Advanced search in
Research products
arrow_drop_down
Searching FieldsTerms
Project
arrow_drop_down
is
arrow_drop_down
Alzheimers Disease Neuroimaging Initiative (1U01AG024904-01)
Include:
The following results are related to Canada. Are you interested to view more results? Visit OpenAIRE - Explore.
996 Research products, page 1 of 100

  • Canada

10
arrow_drop_down
Date (most recent)
arrow_drop_down
  • Open Access
    Authors: 
    Yi Zhao; Bingkai Wang; Chin‐Fu Liu; Andreia V. Faria; Michael I. Miller; Brian S. Caffo; Xi Luo;
    Publisher: arXiv
    Project: CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    Brain segmentation at different levels is generally represented as hierarchical trees. Brain regional atrophy at specific levels was found to be marginally associated with Alzheimer's disease outcomes. In this study, we propose an L1-type regularization for predictors that follow a hierarchical tree structure. Considering a tree as a directed acyclic graph, we interpret the model parameters from a path analysis perspective. Under this concept, the proposed penalty regulates the total effect of each predictor on the outcome. With regularity conditions, it is shown that under the proposed regularization, the estimator of the model coefficient is consistent in L2-norm and the model selection is also consistent. By applying to a brain structural magnetic resonance imaging dataset acquired from the Alzheimer's Disease Neuroimaging Initiative, the proposed approach identifies brain regions where atrophy in these regions demonstrates the declination in memory. With regularization on the total effects, the findings suggest that the impact of atrophy on memory deficits is localized from small brain regions but at various levels of brain segmentation.

  • Publication . Article . Preprint . Other literature type . 2022
    Open Access
    Authors: 
    Wang, Bingkai; Caffo, Brian S.; Luo, Xi; Liu, Chin-Fu; Faria, Andreia V.; Miller, Michael I.; Zhao, Yi;
    Publisher: Wiley
    Project: NIH | Alzheimers Disease Neuroi... (1U01AG024904-01), CIHR

    A compositional tree refers to a tree structure on a set of random variables where each random variable is a node and composition occurs at each non-leaf node of the tree. As a generalization of compositional data, compositional trees handle more complex relationships among random variables and appear in many disciplines, such as brain imaging, genomics and finance. We consider the problem of sparse regression on data that are associated with a compositional tree and propose a transformation-free tree-based regularized regression method for component selection. The regularization penalty is designed based on the tree structure and encourages a sparse tree representation. We prove that our proposed estimator for regression coefficients is both consistent and model selection consistent. In the simulation study, our method shows higher accuracy than competing methods under different scenarios. By analyzing a brain imaging data set from studies of Alzheimer's disease, our method identifies meaningful associations between memory declination and volume of brain regions that are consistent with current understanding.

  • Open Access English
    Authors: 
    Mingming Liu; Jing Yang; Yushi Liu; Bochao Jia; Yun-Fei Chen; Luna Sun; Shujie Ma;
    Project: NIH | Alzheimers Disease Neuroi... (1U01AG024904-01), CIHR

    Uncovering the heterogeneity in the disease progression of Alzheimer's is a key factor to disease understanding and treatment development, so that interventions can be tailored to target the subgroups that will benefit most from the treatment, which is an important goal of precision medicine. However, in practice, one top methodological challenge hindering the heterogeneity investigation is that the true subgroup membership of each individual is often unknown. In this article, we aim to identify latent subgroups of individuals who share a common disorder progress over time, to predict latent subgroup memberships, and to estimate and infer the heterogeneous trajectories among the subgroups. To achieve these goals, we apply a concave fusion learning method proposed in Ma and Huang (2017) and Ma et al. (2019) to conduct subgroup analysis for longitudinal trajectories of the Alzheimer's disease data. The heterogeneous trajectories are represented by subject-specific unknown functions which are approximated by B-splines. The concave fusion method can simultaneously estimate the spline coefficients and merge them together for the subjects belonging to the same subgroup to automatically identify subgroups and recover the heterogeneous trajectories. The resulting estimator of the disease trajectory of each subgroup is supported by an asymptotic distribution. It provides a sound theoretical basis for further conducting statistical inference in subgroup analysis..

