Exploring deep parametric embeddings for breast CADx

Andrew R. Jamieson, Rabi Alam, Maryellen L. Giger

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Computer-aided diagnosis (CADx) involves training supervised classifiers using labeled ("truth-known") data. Often, training data consists of high-dimensional feature vectors extracted from medical images. Unfortunately, very large data sets may be required to train robust classifiers for high-dimensional inputs. To mitigate the risk of classifier over-fitting, CADx schemes may employ feature selection or dimension reduction (DR), for example, principal component analysis (PCA). Recently, a number of novel "structure-preserving" DR methods have been proposed1. Such methods are attractive for use in CADx schemes for two main reasons. First, by providing visualization of highdimensional data structure, and second, since DR can be unsupervised or semi-supervised, unlabeled ("truth-unknown") data may be incorporated2. However, the practical application of state-of-the-art DR techniques such as, t-SNE3, to breast CADx were inhibited by the inability to retain a parametric embedding function capable of mapping new input data to the reduced representation. Deep (more than one hidden layer) neural networks can be used to learn such parametric DR embeddings. We explored the feasibility of such methods for use in CADx by conducting a variety of experiments using simulated feature data, including models based on breast CADx features. Specifically, we investigated the unsupervised parametric t-SNE4 (pt-SNE), the supervised deep t-distributed MCML5 (dt-MCML), and hybrid semi-supervised modifications combining the two.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2011
Subtitle of host publicationComputer-Aided Diagnosis
DOIs
StatePublished - May 13 2011
Externally publishedYes
EventMedical Imaging 2011: Computer-Aided Diagnosis - Lake Buena Vista, FL, United States
Duration: Feb 15 2011Feb 17 2011

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume7963
ISSN (Print)1605-7422

Other

OtherMedical Imaging 2011: Computer-Aided Diagnosis
CountryUnited States
CityLake Buena Vista, FL
Period2/15/112/17/11

Fingerprint

breast
embedding
Breast
classifiers
Classifiers
Principal Component Analysis
education
Computer aided diagnosis
data structures
principal components analysis
Principal component analysis
preserving
Data structures
Feature extraction
Visualization
Neural networks
conduction
Experiments
Datasets

Keywords

  • computer-aided diagnosis
  • deep embedding
  • dimension reduction
  • feature-space
  • machine learning
  • semi-supervised learning

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Cite this

Jamieson, A. R., Alam, R., & Giger, M. L. (2011). Exploring deep parametric embeddings for breast CADx. In Medical Imaging 2011: Computer-Aided Diagnosis [79630Y] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 7963). https://doi.org/10.1117/12.878331

Exploring deep parametric embeddings for breast CADx. / Jamieson, Andrew R.; Alam, Rabi; Giger, Maryellen L.

Medical Imaging 2011: Computer-Aided Diagnosis. 2011. 79630Y (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 7963).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jamieson, AR, Alam, R & Giger, ML 2011, Exploring deep parametric embeddings for breast CADx. in Medical Imaging 2011: Computer-Aided Diagnosis., 79630Y, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 7963, Medical Imaging 2011: Computer-Aided Diagnosis, Lake Buena Vista, FL, United States, 2/15/11. https://doi.org/10.1117/12.878331
Jamieson AR, Alam R, Giger ML. Exploring deep parametric embeddings for breast CADx. In Medical Imaging 2011: Computer-Aided Diagnosis. 2011. 79630Y. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). https://doi.org/10.1117/12.878331
Jamieson, Andrew R. ; Alam, Rabi ; Giger, Maryellen L. / Exploring deep parametric embeddings for breast CADx. Medical Imaging 2011: Computer-Aided Diagnosis. 2011. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).
@inproceedings{cf7975df57574ded9f92b8e145c24112,
title = "Exploring deep parametric embeddings for breast CADx",
abstract = "Computer-aided diagnosis (CADx) involves training supervised classifiers using labeled ({"}truth-known{"}) data. Often, training data consists of high-dimensional feature vectors extracted from medical images. Unfortunately, very large data sets may be required to train robust classifiers for high-dimensional inputs. To mitigate the risk of classifier over-fitting, CADx schemes may employ feature selection or dimension reduction (DR), for example, principal component analysis (PCA). Recently, a number of novel {"}structure-preserving{"} DR methods have been proposed1. Such methods are attractive for use in CADx schemes for two main reasons. First, by providing visualization of highdimensional data structure, and second, since DR can be unsupervised or semi-supervised, unlabeled ({"}truth-unknown{"}) data may be incorporated2. However, the practical application of state-of-the-art DR techniques such as, t-SNE3, to breast CADx were inhibited by the inability to retain a parametric embedding function capable of mapping new input data to the reduced representation. Deep (more than one hidden layer) neural networks can be used to learn such parametric DR embeddings. We explored the feasibility of such methods for use in CADx by conducting a variety of experiments using simulated feature data, including models based on breast CADx features. Specifically, we investigated the unsupervised parametric t-SNE4 (pt-SNE), the supervised deep t-distributed MCML5 (dt-MCML), and hybrid semi-supervised modifications combining the two.",
keywords = "computer-aided diagnosis, deep embedding, dimension reduction, feature-space, machine learning, semi-supervised learning",
author = "Jamieson, {Andrew R.} and Rabi Alam and Giger, {Maryellen L.}",
year = "2011",
month = "5",
day = "13",
doi = "10.1117/12.878331",
language = "English (US)",
isbn = "9780819485052",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
booktitle = "Medical Imaging 2011",

