Interpretation of Brain Morphology in Association to Alzheimer's Disease Dementia Classification Using Graph Convolutional Networks on Triangulated Meshes

Interpretation of Brain Morphology in Association to Alzheimer's Disease Dementia Classification Using Graph Convolutional Networks on Triangulated Meshes. Azcona, E. A., Besson, P., Wu, Y., Punjabi, A., Martersteck, A., Dravid, A., Parrish, T. B., Bandt, S. K., & Katsaggelos, A. K. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), volume 12474 LNCS, pages 95–107, 2020. Springer International Publishing Cham.

Paper doi abstract bibtex

We propose a mesh-based technique to aid in the classification of Alzheimer's disease dementia (ADD) using mesh representations of the cortex and subcortical structures. Deep learning methods for classification tasks that utilize structural neuroimaging often require extensive learning parameters to optimize. Frequently, these approaches for automated medical diagnosis also lack visual interpretability for areas in the brain involved in making a diagnosis. This work: (a) analyzes brain shape using surface information of the cortex and subcortical structures, (b) proposes a residual learning framework for state-of-the-art graph convolutional networks which offer a significant reduction in learnable parameters, and (c) offers visual interpretability of the network via class-specific gradient information that localizes important regions of interest in our inputs. With our proposed method leveraging the use of cortical and subcortical surface information, we outperform other machine learning methods with a 96.35% testing accuracy for the ADD vs. healthy control problem. We confirm the validity of our model by observing its performance in a 25-trial Monte Carlo cross-validation. The generated visualization maps in our study show correspondences with current knowledge regarding the structural localization of pathological changes in the brain associated to dementia of the Alzheimer's type.

@inproceedings{azcona2020interpretation,
abstract = {We propose a mesh-based technique to aid in the classification of Alzheimer's disease dementia (ADD) using mesh representations of the cortex and subcortical structures. Deep learning methods for classification tasks that utilize structural neuroimaging often require extensive learning parameters to optimize. Frequently, these approaches for automated medical diagnosis also lack visual interpretability for areas in the brain involved in making a diagnosis. This work: (a) analyzes brain shape using surface information of the cortex and subcortical structures, (b) proposes a residual learning framework for state-of-the-art graph convolutional networks which offer a significant reduction in learnable parameters, and (c) offers visual interpretability of the network via class-specific gradient information that localizes important regions of interest in our inputs. With our proposed method leveraging the use of cortical and subcortical surface information, we outperform other machine learning methods with a 96.35% testing accuracy for the ADD vs. healthy control problem. We confirm the validity of our model by observing its performance in a 25-trial Monte Carlo cross-validation. The generated visualization maps in our study show correspondences with current knowledge regarding the structural localization of pathological changes in the brain associated to dementia of the Alzheimer's type.},
archivePrefix = {arXiv},
arxivId = {2008.06151},
author = {Azcona, Emanuel A. and Besson, Pierre and Wu, Yunan and Punjabi, Arjun and Martersteck, Adam and Dravid, Amil and Parrish, Todd B. and Bandt, S. Kathleen and Katsaggelos, Aggelos K.},
booktitle = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
doi = {10.1007/978-3-030-61056-2_8},
eprint = {2008.06151},
isbn = {9783030610555},
issn = {16113349},
keywords = {Alzheimer's disease classification,Graph convolutional networks,Neural network interpretability,Triangulated meshes},
organization = {Springer International Publishing Cham},
pages = {95--107},
title = {{Interpretation of Brain Morphology in Association to Alzheimer's Disease Dementia Classification Using Graph Convolutional Networks on Triangulated Meshes}},
url = {https://link.springer.com/10.1007/978-3-030-61056-2_8},
volume = {12474 LNCS},
year = {2020}
}

