A multi-layer binary model with adaptive metabolite selection for multi-type brain tumour classification

Teddy Zhao; Shivaram Avula; Simon Bailey; Sara Burling; Tim Jaspan; Lesley MacPherson; Dipayan Mitra; Paul S. Morgan; Barry L Pizer; Rui Shen; Martin Wilson; Lara Worthington; Theodoros Arvanitis; Andrew Peet; John Apps

A multi-layer binary model with adaptive metabolite selection for multi-type brain tumour classification

Teddy Zhao, Shivaram Avula, Simon Bailey, Sara Burling, Tim Jaspan, Lesley MacPherson, Dipayan Mitra, Paul S. Morgan, Barry L Pizer, Rui Shen, Martin Wilson, Lara Worthington, Theodoros Arvanitis, Andrew Peet, John Apps

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

109 Downloads (Pure)

Abstract

Motivation: Accurate classification of multi-type brain tumours through in vivo proton magnetic resonance spectroscopy remains a significant challenge. Conventional machine learning classifiers consider all reliably observed metabolites as features and classify all brain tumours simultaneously, but their performance is limited for rare tumour types.
Goal(s): This abstract presents a novel multi-layer classification model, binary adaptive metabolite selection (BAMS), for better identifying rare tumour types.
Approach: BAMS generalises the problem by considering only one specific brain tumour type and selecting significant biomarkers in each layer iteratively and dynamically.
Results: In comparison to classic models, BAMS showed significantly improved diagnostic performance for rare brain tumour types.
Impact: A brain tumour classification method that can only work on main types and cannot determine rare types is unlikely to be useful for clinicians. This abstract introduces BAMS that can significantly improve diagnostic performance for rare brain tumour types

Original language	English
Title of host publication	Proc Intl Magn Reson Med
Publisher	International Society for Magnetic Resonance in Medicine
Publication status	Accepted/In press - 31 Jan 2024

ASJC Scopus subject areas

Radiology Nuclear Medicine and imaging
Artificial Intelligence

Access to Document

ISMRM2024ASubmitted manuscript, 236 KBLicence: None: All rights reserved

Cite this

Zhao, T., Avula, S., Bailey, S., Burling, S., Jaspan, T., MacPherson, L., Mitra, D., Morgan, P. S., Pizer, B. L., Shen, R., Wilson, M., Worthington, L., Arvanitis, T., Peet, A., & Apps, J. (Accepted/In press). A multi-layer binary model with adaptive metabolite selection for multi-type brain tumour classification. In Proc Intl Magn Reson Med Article 6117 International Society for Magnetic Resonance in Medicine.

@inproceedings{c458534c3ac242c99403eada0690aacb,

title = "A multi-layer binary model with adaptive metabolite selection for multi-type brain tumour classification",

abstract = "Motivation: Accurate classification of multi-type brain tumours through in vivo proton magnetic resonance spectroscopy remains a significant challenge. Conventional machine learning classifiers consider all reliably observed metabolites as features and classify all brain tumours simultaneously, but their performance is limited for rare tumour types.Goal(s): This abstract presents a novel multi-layer classification model, binary adaptive metabolite selection (BAMS), for better identifying rare tumour types.Approach: BAMS generalises the problem by considering only one specific brain tumour type and selecting significant biomarkers in each layer iteratively and dynamically.Results: In comparison to classic models, BAMS showed significantly improved diagnostic performance for rare brain tumour types.Impact: A brain tumour classification method that can only work on main types and cannot determine rare types is unlikely to be useful for clinicians. This abstract introduces BAMS that can significantly improve diagnostic performance for rare brain tumour types",

author = "Teddy Zhao and Shivaram Avula and Simon Bailey and Sara Burling and Tim Jaspan and Lesley MacPherson and Dipayan Mitra and Morgan, {Paul S.} and Pizer, {Barry L} and Rui Shen and Martin Wilson and Lara Worthington and Theodoros Arvanitis and Andrew Peet and John Apps",

year = "2024",

month = jan,

day = "31",

language = "English",

booktitle = "Proc Intl Magn Reson Med",

publisher = "International Society for Magnetic Resonance in Medicine",

}

Zhao, T, Avula, S, Bailey, S, Burling, S, Jaspan, T, MacPherson, L, Mitra, D, Morgan, PS, Pizer, BL, Shen, R, Wilson, M, Worthington, L, Arvanitis, T , Peet, A & Apps, J 2024, A multi-layer binary model with adaptive metabolite selection for multi-type brain tumour classification. in Proc Intl Magn Reson Med., 6117, International Society for Magnetic Resonance in Medicine.

TY - GEN

T1 - A multi-layer binary model with adaptive metabolite selection for multi-type brain tumour classification

AU - Zhao, Teddy

AU - Avula, Shivaram

AU - Bailey, Simon

AU - Burling, Sara

AU - Jaspan, Tim

AU - MacPherson, Lesley

AU - Mitra, Dipayan

AU - Morgan, Paul S.

AU - Pizer, Barry L

AU - Shen, Rui

AU - Wilson, Martin

AU - Worthington, Lara

AU - Arvanitis, Theodoros

AU - Peet, Andrew

AU - Apps, John

PY - 2024/1/31

Y1 - 2024/1/31

N2 - Motivation: Accurate classification of multi-type brain tumours through in vivo proton magnetic resonance spectroscopy remains a significant challenge. Conventional machine learning classifiers consider all reliably observed metabolites as features and classify all brain tumours simultaneously, but their performance is limited for rare tumour types.Goal(s): This abstract presents a novel multi-layer classification model, binary adaptive metabolite selection (BAMS), for better identifying rare tumour types.Approach: BAMS generalises the problem by considering only one specific brain tumour type and selecting significant biomarkers in each layer iteratively and dynamically.Results: In comparison to classic models, BAMS showed significantly improved diagnostic performance for rare brain tumour types.Impact: A brain tumour classification method that can only work on main types and cannot determine rare types is unlikely to be useful for clinicians. This abstract introduces BAMS that can significantly improve diagnostic performance for rare brain tumour types

AB - Motivation: Accurate classification of multi-type brain tumours through in vivo proton magnetic resonance spectroscopy remains a significant challenge. Conventional machine learning classifiers consider all reliably observed metabolites as features and classify all brain tumours simultaneously, but their performance is limited for rare tumour types.Goal(s): This abstract presents a novel multi-layer classification model, binary adaptive metabolite selection (BAMS), for better identifying rare tumour types.Approach: BAMS generalises the problem by considering only one specific brain tumour type and selecting significant biomarkers in each layer iteratively and dynamically.Results: In comparison to classic models, BAMS showed significantly improved diagnostic performance for rare brain tumour types.Impact: A brain tumour classification method that can only work on main types and cannot determine rare types is unlikely to be useful for clinicians. This abstract introduces BAMS that can significantly improve diagnostic performance for rare brain tumour types

UR - https://www.ismrm.org/24m/

M3 - Conference contribution

BT - Proc Intl Magn Reson Med

PB - International Society for Magnetic Resonance in Medicine

ER -

A multi-layer binary model with adaptive metabolite selection for multi-type brain tumour classification

Abstract

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this