Showing papers by "Jerome A. Yesavage published in 2017"

Longitudinal measurement and hierarchical classification framework for the prediction of Alzheimer's disease.

Abstract: Accurate prediction of Alzheimer's disease (AD) is important for the early diagnosis and treatment of this condition. Mild cognitive impairment (MCI) is an early stage of AD. Therefore, patients with MCI who are at high risk of fully developing AD should be identified to accurately predict AD. However, the relationship between brain images and AD is difficult to construct because of the complex characteristics of neuroimaging data. To address this problem, we present a longitudinal measurement of MCI brain images and a hierarchical classification method for AD prediction. Longitudinal images obtained from individuals with MCI were investigated to acquire important information on the longitudinal changes, which can be used to classify MCI subjects as either MCI conversion (MCIc) or MCI non-conversion (MCInc) individuals. Moreover, a hierarchical framework was introduced to the classifier to manage high feature dimensionality issues and incorporate spatial information for improving the prediction accuracy. The proposed method was evaluated using 131 patients with MCI (70 MCIc and 61 MCInc) based on MRI scans taken at different time points. Results showed that the proposed method achieved 79.4% accuracy for the classification of MCIc versus MCInc, thereby demonstrating very promising performance for AD prediction.

Cascaded Multi-view Canonical Correlation (CaMCCo) for Early Diagnosis of Alzheimer's Disease via Fusion of Clinical, Imaging and Omic Features.

Abstract: The introduction of mild cognitive impairment (MCI) as a diagnostic category adds to the challenges of diagnosing Alzheimer’s Disease (AD). No single marker has been proven to accurately categorize patients into their respective diagnostic groups. Thus, previous studies have attempted to develop fused predictors of AD and MCI. These studies have two main limitations. Most do not simultaneously consider all diagnostic categories and provide suboptimal fused representations using the same set of modalities for prediction of all classes. In this work, we present a combined framework, cascaded multiview canonical correlation (CaMCCo), for fusion and cascaded classification that incorporates all diagnostic categories and optimizes classification by selectively combining a subset of modalities at each level of the cascade. CaMCCo is evaluated on a data cohort comprising 149 patients for whom neurophysiological, neuroimaging, proteomic and genomic data were available. Results suggest that fusion of select modalities for each classification task outperforms (mean AUC = 0.92) fusion of all modalities (mean AUC = 0.54) and individual modalities (mean AUC = 0.90, 0.53, 0.71, 0.73, 0.62, 0.68). In addition, CaMCCo outperforms all other multi-class classification methods for MCI prediction (PPV: 0.80 vs. 0.67, 0.63).

Altered Microstructural Caudate Integrity in Posttraumatic Stress Disorder but Not Traumatic Brain Injury.

Abstract: Objective Given the high prevalence and comorbidity of combat-related PTSD and TBI in Veterans, it is often difficult to disentangle the contributions of each disorder. Examining these pathologies separately may help to understand the neurobiological basis of memory impairment in PTSD and TBI independently of each other. Thus, we investigated whether a) PTSD and TBI are characterized by subcortical structural abnormalities by examining diffusion tensor imaging (DTI) metrics and volume and b) if these abnormalities were specific to PTSD versus TBI. Method We investigated whether individuals with PTSD or TBI display subcortical structural abnormalities in memory regions by examining DTI metrics and volume of the hippocampus and caudate in three groups of Veterans: Veterans with PTSD, Veterans with TBI, and Veterans with neither PTSD nor TBI (Veteran controls). Results While our results demonstrated no macrostructural differences among the groups in these regions, there were significant alterations in microstructural DTI indices in the caudate for the PTSD group but not the TBI group compared to Veteran controls. Conclusions The result of increased mean, radial, and axial diffusivity, and decreased fractional anisotropy in the caudate in absence of significant volume atrophy in the PTSD group suggests the presence of subtle abnormalities evident only at a microstructural level. The caudate is thought to play a role in the physiopathology of PTSD, and the habit-like behavioral features of the disorder could be due to striatal-dependent habit learning mechanisms. Thus, DTI appears to be a vital tool to investigate subcortical pathology, greatly enhancing the ability to detect subtle brain changes in complex disorders.