  • Open Access
    Authors: 
    Liu, Mingming; Yang, Jing; Liu, Yushi; Jia, Bochao; Chen, Yun-Fei; Sun, Luna; Ma, Shujie;
    Publisher: Taylor & Francis
    Project: CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    Uncovering the heterogeneity in the disease progression of Alzheimer's is a key factor to disease understanding and treatment development, so that interventions can be tailored to target the subgroups that will benefit most from the treatment, which is an important goal of precision medicine. However, in practice, one top methodological challenge hindering the heterogeneity investigation is that the true subgroup membership of each individual is often unknown. In this article, we aim to identify latent subgroups of individuals who share a common disorder progress over time, to predict latent subgroup memberships, and to estimate and infer the heterogeneous trajectories among the subgroups. To achieve these goals, we apply a concave fusion learning method to conduct subgroup analysis for longitudinal trajectories of the Alzheimer's disease data. The heterogeneous trajectories are represented by subject-specific unknown functions which are approximated by B-splines. The concave fusion method can simultaneously estimate the spline coefficients and merge them together for the subjects belonging to the same subgroup to automatically identify subgroups and recover the heterogeneous trajectories. The resulting estimator of the disease trajectory of each subgroup is supported by an asymptotic distribution. It provides a sound theoretical basis for further conducting statistical inference in subgroup analysis.

  • Publication . Article . Preprint . 2021 . Embargo End Date: 01 Jan 2020
    Open Access
    Authors: 
    Alejandro Schuler; David Walsh; Diana Hall; Jon Walsh; Charles Fisher;
    Publisher: arXiv
    Project: NIH | Alzheimer's Disease Coope... (3U19AG010483-26S4), CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    Abstract Estimating causal effects from randomized experiments is central to clinical research. Reducing the statistical uncertainty in these analyses is an important objective for statisticians. Registries, prior trials, and health records constitute a growing compendium of historical data on patients under standard-of-care that may be exploitable to this end. However, most methods for historical borrowing achieve reductions in variance by sacrificing strict type-I error rate control. Here, we propose a use of historical data that exploits linear covariate adjustment to improve the efficiency of trial analyses without incurring bias. Specifically, we train a prognostic model on the historical data, then estimate the treatment effect using a linear regression while adjusting for the trial subjects’ predicted outcomes (their prognostic scores). We prove that, under certain conditions, this prognostic covariate adjustment procedure attains the minimum variance possible among a large class of estimators. When those conditions are not met, prognostic covariate adjustment is still more efficient than raw covariate adjustment and the gain in efficiency is proportional to a measure of the predictive accuracy of the prognostic model above and beyond the linear relationship with the raw covariates. We demonstrate the approach using simulations and a reanalysis of an Alzheimer’s disease clinical trial and observe meaningful reductions in mean-squared error and the estimated variance. Lastly, we provide a simplified formula for asymptotic variance that enables power calculations that account for these gains. Sample size reductions between 10% and 30% are attainable when using prognostic models that explain a clinically realistic percentage of the outcome variance.

