Differential dementia diagnosis on incomplete data with latent trees

Christian Ledig, Sebastian Kaltwang, Antti Tolonen, Juha Koikkalainen, Philip Scheltens, Frederik Barkhof, Hanneke Rhodius-Meester, Betty Tijms, Afina W. Lemstra, Wiesje van der Flier, Jyrki Lötjönen, Daniel Rueckert

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

Abstract

Incomplete patient data is a substantial problem that is not sufficiently addressed in current clinical research. Many published methods assume both completeness and validity of study data. However,this assumption is often violated as individual features might be unavailable due to missing patient examination or distorted/wrong due to inaccurate measurements or human error. In this work we propose to use the Latent Tree (LT) generative model to address current limitations due to missing data. We show on 491 subjects of a challenging dementia dataset that LT feature estimation is more robust towards incomplete data as compared to mean or Gaussian Mixture Model imputation and has a synergistic effect when combined with common classifiers (we use SVM as example). We show that LTs allow the inclusion of incomplete samples into classifier training. Using LTs,we obtain a balanced accuracy of 62% for the classification of all patients into five distinct dementia types even though 20% of the features are missing in both training and testing data (68% on complete data). Further,we confirm the potential of LTs to detect outlier samples within the dataset.
Original languageEnglish
Title of host publicationMedical Image Computing and Computer-Assisted Intervention
Subtitle of host publicationMICCAI 2016
PublisherSpringer
Pages44-52
ISBN (Electronic)978-3-319-46723-8
ISBN (Print)978-3-319-46722-1
DOIs
Publication statusPublished - 2 Oct 2016
MoE publication typeA4 Article in a conference publication
Event19th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016 - Athens, Greece
Duration: 17 Oct 201621 Oct 2016

Publication series

NameLecture Notes in Computer Science LNCS
Volume9901

Conference

Conference19th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016
Abbreviated titleMICCAI 2016
CountryGreece
CityAthens
Period17/10/1621/10/16

Fingerprint

Classifiers
Testing

Keywords

  • Latent Trees
  • differential diagnosis
  • dementia
  • incomplete data

Cite this

Ledig, C., Kaltwang, S., Tolonen, A., Koikkalainen, J., Scheltens, P., Barkhof, F., ... Rueckert, D. (2016). Differential dementia diagnosis on incomplete data with latent trees. In Medical Image Computing and Computer-Assisted Intervention: MICCAI 2016 (pp. 44-52). Springer. Lecture Notes in Computer Science, Vol.. 9901 https://doi.org/10.1007/978-3-319-46723-8_6
Ledig, Christian ; Kaltwang, Sebastian ; Tolonen, Antti ; Koikkalainen, Juha ; Scheltens, Philip ; Barkhof, Frederik ; Rhodius-Meester, Hanneke ; Tijms, Betty ; Lemstra, Afina W. ; van der Flier, Wiesje ; Lötjönen, Jyrki ; Rueckert, Daniel. / Differential dementia diagnosis on incomplete data with latent trees. Medical Image Computing and Computer-Assisted Intervention: MICCAI 2016. Springer, 2016. pp. 44-52 (Lecture Notes in Computer Science, Vol. 9901).
@inproceedings{74727d47a3a347a1a4bd09481ad2d07b,
title = "Differential dementia diagnosis on incomplete data with latent trees",
abstract = "Incomplete patient data is a substantial problem that is not sufficiently addressed in current clinical research. Many published methods assume both completeness and validity of study data. However,this assumption is often violated as individual features might be unavailable due to missing patient examination or distorted/wrong due to inaccurate measurements or human error. In this work we propose to use the Latent Tree (LT) generative model to address current limitations due to missing data. We show on 491 subjects of a challenging dementia dataset that LT feature estimation is more robust towards incomplete data as compared to mean or Gaussian Mixture Model imputation and has a synergistic effect when combined with common classifiers (we use SVM as example). We show that LTs allow the inclusion of incomplete samples into classifier training. Using LTs,we obtain a balanced accuracy of 62{\%} for the classification of all patients into five distinct dementia types even though 20{\%} of the features are missing in both training and testing data (68{\%} on complete data). Further,we confirm the potential of LTs to detect outlier samples within the dataset.",
keywords = "Latent Trees, differential diagnosis, dementia, incomplete data",
author = "Christian Ledig and Sebastian Kaltwang and Antti Tolonen and Juha Koikkalainen and Philip Scheltens and Frederik Barkhof and Hanneke Rhodius-Meester and Betty Tijms and Lemstra, {Afina W.} and {van der Flier}, Wiesje and Jyrki L{\"o}tj{\"o}nen and Daniel Rueckert",
year = "2016",
month = "10",
day = "2",
doi = "10.1007/978-3-319-46723-8_6",
language = "English",
isbn = "978-3-319-46722-1",
series = "Lecture Notes in Computer Science LNCS",
publisher = "Springer",
pages = "44--52",
booktitle = "Medical Image Computing and Computer-Assisted Intervention",
address = "Germany",