Downloads: 0

{"_id":"6drPFPwxBtEnxfQi4","bibbaseid":"azcona-besson-wu-punjabi-martersteck-dravid-parrish-bandt-etal-interpretationofbrainmorphologyinassociationtoalzheimersdiseasedementiaclassificationusinggraphconvolutionalnetworksontriangulatedmeshes-2020","author_short":["Azcona, E. A.","Besson, P.","Wu, Y.","Punjabi, A.","Martersteck, A.","Dravid, A.","Parrish, T. B.","Bandt, S. K.","Katsaggelos, A. K."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","abstract":"We propose a mesh-based technique to aid in the classification of Alzheimer's disease dementia (ADD) using mesh representations of the cortex and subcortical structures. Deep learning methods for classification tasks that utilize structural neuroimaging often require extensive learning parameters to optimize. Frequently, these approaches for automated medical diagnosis also lack visual interpretability for areas in the brain involved in making a diagnosis. This work: (a) analyzes brain shape using surface information of the cortex and subcortical structures, (b) proposes a residual learning framework for state-of-the-art graph convolutional networks which offer a significant reduction in learnable parameters, and (c) offers visual interpretability of the network via class-specific gradient information that localizes important regions of interest in our inputs. With our proposed method leveraging the use of cortical and subcortical surface information, we outperform other machine learning methods with a 96.35% testing accuracy for the ADD vs. healthy control problem. We confirm the validity of our model by observing its performance in a 25-trial Monte Carlo cross-validation. The generated visualization maps in our study show correspondences with current knowledge regarding the structural localization of pathological changes in the brain associated to dementia of the Alzheimer's type.","archiveprefix":"arXiv","arxivid":"2008.06151","author":[{"propositions":[],"lastnames":["Azcona"],"firstnames":["Emanuel","A."],"suffixes":[]},{"propositions":[],"lastnames":["Besson"],"firstnames":["Pierre"],"suffixes":[]},{"propositions":[],"lastnames":["Wu"],"firstnames":["Yunan"],"suffixes":[]},{"propositions":[],"lastnames":["Punjabi"],"firstnames":["Arjun"],"suffixes":[]},{"propositions":[],"lastnames":["Martersteck"],"firstnames":["Adam"],"suffixes":[]},{"propositions":[],"lastnames":["Dravid"],"firstnames":["Amil"],"suffixes":[]},{"propositions":[],"lastnames":["Parrish"],"firstnames":["Todd","B."],"suffixes":[]},{"propositions":[],"lastnames":["Bandt"],"firstnames":["S.","Kathleen"],"suffixes":[]},{"propositions":[],"lastnames":["Katsaggelos"],"firstnames":["Aggelos","K."],"suffixes":[]}],"booktitle":"Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)","doi":"10.1007/978-3-030-61056-2_8","eprint":"2008.06151","isbn":"9783030610555","issn":"16113349","keywords":"Alzheimer's disease classification,Graph convolutional networks,Neural network interpretability,Triangulated meshes","organization":"Springer International Publishing Cham","pages":"95–107","title":"Interpretation of Brain Morphology in Association to Alzheimer's Disease Dementia Classification Using Graph Convolutional Networks on Triangulated Meshes","url":"https://link.springer.com/10.1007/978-3-030-61056-2_8","volume":"12474 LNCS","year":"2020","bibtex":"@inproceedings{azcona2020interpretation,\nabstract = {We propose a mesh-based technique to aid in the classification of Alzheimer's disease dementia (ADD) using mesh representations of the cortex and subcortical structures. Deep learning methods for classification tasks that utilize structural neuroimaging often require extensive learning parameters to optimize. Frequently, these approaches for automated medical diagnosis also lack visual interpretability