  • Open Access
    Authors: 
    Samar S. M. Elsheikh; Samar S. M. Elsheikh; Emile R. Chimusa; Alzheimer's Disease Neuroimaging Initiative; Nicola J. Mulder; Alessandro Crimi; Alessandro Crimi; Alessandro Crimi;
    Publisher: Frontiers Media SA
    Project: CIHR , EC | Sano (857533), NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    ABSTRACT Networks are present in many aspects of our lives, and networks in neuroscience have recently gained much attention leading to novel representations of brain connectivity. Indeed, there is still room for investigation of the genetic contribution to brain connectivity. The integration of neuroimaging and genetics allows a better understanding of the effects of the genetic variations on brain structural and functional connections, but few studies have successfully investigated the longitudinal association of such a mutual interplay. Nevertheless, several Alzheimer’s disease-associated genetic variants have been identified through omic studies, and the current work uses whole-brain tractography in a longitudinal case-control study design and measures the structural connectivity changes of brain networks to study the neurodegeneration of Alzheimer’s. This is performed by examining the effect of targeted genetic risk factors on local and global brain connectivity. We investigated the degree to which changes in brain connectivity are affected by gene expression. More specifically, we used the most common brain connectivity measures such as efficiency, characteristic path length, betweenness centrality, Louvain modularity and transitivity (a variation of clustering coefficient). Furthermore, we examined the extent to which Clinical Dementia Rating reflects brain connections longitudinally and genetic variation. Here, we show that the expression of PLAU and HFE genes increases the change in betweenness centrality related to the fusiform gyrus and clustering coefficient of cingulum bundle over time, respectively. APP and BLMH gene expression associates with local connectivity. We also show that betweenness centrality has a high contribution to dementia in distinct brain regions. Our findings provide insights into the complex longitudinal interplay between genetics and neuroimaging characteristics and highlight the role of Alzheimer’s genetic risk factors in the estimation of regional brain connection alterations. These regional relationship patterns can be useful for early disease treatment and neurodegeneration prediction.

  • Open Access
    Authors: 
    Martin Dyrba; Moritz Hanzig; Slawek Altenstein; Sebastian Bader; Tommaso Ballarini; Frederic Brosseron; Katharina Buerger; Daniel Cantré; Peter Dechent; Laura Dobisch; +31 more
    Publisher: arXiv
    Project: CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    Background: Although convolutional neural networks (CNN) achieve high diagnostic accuracy for detecting Alzheimer's disease (AD) dementia based on magnetic resonance imaging (MRI) scans, they are not yet applied in clinical routine. One important reason for this is a lack of model comprehensibility. Recently developed visualization methods for deriving CNN relevance maps may help to fill this gap. We investigated whether models with higher accuracy also rely more on discriminative brain regions predefined by prior knowledge. Methods: We trained a CNN for the detection of AD in N=663 T1-weighted MRI scans of patients with dementia and amnestic mild cognitive impairment (MCI) and verified the accuracy of the models via cross-validation and in three independent samples including N=1655 cases. We evaluated the association of relevance scores and hippocampus volume to validate the clinical utility of this approach. To improve model comprehensibility, we implemented an interactive visualization of 3D CNN relevance maps. Results: Across three independent datasets, group separation showed high accuracy for AD dementia vs. controls (AUC$\geq$0.92) and moderate accuracy for MCI vs. controls (AUC$\approx$0.75). Relevance maps indicated that hippocampal atrophy was considered as the most informative factor for AD detection, with additional contributions from atrophy in other cortical and subcortical regions. Relevance scores within the hippocampus were highly correlated with hippocampal volumes (Pearson's r$\approx$-0.86, p<0.001). Conclusion: The relevance maps highlighted atrophy in regions that we had hypothesized a priori. This strengthens the comprehensibility of the CNN models, which were trained in a purely data-driven manner based on the scans and diagnosis labels. 24 pages, 9 figures/tables, supplementary material, source code available on GitHub

  • Publication . Preprint . Article . 2021 . Embargo End Date: 01 Jan 2021
    Open Access
    Authors: 
    Yunyou Huang; Nana Wang; Suqin Tang; Li Ma; Tianshu Hao; Zihan Jiang; Fan Zhang; Guoxin Kang; Xiuxia Miao; Xianglong Guan; +3 more
    Publisher: arXiv
    Project: CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    This paper quantitatively reveals the state-of-the-art and state-of-the-practice AI systems only achieve acceptable performance on the stringent conditions that all categories of subjects are known, which we call closed clinical settings, but fail to work in real-world clinical settings. Compared to the diagnosis task in the closed setting, real-world clinical settings pose severe challenges, and we must treat them differently. We build a clinical AI benchmark named Clinical AIBench to set up real-world clinical settings to facilitate researches. We propose an open, dynamic machine learning framework and develop an AI system named OpenClinicalAI to diagnose diseases in real-world clinical settings. The first versions of Clinical AIBench and OpenClinicalAI target Alzheimer’s disease. In the real-world clinical setting, OpenClinicalAI significantly out-performs the state-of-the-art AI system. In addition, OpenClinicalAI develops personalized diagnosis strategies to avoid unnecessary testing and seamlessly collaborates with clinicians. It is promising to be embedded in the current medical systems to improve medical services.One-Sentence SummaryWe propose a clinical AI benchmark and an open, dynamic machine learning framework to enable AI diagnosis systems to land in real-world clinical settings.