Repetitive transcranial magnetic stimulation (rTMS) for treatment-resistant major depression (TRMD) Veteran patients: study protocol for a randomized controlled trial

Abstract: Evaluation of repetitive transcranial magnetic stimulation (rTMS) for treatment-resistant major depression (TRMD) in Veterans offers unique clinical trial challenges. Here we describe a randomized, double-blinded, intent-to-treat, two-arm, superiority parallel design, a multicenter study funded by the Cooperative Studies Program (CSP No. 556) of the US Department of Veterans Affairs. We recruited medical providers with clinical expertise in treating TRMD at nine Veterans Affairs (VA) medical centers as the trial local investigators. We plan to enroll 360 Veterans diagnosed with TRMD at the nine VA medical centers over a 3-year period. We will randomize participants into a double-blinded clinical trial to left prefrontal rTMS treatment or to sham (control) rTMS treatment (180 participants each group) for up to 30 treatment sessions. All participants will meet Diagnostic and statistical manual of mental disorders, 4 th edition (DSM-IV) criteria for major depression and will have failed at least two prior pharmacological interventions. In contrast with other rTMS clinical trials, we will not exclude Veterans with posttraumatic stress disorder (PTSD) or history of substance abuse and we will obtain detailed history regarding these disorders. Furthermore, we will maintain participants on stable anti-depressant medication throughout the trial. We will evaluate all participants on a wide variety of potential predictors of treatment response including cognitive, psychological and functional parameters. The primary dependent measure will be remission rate (Hamilton Rating Scale for Depression (HRSD24) ≤ 10), and secondary analyses will be conducted on other indices. Comparisons between the rTMS and the sham groups will be made at the end of the acute treatment phase to test the primary hypothesis. The unique challenges to performing such a large technically challenging clinical trial with Veterans and potential avenues for improvement of the design in future trials will be described. ClinicalTrials.gov, NCT01191333 . Registered on 26 August 2010. This report is based on the protocol version 4.6 amended in February 2016. All items from the World Health Organization Trial Registration Data Set are listed in Appendix A.

Pattern Discovery in Brain Imaging Genetics via SCCA Modeling with a Generic Non-convex Penalty

Abstract: Brain imaging genetics intends to uncover associations between genetic markers and neuroimaging quantitative traits. Sparse canonical correlation analysis (SCCA) can discover bi-multivariate associations and select relevant features, and is becoming popular in imaging genetic studies. The L1-norm function is not only convex, but also singular at the origin, which is a necessary condition for sparsity. Thus most SCCA methods impose [Formula: see text]-norm onto the individual feature or the structure level of features to pursuit corresponding sparsity. However, the [Formula: see text]-norm penalty over-penalizes large coefficients and may incurs estimation bias. A number of non-convex penalties are proposed to reduce the estimation bias in regression tasks. But using them in SCCA remains largely unexplored. In this paper, we design a unified non-convex SCCA model, based on seven non-convex functions, for unbiased estimation and stable feature selection simultaneously. We also propose an efficient optimization algorithm. The proposed method obtains both higher correlation coefficients and better canonical loading patterns. Specifically, these SCCA methods with non-convex penalties discover a strong association between the APOE e4 rs429358 SNP and the hippocampus region of the brain. They both are Alzheimer's disease related biomarkers, indicating the potential and power of the non-convex methods in brain imaging genetics.

Susceptibility-Based Neuroimaging: Standard Methods, Clinical Applications, and Future Directions.

Abstract: The evaluation of neuropathologies using MRI methods that leverage tissue susceptibility have become standard practice, especially to detect blood products or mineralization. Additionally, emerging MRI techniques have the ability to provide new information based on tissue susceptibility properties in a robust and quantitative manner. This paper discusses these advanced susceptibility imaging techniques and their clinical applications.

Preliminary Investigation of Sex Differences in Procedural Skill Learning in Veterans with Co-Morbidities

Abstract: Sex differences in procedural skill learning have not been well characterized. Skill learning is an important area to explore in clinical settings that involve rehabilitation and deficit remediation, especially for returning Veterans that have a range of co-morbid conditions (traumatic brain injury, posttraumatic stress disorder, and depression) and possess impairments in multiple domains. Sixty-five (55 males, 10 females) Veterans completed two procedural learning tasks and answered self-report questionnaires. Participants’ performance and total learning slope were analyzed to determine sex differences in learning. Our results revealed sex differences in both tasks demonstrating females tend to perform better than males with a large effect size for these mean differences. While females performed better on the procedural learning tasks compared to males, their rate of learning was equivalent. Skill learning is an important requisite for rehabilitation, as skill learning is necessary to perform daily activities in new settings. Ultimately, these results provide insight into skill learning in Veterans with a range of co-morbid conditions and provide support for further investigation of sex differences in procedural learning.