for areas in the brain involved in making a diagnosis. This work: (a) analyzes brain shape using surface information of the cortex and subcortical structures, (b) proposes a residual learning framework for state-of-the-art graph convolutional networks which offer a significant reduction in learnable parameters, and (c) offers visual interpretability of the network via class-specific gradient information that localizes important regions of interest in our inputs. With our proposed method leveraging the use of cortical and subcortical surface information, we outperform other machine learning methods with a 96.35% testing accuracy for the ADD vs. healthy control problem. We confirm the validity of our model by observing its performance in a 25-trial Monte Carlo cross-validation. The generated visualization maps in our study show correspondences with current knowledge regarding the structural localization of pathological changes in the brain associated to dementia of the Alzheimer's type.},\narchivePrefix = {arXiv},\narxivId = {2008.06151},\nauthor = {Azcona, Emanuel A. and Besson, Pierre and Wu, Yunan and Punjabi, Arjun and Martersteck, Adam and Dravid, Amil and Parrish, Todd B. and Bandt, S. Kathleen and Katsaggelos, Aggelos K.},\nbooktitle = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},\ndoi = {10.1007/978-3-030-61056-2_8},\neprint = {2008.06151},\nisbn = {9783030610555},\nissn = {16113349},\nkeywords = {Alzheimer's disease classification,Graph convolutional networks,Neural network interpretability,Triangulated meshes},\norganization = {Springer International Publishing Cham},\npages = {95--107},\ntitle = {{Interpretation of Brain Morphology in Association to Alzheimer's Disease Dementia Classification Using Graph Convolutional Networks on Triangulated Meshes}},\nurl = {https://link.springer.com/10.1007/978-3-030-61056-2_8},\nvolume = {12474 LNCS},\nyear = {2020}\n}\n","author_short":["Azcona, E. A.","Besson, P.","Wu, Y.","Punjabi, A.","Martersteck, A.","Dravid, A.","Parrish, T. B.","Bandt, S. K.","Katsaggelos, A. K."],"key":"azcona2020interpretation","id":"azcona2020interpretation","bibbaseid":"azcona-besson-wu-punjabi-martersteck-dravid-parrish-bandt-etal-interpretationofbrainmorphologyinassociationtoalzheimersdiseasedementiaclassificationusinggraphconvolutionalnetworksontriangulatedmeshes-2020","role":"author","urls":{"Paper":"https://link.springer.com/10.1007/978-3-030-61056-2_8"},"keyword":["Alzheimer's disease classification","Graph convolutional networks","Neural network interpretability","Triangulated meshes"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://sites.northwestern.edu/ivpl/files/2023/06/IVPL_Updated_publications-1.bib","dataSources":["KTWAakbPXLGfYseXn","ePKPjG8C6yvpk4mEK","37Qfzv6wRptYkoSCL","zFPgsTDAW8aDnb5iN","E6Bth2QB5BYjBMZE7","nbnEjsN7MJhurAK9x","PNQZj6FjzoxxJk4Yi","7FpDWDGJ4KgpDiGfB","bod9ms4MQJHuJgPpp","QR9t5P2cLdJuzhfzK","D8k2SxfC5dKNRFgro","7Dwzbxq93HWrJEhT6","qhF8zxmGcJfvtdeAg","fvDEHD49E2ZRwE3fb","H7crv8NWhZup4d4by","DHqokWsryttGh7pJE","vRJd4wNg9HpoZSMHD","sYxQ6pxFgA59JRhxi","w2WahSbYrbcCKBDsC","XasdXLL99y5rygCmq","3gkSihZQRfAD2KBo3","t5XMbyZbtPBo4wBGS","bEpHM2CtrwW2qE8FP","teJzFLHexaz5AQW5z"],"keywords":["alzheimer's disease classification","graph convolutional networks","neural network interpretability","triangulated meshes"],"search_terms":["interpretation","brain","morphology","association","alzheimer","disease","dementia","classification","using","graph","convolutional","networks","triangulated","meshes","azcona","besson","wu","punjabi","martersteck","dravid","parrish","bandt","katsaggelos"],"title":"Interpretation of Brain Morphology in Association to Alzheimer's Disease Dementia Classification Using Graph Convolutional Networks on Triangulated Meshes","year":2020}