  • Open Access English
    Authors: 
    Gharsallaoui, Mohammed Amine; Rekik, Islem;
    Project: NIH | Alzheimers Disease Neuroi... (1U01AG024904-01), CIHR

    Graph neural networks (GNNs) have witnessed an unprecedented proliferation in tackling several problems in computer vision, computer-aided diagnosis, and related fields. While prior studies have focused on boosting the model accuracy, quantifying the reproducibility of the most discriminative features identified by GNNs is still an intact problem that yields concerns about their reliability in clinical applications in particular. Specifically, the reproducibility of biological markers across clinical datasets and distribution shifts across classes (e.g., healthy and disordered brains) is of paramount importance in revealing the underpinning mechanisms of diseases as well as propelling the development of personalized treatment. Motivated by these issues, we propose, for the first time, reproducibility-based GNN selection (RG-Select), a framework for GNN reproducibility assessment via the quantification of the most discriminative features (i.e., biomarkers) shared between different models. To ascertain the soundness of our framework, the reproducibility assessment embraces variations of different factors such as training strategies and data perturbations. Despite these challenges, our framework successfully yielded replicable conclusions across different training strategies and various clinical datasets. Our findings could thus pave the way for the development of biomarker trustworthiness and reliability assessment methods for computer-aided diagnosis and prognosis tasks. RG-Select code is available on GitHub at https://github.com/basiralab/RG-Select.

  • Publication . Preprint . Article . 2021 . Embargo End Date: 01 Jan 2021
    Open Access
    Authors: 
    Starmans, Martijn P. A.; van der Voort, Sebastian R.; Phil, Thomas; Timbergen, Milea J. M.; Vos, Melissa; Padmos, Guillaume A.; Kessels, Wouter; Hanff, David; Grunhagen, Dirk J.; Verhoef, Cornelis; +36 more
    Publisher: arXiv
    Project: NIH | Alzheimers Disease Neuroi... (1U01AG024904-01), EC | EuCanImage (952103), CIHR

    Radiomics uses quantitative medical imaging features to predict clinical outcomes. Currently, in a new clinical application, finding the optimal radiomics method out of the wide range of available options has to be done manually through a heuristic trial-and-error process. In this study we propose a framework for automatically optimizing the construction of radiomics workflows per application. To this end, we formulate radiomics as a modular workflow and include a large collection of common algorithms for each component. To optimize the workflow per application, we employ automated machine learning using a random search and ensembling. We evaluate our method in twelve different clinical applications, resulting in the following area under the curves: 1) liposarcoma (0.83); 2) desmoid-type fibromatosis (0.82); 3) primary liver tumors (0.80); 4) gastrointestinal stromal tumors (0.77); 5) colorectal liver metastases (0.61); 6) melanoma metastases (0.45); 7) hepatocellular carcinoma (0.75); 8) mesenteric fibrosis (0.80); 9) prostate cancer (0.72); 10) glioma (0.71); 11) Alzheimer's disease (0.87); and 12) head and neck cancer (0.84). We show that our framework has a competitive performance compared human experts, outperforms a radiomics baseline, and performs similar or superior to Bayesian optimization and more advanced ensemble approaches. Concluding, our method fully automatically optimizes the construction of radiomics workflows, thereby streamlining the search for radiomics biomarkers in new applications. To facilitate reproducibility and future research, we publicly release six datasets, the software implementation of our framework, and the code to reproduce this study. Comment: 33 pages, 4 figures, 4 tables, 2 supplementary figures, 3 supplementary table, submitted to Medical Image Analysis; revision;

