A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images

Yashbir Singh; William Jons; John E. Eaton; Joseph D. Sobek; Jaidip Jagtap; Gian Marco Conte; Eric G. Fuemmeler; Kuan Zhang; Yujia Wei; Diana Victoria Vera Garcia; Bradley J Erickson

doi:10.1117/12.2607312

A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images

Yashbir Singh, William Jons, John E. Eaton, Joseph D. Sobek, Jaidip Jagtap, Gian Marco Conte, Eric G. Fuemmeler, Kuan Zhang, Yujia Wei, Diana Victoria Vera Garcia, Bradley J Erickson

Bioscience/Biotechnology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

Machine-learning-based solutions need sufficient manually labeled training data to produce accurate predictions, which can hinder their performance for rare diseases with limited data. We show how to use a newly developed algebraic topology-based machine learning method that analyzes the visual pattern of the data to accurately predict hepatic decompensation in patients with Primary Sclerosing Cholangitis. The results demonstrate that the proposed methodology discriminates between Early Decompensation and Not Early groups. We found that the algebraic topology-based machinelearning approach allows us to make accurate predictions from small datasets such as predicting early and not early hepatic decompensation.

Original language	English (US)
Title of host publication	Medical Imaging 2022
Subtitle of host publication	Imaging Informatics for Healthcare, Research, and Applications
Editors	Thomas M. Deserno, Thomas M. Deserno, Brian J. Park
Publisher	SPIE
ISBN (Electronic)	9781510649491
DOIs	https://doi.org/10.1117/12.2607312
State	Published - 2022
Event	Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications - Virtual, Online Duration: Mar 21 2022 → Mar 27 2022

Publication series

Name	Progress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume	12037
ISSN (Print)	1605-7422

Conference

Conference	Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications
City	Virtual, Online
Period	3/21/22 → 3/27/22

Bibliographical note

Publisher Copyright:
© 2022 SPIE.

Keywords

Hepatic Decompensation
Machine learning
Persistent Homology
Topological data Analysis

Publisher link

10.1117/12.2607312

Find It

Find It at U of M Libraries

Cite this

Singh, Y., Jons, W., Eaton, J. E., Sobek, J. D., Jagtap, J., Conte, G. M., Fuemmeler, E. G., Zhang, K., Wei, Y., Garcia, D. V. V., & Erickson, B. J. (2022). A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images. In T. M. Deserno, T. M. Deserno, & B. J. Park (Eds.), Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications Article 120370I (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 12037). SPIE. https://doi.org/10.1117/12.2607312

A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images. / Singh, Yashbir; Jons, William; Eaton, John E. et al.
Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications. ed. / Thomas M. Deserno; Thomas M. Deserno; Brian J. Park. SPIE, 2022. 120370I (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 12037).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Singh, Y, Jons, W, Eaton, JE, Sobek, JD, Jagtap, J, Conte, GM, Fuemmeler, EG, Zhang, K, Wei, Y, Garcia, DVV & Erickson, BJ 2022, A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images. in TM Deserno, TM Deserno & BJ Park (eds), Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications., 120370I, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 12037, SPIE, Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications, Virtual, Online, 3/21/22. https://doi.org/10.1117/12.2607312

Singh Y, Jons W, Eaton JE, Sobek JD, Jagtap J, Conte GM et al. A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images. In Deserno TM, Deserno TM, Park BJ, editors, Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications. SPIE. 2022. 120370I. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). doi: 10.1117/12.2607312

Singh, Yashbir ; Jons, William ; Eaton, John E. et al. / A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images. Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications. editor / Thomas M. Deserno ; Thomas M. Deserno ; Brian J. Park. SPIE, 2022. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).

@inproceedings{e24cbb4e01dd4fcfa1c857643579308f,

title = "A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images",

abstract = "Machine-learning-based solutions need sufficient manually labeled training data to produce accurate predictions, which can hinder their performance for rare diseases with limited data. We show how to use a newly developed algebraic topology-based machine learning method that analyzes the visual pattern of the data to accurately predict hepatic decompensation in patients with Primary Sclerosing Cholangitis. The results demonstrate that the proposed methodology discriminates between Early Decompensation and Not Early groups. We found that the algebraic topology-based machinelearning approach allows us to make accurate predictions from small datasets such as predicting early and not early hepatic decompensation.",

keywords = "Hepatic Decompensation, Machine learning, Persistent Homology, Topological data Analysis",

author = "Yashbir Singh and William Jons and Eaton, {John E.} and Sobek, {Joseph D.} and Jaidip Jagtap and Conte, {Gian Marco} and Fuemmeler, {Eric G.} and Kuan Zhang and Yujia Wei and Garcia, {Diana Victoria Vera} and Erickson, {Bradley J}",

note = "Publisher Copyright: {\textcopyright} 2022 SPIE.; Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications ; Conference date: 21-03-2022 Through 27-03-2022",

year = "2022",

doi = "10.1117/12.2607312",

language = "English (US)",

series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

publisher = "SPIE",

editor = "Deserno, {Thomas M.} and Deserno, {Thomas M.} and Park, {Brian J.}",

booktitle = "Medical Imaging 2022",

address = "United States",

}

TY - GEN

T1 - A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images

AU - Singh, Yashbir

AU - Jons, William

AU - Eaton, John E.

AU - Sobek, Joseph D.

AU - Jagtap, Jaidip

AU - Conte, Gian Marco

AU - Fuemmeler, Eric G.

AU - Zhang, Kuan

AU - Wei, Yujia

AU - Garcia, Diana Victoria Vera

AU - Erickson, Bradley J

PY - 2022

Y1 - 2022

N2 - Machine-learning-based solutions need sufficient manually labeled training data to produce accurate predictions, which can hinder their performance for rare diseases with limited data. We show how to use a newly developed algebraic topology-based machine learning method that analyzes the visual pattern of the data to accurately predict hepatic decompensation in patients with Primary Sclerosing Cholangitis. The results demonstrate that the proposed methodology discriminates between Early Decompensation and Not Early groups. We found that the algebraic topology-based machinelearning approach allows us to make accurate predictions from small datasets such as predicting early and not early hepatic decompensation.

AB - Machine-learning-based solutions need sufficient manually labeled training data to produce accurate predictions, which can hinder their performance for rare diseases with limited data. We show how to use a newly developed algebraic topology-based machine learning method that analyzes the visual pattern of the data to accurately predict hepatic decompensation in patients with Primary Sclerosing Cholangitis. The results demonstrate that the proposed methodology discriminates between Early Decompensation and Not Early groups. We found that the algebraic topology-based machinelearning approach allows us to make accurate predictions from small datasets such as predicting early and not early hepatic decompensation.

KW - Hepatic Decompensation

KW - Machine learning

KW - Persistent Homology

KW - Topological data Analysis

UR - http://www.scopus.com/inward/record.url?scp=85132046239&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85132046239&partnerID=8YFLogxK

U2 - 10.1117/12.2607312

DO - 10.1117/12.2607312

M3 - Conference contribution

AN - SCOPUS:85132046239

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

BT - Medical Imaging 2022

A2 - Deserno, Thomas M.

A2 - Park, Brian J.

PB - SPIE

T2 - Medical Imaging 2022: Imaging Informatics for Healthcare, Research, and Applications

Y2 - 21 March 2022 through 27 March 2022

ER -

A New TDA-based machine learning Classifier Framework for Predicting Hepatic Decompensation from MR Images

Abstract

Publication series

Conference

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this