}

TY - GEN

T1 - Exploring deep parametric embeddings for breast CADx

AU - Jamieson, Andrew R.

AU - Alam, Rabi

AU - Giger, Maryellen L.

PY - 2011/5/13

Y1 - 2011/5/13

N2 - Computer-aided diagnosis (CADx) involves training supervised classifiers using labeled ("truth-known") data. Often, training data consists of high-dimensional feature vectors extracted from medical images. Unfortunately, very large data sets may be required to train robust classifiers for high-dimensional inputs. To mitigate the risk of classifier over-fitting, CADx schemes may employ feature selection or dimension reduction (DR), for example, principal component analysis (PCA). Recently, a number of novel "structure-preserving" DR methods have been proposed1. Such methods are attractive for use in CADx schemes for two main reasons. First, by providing visualization of highdimensional data structure, and second, since DR can be unsupervised or semi-supervised, unlabeled ("truth-unknown") data may be incorporated2. However, the practical application of state-of-the-art DR techniques such as, t-SNE3, to breast CADx were inhibited by the inability to retain a parametric embedding function capable of mapping new input data to the reduced representation. Deep (more than one hidden layer) neural networks can be used to learn such parametric DR embeddings. We explored the feasibility of such methods for use in CADx by conducting a variety of experiments using simulated feature data, including models based on breast CADx features. Specifically, we investigated the unsupervised parametric t-SNE4 (pt-SNE), the supervised deep t-distributed MCML5 (dt-MCML), and hybrid semi-supervised modifications combining the two.

AB - Computer-aided diagnosis (CADx) involves training supervised classifiers using labeled ("truth-known") data. Often, training data consists of high-dimensional feature vectors extracted from medical images. Unfortunately, very large data sets may be required to train robust classifiers for high-dimensional inputs. To mitigate the risk of classifier over-fitting, CADx schemes may employ feature selection or dimension reduction (DR), for example, principal component analysis (PCA). Recently, a number of novel "structure-preserving" DR methods have been proposed1. Such methods are attractive for use in CADx schemes for two main reasons. First, by providing visualization of highdimensional data structure, and second, since DR can be unsupervised or semi-supervised, unlabeled ("truth-unknown") data may be incorporated2. However, the practical application of state-of-the-art DR techniques such as, t-SNE3, to breast CADx were inhibited by the inability to retain a parametric embedding function capable of mapping new input data to the reduced representation. Deep (more than one hidden layer) neural networks can be used to learn such parametric DR embeddings. We explored the feasibility of such methods for use in CADx by conducting a variety of experiments using simulated feature data, including models based on breast CADx features. Specifically, we investigated the unsupervised parametric t-SNE4 (pt-SNE), the supervised deep t-distributed MCML5 (dt-MCML), and hybrid semi-supervised modifications combining the two.

KW - computer-aided diagnosis

KW - deep embedding

KW - dimension reduction

KW - feature-space

KW - machine learning

KW - semi-supervised learning

UR - http://www.scopus.com/inward/record.url?scp=79955752000&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955752000&partnerID=8YFLogxK

U2 - 10.1117/12.878331

DO - 10.1117/12.878331

M3 - Conference contribution

AN - SCOPUS:79955752000

SN - 9780819485052

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2011

ER -