Advanced search in
Research products
arrow_drop_down
Searching FieldsTerms
Project
arrow_drop_down
is
arrow_drop_down
Alzheimers Disease Neuroimaging Initiative (1U01AG024904-01)
Include:
The following results are related to Canada. Are you interested to view more results? Visit OpenAIRE - Explore.
996 Research products, page 1 of 100
  • Open Access
    Authors: 
    Yi Zhao; Bingkai Wang; Chin‐Fu Liu; Andreia V. Faria; Michael I. Miller; Brian S. Caffo; Xi Luo;
    Publisher: arXiv
    Project: CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    Brain segmentation at different levels is generally represented as hierarchical trees. Brain regional atrophy at specific levels was found to be marginally associated with Alzheimer's disease outcomes. In this study, we propose an L1-type regularization for predictors that follow a hierarchical tree structure. Considering a tree as a directed acyclic graph, we interpret the model parameters from a path analysis perspective. Under this concept, the proposed penalty regulates the total effect of each predictor on the outcome. With regularity conditions, it is shown that under the proposed regularization, the estimator of the model coefficient is consistent in L2-norm and the model selection is also consistent. By applying to a brain structural magnetic resonance imaging dataset acquired from the Alzheimer's Disease Neuroimaging Initiative, the proposed approach identifies brain regions where atrophy in these regions demonstrates the declination in memory. With regularization on the total effects, the findings suggest that the impact of atrophy on memory deficits is localized from small brain regions but at various levels of brain segmentation.

  • Publication . Article . Preprint . Other literature type . 2022
    Open Access
    Authors: 
    Wang, Bingkai; Caffo, Brian S.; Luo, Xi; Liu, Chin-Fu; Faria, Andreia V.; Miller, Michael I.; Zhao, Yi;
    Publisher: Wiley
    Project: NIH | Alzheimers Disease Neuroi... (1U01AG024904-01), CIHR

    A compositional tree refers to a tree structure on a set of random variables where each random variable is a node and composition occurs at each non-leaf node of the tree. As a generalization of compositional data, compositional trees handle more complex relationships among random variables and appear in many disciplines, such as brain imaging, genomics and finance. We consider the problem of sparse regression on data that are associated with a compositional tree and propose a transformation-free tree-based regularized regression method for component selection. The regularization penalty is designed based on the tree structure and encourages a sparse tree representation. We prove that our proposed estimator for regression coefficients is both consistent and model selection consistent. In the simulation study, our method shows higher accuracy than competing methods under different scenarios. By analyzing a brain imaging data set from studies of Alzheimer's disease, our method identifies meaningful associations between memory declination and volume of brain regions that are consistent with current understanding.

  • Open Access English
    Authors: 
    Mingming Liu; Jing Yang; Yushi Liu; Bochao Jia; Yun-Fei Chen; Luna Sun; Shujie Ma;
    Project: NIH | Alzheimers Disease Neuroi... (1U01AG024904-01), CIHR

    Uncovering the heterogeneity in the disease progression of Alzheimer's is a key factor to disease understanding and treatment development, so that interventions can be tailored to target the subgroups that will benefit most from the treatment, which is an important goal of precision medicine. However, in practice, one top methodological challenge hindering the heterogeneity investigation is that the true subgroup membership of each individual is often unknown. In this article, we aim to identify latent subgroups of individuals who share a common disorder progress over time, to predict latent subgroup memberships, and to estimate and infer the heterogeneous trajectories among the subgroups. To achieve these goals, we apply a concave fusion learning method proposed in Ma and Huang (2017) and Ma et al. (2019) to conduct subgroup analysis for longitudinal trajectories of the Alzheimer's disease data. The heterogeneous trajectories are represented by subject-specific unknown functions which are approximated by B-splines. The concave fusion method can simultaneously estimate the spline coefficients and merge them together for the subjects belonging to the same subgroup to automatically identify subgroups and recover the heterogeneous trajectories. The resulting estimator of the disease trajectory of each subgroup is supported by an asymptotic distribution. It provides a sound theoretical basis for further conducting statistical inference in subgroup analysis..