}

Ledig, C, Kaltwang, S, Tolonen, A, Koikkalainen, J, Scheltens, P, Barkhof, F, Rhodius-Meester, H, Tijms, B, Lemstra, AW, van der Flier, W, Lötjönen, J & Rueckert, D 2016, Differential dementia diagnosis on incomplete data with latent trees. in Medical Image Computing and Computer-Assisted Intervention: MICCAI 2016. Springer, Lecture Notes in Computer Science, vol. 9901, pp. 44-52, 19th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016, Athens, Greece, 17/10/16. https://doi.org/10.1007/978-3-319-46723-8_6

Differential dementia diagnosis on incomplete data with latent trees. / Ledig, Christian; Kaltwang, Sebastian; Tolonen, Antti; Koikkalainen, Juha; Scheltens, Philip; Barkhof, Frederik; Rhodius-Meester, Hanneke; Tijms, Betty; Lemstra, Afina W.; van der Flier, Wiesje; Lötjönen, Jyrki; Rueckert, Daniel.

Medical Image Computing and Computer-Assisted Intervention: MICCAI 2016. Springer, 2016. p. 44-52 (Lecture Notes in Computer Science, Vol. 9901).

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - Differential dementia diagnosis on incomplete data with latent trees

AU - Ledig, Christian

AU - Kaltwang, Sebastian

AU - Tolonen, Antti

AU - Koikkalainen, Juha

AU - Scheltens, Philip

AU - Barkhof, Frederik

AU - Rhodius-Meester, Hanneke

AU - Tijms, Betty

AU - Lemstra, Afina W.

AU - van der Flier, Wiesje

AU - Lötjönen, Jyrki

AU - Rueckert, Daniel

PY - 2016/10/2

Y1 - 2016/10/2

N2 - Incomplete patient data is a substantial problem that is not sufficiently addressed in current clinical research. Many published methods assume both completeness and validity of study data. However,this assumption is often violated as individual features might be unavailable due to missing patient examination or distorted/wrong due to inaccurate measurements or human error. In this work we propose to use the Latent Tree (LT) generative model to address current limitations due to missing data. We show on 491 subjects of a challenging dementia dataset that LT feature estimation is more robust towards incomplete data as compared to mean or Gaussian Mixture Model imputation and has a synergistic effect when combined with common classifiers (we use SVM as example). We show that LTs allow the inclusion of incomplete samples into classifier training. Using LTs,we obtain a balanced accuracy of 62% for the classification of all patients into five distinct dementia types even though 20% of the features are missing in both training and testing data (68% on complete data). Further,we confirm the potential of LTs to detect outlier samples within the dataset.

AB - Incomplete patient data is a substantial problem that is not sufficiently addressed in current clinical research. Many published methods assume both completeness and validity of study data. However,this assumption is often violated as individual features might be unavailable due to missing patient examination or distorted/wrong due to inaccurate measurements or human error. In this work we propose to use the Latent Tree (LT) generative model to address current limitations due to missing data. We show on 491 subjects of a challenging dementia dataset that LT feature estimation is more robust towards incomplete data as compared to mean or Gaussian Mixture Model imputation and has a synergistic effect when combined with common classifiers (we use SVM as example). We show that LTs allow the inclusion of incomplete samples into classifier training. Using LTs,we obtain a balanced accuracy of 62% for the classification of all patients into five distinct dementia types even though 20% of the features are missing in both training and testing data (68% on complete data). Further,we confirm the potential of LTs to detect outlier samples within the dataset.

KW - Latent Trees

KW - differential diagnosis

KW - dementia

KW - incomplete data

U2 - 10.1007/978-3-319-46723-8_6

DO - 10.1007/978-3-319-46723-8_6

M3 - Conference article in proceedings

SN - 978-3-319-46722-1

T3 - Lecture Notes in Computer Science LNCS

SP - 44

EP - 52

BT - Medical Image Computing and Computer-Assisted Intervention

PB - Springer

ER -

Ledig C, Kaltwang S, Tolonen A, Koikkalainen J, Scheltens P, Barkhof F et al. Differential dementia diagnosis on incomplete data with latent trees. In Medical Image Computing and Computer-Assisted Intervention: MICCAI 2016. Springer. 2016. p. 44-52. (Lecture Notes in Computer Science, Vol. 9901). https://doi.org/10.1007/978-3-319-46723-8_6