vitro Growth of Cryptosporidium parvum 
Cyclosporine and nonimmunosuppressive cyclosporin (CS) analogs were demonstrated to be potent inhibitors of the growth of the intracellular parasite Cryptosporidium parvum in short-term (48-h) in vitro cultures. Fifty-percent inhibitory concentrations (IC50s) were 0.4 μM for SDZ 033-243, 1.0 μM for SDZ PSC-833, and 1.5 μM for cyclosporine. Two other analogs were less effective than cyclosporine: the IC50 of SDZ 205-549 was 5 μM, and that of SDZ 209-313 was 7 μM. These were much lower than the IC50 of 85 μM of paromomycin, a standard positive control for in vitro drug assays for this parasite. In addition, intracellular growth of excysted sporozoites that had been incubated for 1 h in cyclosporine was significantly reduced, suggesting that the drug can inhibit sporozoite invasion. The cellular activities of the CS analogs used have been characterized for mammalian cells and protozoa. The two analogs that were most active in inhibiting C. parvum, SDZ PSC-833 and SDZ 033-243, bind weakly to cyclophilin, a peptidyl proline isomerase which is the primary target of cyclosporine and CS analogs. However, they are potent modifiers of the activity of the P glycoproteins/multidrug resistance (MDR) transporters, members of the ATP-binding cassette (ABC) superfamily. Hence, both cyclophilin and some ABC transporters may be targets for this class of drugs, although drugs that preferentially interact with the latter are more potent. Cyclosporine (0.5 μM) had no significant chemosensitizing activity. That is, it did not significantly increase sensitivity to paromomycin, suggesting that an ABC transporter is not critical in the efflux of this drug. Cyclosporine at concentrations up to 50 μM was not toxic to host Caco-2 cells in the CellTiter 96 assay. The results of this study complement those of studies of the inhibitory effect of cyclosporine and CS analogs on other apicomplexan parasites, Plasmodium falciparum, Plasmodium vivax, and Toxoplasma gondii.
PMCID: PMC105553  PMID: 9559794
14.  Chemical Characterization of a New Surface Antigenic Polysaccharide from a Mutant of Staphylococcus aureus 
Journal of Bacteriology  1971;108(2):874-884.
A mutant of Staphylococcus aureus H was isolated by virtue of its inability to agglutinate with antibodies against teichoic acid of S. aureus. Immunological studies revealed that the mutant, S. aureus T, possessed a new surface antigen in addition to having the antigenic determinant of the wild-type strain, the ribitol teichoic acid. The presence of this additional surface component rendered strain T resistant to staphylococcal typing phages, presumably by masking the phage-receptor sites. The polymer was separated from teichoic acid by chromatography on diethylaminoethyl cellulose and was shown to be composed of two amino sugars, N-acetyl-d-fucosamine and N-acetyl-d-mannosamin uronic acid.
PMCID: PMC247155  PMID: 5001874

Results 1-14 (14)