  • Open Access
    Authors: 
    Liu, Mingming; Yang, Jing; Liu, Yushi; Jia, Bochao; Chen, Yun-Fei; Sun, Luna; Ma, Shujie;
    Publisher: Taylor & Francis
    Project: CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    Uncovering the heterogeneity in the disease progression of Alzheimer's is a key factor to disease understanding and treatment development, so that interventions can be tailored to target the subgroups that will benefit most from the treatment, which is an important goal of precision medicine. However, in practice, one top methodological challenge hindering the heterogeneity investigation is that the true subgroup membership of each individual is often unknown. In this article, we aim to identify latent subgroups of individuals who share a common disorder progress over time, to predict latent subgroup memberships, and to estimate and infer the heterogeneous trajectories among the subgroups. To achieve these goals, we apply a concave fusion learning method to conduct subgroup analysis for longitudinal trajectories of the Alzheimer's disease data. The heterogeneous trajectories are represented by subject-specific unknown functions which are approximated by B-splines. The concave fusion method can simultaneously estimate the spline coefficients and merge them together for the subjects belonging to the same subgroup to automatically identify subgroups and recover the heterogeneous trajectories. The resulting estimator of the disease trajectory of each subgroup is supported by an asymptotic distribution. It provides a sound theoretical basis for further conducting statistical inference in subgroup analysis.

  • Publication . Article . Preprint . 2021 . Embargo End Date: 01 Jan 2020
    Open Access
    Authors: 
    Alejandro Schuler; David Walsh; Diana Hall; Jon Walsh; Charles Fisher;
    Publisher: arXiv
    Project: NIH | Alzheimer's Disease Coope... (3U19AG010483-26S4), CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    Abstract Estimating causal effects from randomized experiments is central to clinical research. Reducing the statistical uncertainty in these analyses is an important objective for statisticians. Registries, prior trials, and health records constitute a growing compendium of historical data on patients under standard-of-care that may be exploitable to this end. However, most methods for historical borrowing achieve reductions in variance by sacrificing strict type-I error rate control. Here, we propose a use of historical data that exploits linear covariate adjustment to improve the efficiency of trial analyses without incurring bias. Specifically, we train a prognostic model on the historical data, then estimate the treatment effect using a linear regression while adjusting for the trial subjects’ predicted outcomes (their prognostic scores). We prove that, under certain conditions, this prognostic covariate adjustment procedure attains the minimum variance possible among a large class of estimators. When those conditions are not met, prognostic covariate adjustment is still more efficient than raw covariate adjustment and the gain in efficiency is proportional to a measure of the predictive accuracy of the prognostic model above and beyond the linear relationship with the raw covariates. We demonstrate the approach using simulations and a reanalysis of an Alzheimer’s disease clinical trial and observe meaningful reductions in mean-squared error and the estimated variance. Lastly, we provide a simplified formula for asymptotic variance that enables power calculations that account for these gains. Sample size reductions between 10% and 30% are attainable when using prognostic models that explain a clinically realistic percentage of the outcome variance.

