A Deep Network for Explainable Prediction of Non-Imaging Phenotypes using Anatomical Multi-View Data

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wei, Yuxiang, Chen, Yuqian, Xue, Tengfei, Zekelman, Leo, Makris, Nikos, Rathi, Yogesh, Cai, Weidong, Zhang, Fan, Donnell, Lauren J. O'
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911758105444352
author Wei, Yuxiang
Chen, Yuqian
Xue, Tengfei
Zekelman, Leo
Makris, Nikos
Rathi, Yogesh
Cai, Weidong
Zhang, Fan
Donnell, Lauren J. O'
author_facet Wei, Yuxiang
Chen, Yuqian
Xue, Tengfei
Zekelman, Leo
Makris, Nikos
Rathi, Yogesh
Cai, Weidong
Zhang, Fan
Donnell, Lauren J. O'
contents Large datasets often contain multiple distinct feature sets, or views, that offer complementary information that can be exploited by multi-view learning methods to improve results. We investigate anatomical multi-view data, where each brain anatomical structure is described with multiple feature sets. In particular, we focus on sets of white matter microstructure and connectivity features from diffusion MRI, as well as sets of gray matter area and thickness features from structural MRI. We investigate machine learning methodology that applies multi-view approaches to improve the prediction of non-imaging phenotypes, including demographics (age), motor (strength), and cognition (picture vocabulary). We present an explainable multi-view network (EMV-Net) that can use different anatomical views to improve prediction performance. In this network, each individual anatomical view is processed by a view-specific feature extractor and the extracted information from each view is fused using a learnable weight. This is followed by a wavelet transform-based module to obtain complementary information across views which is then applied to calibrate the view-specific information. Additionally, the calibrator produces an attention-based calibration score to indicate anatomical structures' importance for interpretation.
format Preprint
id arxiv_https___arxiv_org_abs_2401_04579
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Deep Network for Explainable Prediction of Non-Imaging Phenotypes using Anatomical Multi-View Data
Wei, Yuxiang
Chen, Yuqian
Xue, Tengfei
Zekelman, Leo
Makris, Nikos
Rathi, Yogesh
Cai, Weidong
Zhang, Fan
Donnell, Lauren J. O'
Quantitative Methods
Artificial Intelligence
Image and Video Processing
Large datasets often contain multiple distinct feature sets, or views, that offer complementary information that can be exploited by multi-view learning methods to improve results. We investigate anatomical multi-view data, where each brain anatomical structure is described with multiple feature sets. In particular, we focus on sets of white matter microstructure and connectivity features from diffusion MRI, as well as sets of gray matter area and thickness features from structural MRI. We investigate machine learning methodology that applies multi-view approaches to improve the prediction of non-imaging phenotypes, including demographics (age), motor (strength), and cognition (picture vocabulary). We present an explainable multi-view network (EMV-Net) that can use different anatomical views to improve prediction performance. In this network, each individual anatomical view is processed by a view-specific feature extractor and the extracted information from each view is fused using a learnable weight. This is followed by a wavelet transform-based module to obtain complementary information across views which is then applied to calibrate the view-specific information. Additionally, the calibrator produces an attention-based calibration score to indicate anatomical structures' importance for interpretation.
title A Deep Network for Explainable Prediction of Non-Imaging Phenotypes using Anatomical Multi-View Data
topic Quantitative Methods
Artificial Intelligence
Image and Video Processing
url https://arxiv.org/abs/2401.04579