Brain morphometry by probabilistic latent semantic analysis

Authors:
U. Castellani;A. Perina;V. Murino;M. Bellani;G. Rambaldelli;M. Tansella;P. Brambilla
Affiliations:
VIPS lab, University of Verona, Italy;VIPS lab, University of Verona, Italy;VIPS lab, University of Verona, Italy and Istituto Italiano di Tecnologia, Italy;Istituto Italiano di Tecnologia, Italy;Department of Medicine and Public Health, University of Verona, Italy;Department of Medicine and Public Health, University of Verona, Italy;Department of Medicine and Public Health, University of Verona, Italy and ICBN Center, University of Udine and Verona, Italy
Venue:
MICCAI'10 Proceedings of the 13th international conference on Medical image computing and computer-assisted intervention: Part II
Year:
2010

Citing 8
Cited 7

Surface shape and curvature scales

Image and Vision Computing
Unsupervised learning by probabilistic latent semantic analysis

Machine Learning
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Laplace-Beltrami eigenvalues and topological features of eigenfunctions for statistical shape analysis

Computer-Aided Design
A Unified Framework for MR Based Disease Classification

IPMI '09 Proceedings of the 21st International Conference on Information Processing in Medical Imaging
Multivariate High-Dimensional Cortical Folding Analysis, Combining Complexity and Shape, in Neonates with Congenital Heart Disease

IPMI '09 Proceedings of the 21st International Conference on Information Processing in Medical Imaging
Feature-Based Morphometry

MICCAI '09 Proceedings of the 12th International Conference on Medical Image Computing and Computer-Assisted Intervention: Part II
Global medical shape analysis using the Laplace-Beltrami spectrum

MICCAI'07 Proceedings of the 10th international conference on Medical image computing and computer-assisted intervention - Volume Part I

Biologically-aware latent dirichlet allocation (BaLDA) for the classification of expression microarray

PRIB'10 Proceedings of the 5th IAPR international conference on Pattern recognition in bioinformatics
Hybrid generative-discriminative nucleus classification of renal cell carcinoma

SIMBAD'11 Proceedings of the First international conference on Similarity-based pattern recognition
Renal cancer cell classification using generative embeddings and information theoretic kernels

PRIB'11 Proceedings of the 6th IAPR international conference on Pattern recognition in bioinformatics
A comparison on score spaces for expression microarray data classification

PRIB'11 Proceedings of the 6th IAPR international conference on Pattern recognition in bioinformatics
Combining information theoretic kernels with generative embeddings for classification

Neurocomputing
Investigating Topic Models' Capabilities in Expression Microarray Data Classification

IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
Exploiting geometry in counting grids

SIMBAD'13 Proceedings of the Second international conference on Similarity-Based Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

The paper proposes a new shape morphometry approach that combines advanced classification techniques with geometric features to identify morphological abnormalities on the brain surface. Our aim is to improve the classification accuracy in distinguishing between normal subjects and schizophrenic patients. The approach is inspired by natural language processing. Local brain surface geometric patterns are quantized to visual words, and their co-occurrences are encoded as visual topic. To do this, a generative model, the probabilistic Latent Semantic Analysis is learned from quantized shape descriptors (visual words). Finally, we extract from the learned models a generative score, that is used as input of a Support Vector Machine (SVM), defining an hybrid generative/discriminative classification algorithm. An exhaustive experimental section is proposed on a dataset consisting of MRI scans from 64 patients and 60 control subjects. Promising results are reporting by observing accuracies up to 86.13%.