  • Open Access
    Authors: 
    Samar S. M. Elsheikh; Samar S. M. Elsheikh; Emile R. Chimusa; Alzheimer's Disease Neuroimaging Initiative; Nicola J. Mulder; Alessandro Crimi; Alessandro Crimi; Alessandro Crimi;
    Publisher: Frontiers Media SA
    Project: CIHR , EC | Sano (857533), NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    ABSTRACT Networks are present in many aspects of our lives, and networks in neuroscience have recently gained much attention leading to novel representations of brain connectivity. Indeed, there is still room for investigation of the genetic contribution to brain connectivity. The integration of neuroimaging and genetics allows a better understanding of the effects of the genetic variations on brain structural and functional connections, but few studies have successfully investigated the longitudinal association of such a mutual interplay. Nevertheless, several Alzheimer’s disease-associated genetic variants have been identified through omic studies, and the current work uses whole-brain tractography in a longitudinal case-control study design and measures the structural connectivity changes of brain networks to study the neurodegeneration of Alzheimer’s. This is performed by examining the effect of targeted genetic risk factors on local and global brain connectivity. We investigated the degree to which changes in brain connectivity are affected by gene expression. More specifically, we used the most common brain connectivity measures such as efficiency, characteristic path length, betweenness centrality, Louvain modularity and transitivity (a variation of clustering coefficient). Furthermore, we examined the extent to which Clinical Dementia Rating reflects brain connections longitudinally and genetic variation. Here, we show that the expression of PLAU and HFE genes increases the change in betweenness centrality related to the fusiform gyrus and clustering coefficient of cingulum bundle over time, respectively. APP and BLMH gene expression associates with local connectivity. We also show that betweenness centrality has a high contribution to dementia in distinct brain regions. Our findings provide insights into the complex longitudinal interplay between genetics and neuroimaging characteristics and highlight the role of Alzheimer’s genetic risk factors in the estimation of regional brain connection alterations. These regional relationship patterns can be useful for early disease treatment and neurodegeneration prediction.

  • Open Access
    Authors: 
    Martin Dyrba; Moritz Hanzig; Slawek Altenstein; Sebastian Bader; Tommaso Ballarini; Frederic Brosseron; Katharina Buerger; Daniel Cantré; Peter Dechent; Laura Dobisch; +31 more
    Publisher: arXiv
    Project: CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    Background: Although convolutional neural networks (CNN) achieve high diagnostic accuracy for detecting Alzheimer's disease (AD) dementia based on magnetic resonance imaging (MRI) scans, they are not yet applied in clinical routine. One important reason for this is a lack of model comprehensibility. Recently developed visualization methods for deriving CNN relevance maps may help to fill this gap. We investigated whether models with higher accuracy also rely more on discriminative brain regions predefined by prior knowledge. Methods: We trained a CNN for the detection of AD in N=663 T1-weighted MRI scans of patients with dementia and amnestic mild cognitive impairment (MCI) and verified the accuracy of the models via cross-validation and in three independent samples including N=1655 cases. We evaluated the association of relevance scores and hippocampus volume to validate the clinical utility of this approach. To improve model comprehensibility, we implemented an interactive visualization of 3D CNN relevance maps. Results: Across three independent datasets, group separation showed high accuracy for AD dementia vs. controls (AUC$\geq$0.92) and moderate accuracy for MCI vs. controls (AUC$\approx$0.75). Relevance maps indicated that hippocampal atrophy was considered as the most informative factor for AD detection, with additional contributions from atrophy in other cortical and subcortical regions. Relevance scores within the hippocampus were highly correlated with hippocampal volumes (Pearson's r$\approx$-0.86, p<0.001). Conclusion: The relevance maps highlighted atrophy in regions that we had hypothesized a priori. This strengthens the comprehensibility of the CNN models, which were trained in a purely data-driven manner based on the scans and diagnosis labels. 24 pages, 9 figures/tables, supplementary material, source code available on GitHub

  • Publication . Preprint . Article . 2021 . Embargo End Date: 01 Jan 2021
    Open Access
    Authors: 
    Yunyou Huang; Nana Wang; Suqin Tang; Li Ma; Tianshu Hao; Zihan Jiang; Fan Zhang; Guoxin Kang; Xiuxia Miao; Xianglong Guan; +3 more
    Publisher: arXiv
    Project: CIHR , NIH | Alzheimers Disease Neuroi... (1U01AG024904-01)

    This paper quantitatively reveals the state-of-the-art and state-of-the-practice AI systems only achieve acceptable performance on the stringent conditions that all categories of subjects are known, which we call closed clinical settings, but fail to work in real-world clinical settings. Compared to the diagnosis task in the closed setting, real-world clinical settings pose severe challenges, and we must treat them differently. We build a clinical AI benchmark named Clinical AIBench to set up real-world clinical settings to facilitate researches. We propose an open, dynamic machine learning framework and develop an AI system named OpenClinicalAI to diagnose diseases in real-world clinical settings. The first versions of Clinical AIBench and OpenClinicalAI target Alzheimer’s disease. In the real-world clinical setting, OpenClinicalAI significantly out-performs the state-of-the-art AI system. In addition, OpenClinicalAI develops personalized diagnosis strategies to avoid unnecessary testing and seamlessly collaborates with clinicians. It is promising to be embedded in the current medical systems to improve medical services.One-Sentence SummaryWe propose a clinical AI benchmark and an open, dynamic machine learning framework to enable AI diagnosis systems to land in real-world clinical settings.

  • Open Access English
    Authors: 
    Gharsallaoui, Mohammed Amine; Rekik, Islem;
    Project: NIH | Alzheimers Disease Neuroi... (1U01AG024904-01), CIHR

    Graph neural networks (GNNs) have witnessed an unprecedented proliferation in tackling several problems in computer vision, computer-aided diagnosis, and related fields. While prior studies have focused on boosting the model accuracy, quantifying the reproducibility of the most discriminative features identified by GNNs is still an intact problem that yields concerns about their reliability in clinical applications in particular. Specifically, the reproducibility of biological markers across clinical datasets and distribution shifts across classes (e.g., healthy and disordered brains) is of paramount importance in revealing the underpinning mechanisms of diseases as well as propelling the development of personalized treatment. Motivated by these issues, we propose, for the first time, reproducibility-based GNN selection (RG-Select), a framework for GNN reproducibility assessment via the quantification of the most discriminative features (i.e., biomarkers) shared between different models. To ascertain the soundness of our framework, the reproducibility assessment embraces variations of different factors such as training strategies and data perturbations. Despite these challenges, our framework successfully yielded replicable conclusions across different training strategies and various clinical datasets. Our findings could thus pave the way for the development of biomarker trustworthiness and reliability assessment methods for computer-aided diagnosis and prognosis tasks. RG-Select code is available on GitHub at https://github.com/basiralab/RG-Select.

  • Publication . Preprint . Article . 2021 . Embargo End Date: 01 Jan 2021
    Open Access
    Authors: 
    Starmans, Martijn P. A.; van der Voort, Sebastian R.; Phil, Thomas; Timbergen, Milea J. M.; Vos, Melissa; Padmos, Guillaume A.; Kessels, Wouter; Hanff, David; Grunhagen, Dirk J.; Verhoef, Cornelis; +36 more
    Publisher: arXiv
    Project: NIH | Alzheimers Disease Neuroi... (1U01AG024904-01), EC | EuCanImage (952103), CIHR

    Radiomics uses quantitative medical imaging features to predict clinical outcomes. Currently, in a new clinical application, finding the optimal radiomics method out of the wide range of available options has to be done manually through a heuristic trial-and-error process. In this study we propose a framework for automatically optimizing the construction of radiomics workflows per application. To this end, we formulate radiomics as a modular workflow and include a large collection of common algorithms for each component. To optimize the workflow per application, we employ automated machine learning using a random search and ensembling. We evaluate our method in twelve different clinical applications, resulting in the following area under the curves: 1) liposarcoma (0.83); 2) desmoid-type fibromatosis (0.82); 3) primary liver tumors (0.80); 4) gastrointestinal stromal tumors (0.77); 5) colorectal liver metastases (0.61); 6) melanoma metastases (0.45); 7) hepatocellular carcinoma (0.75); 8) mesenteric fibrosis (0.80); 9) prostate cancer (0.72); 10) glioma (0.71); 11) Alzheimer's disease (0.87); and 12) head and neck cancer (0.84). We show that our framework has a competitive performance compared human experts, outperforms a radiomics baseline, and performs similar or superior to Bayesian optimization and more advanced ensemble approaches. Concluding, our method fully automatically optimizes the construction of radiomics workflows, thereby streamlining the search for radiomics biomarkers in new applications. To facilitate reproducibility and future research, we publicly release six datasets, the software implementation of our framework, and the code to reproduce this study. Comment: 33 pages, 4 figures, 4 tables, 2 supplementary figures, 3 supplementary table, submitted to Medical Image Analysis; revision;