The use of machine learning/deep learning in PET/CT interpretation to aid in outcome prediction in lymphoma

Russell Thomas Frood

The use of machine learning/deep learning in PET/CT interpretation to aid in outcome prediction in lymphoma

Russell Thomas Frood

Research output: Thesis › Doctoral Thesis

Abstract

Lymphoma is a haematopoietic malignancy consisting of two broad categories: Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL). These categories can be further split into subtypes with classical HL (cHL) and diffuse large B cell lymphoma (DLBCL) being the commonest subtypes. The gold standard imaging modality for staging and response assessment for cHL and DLBCL is 2-deoxy-2-[fluorine-18]fluoro-D-glucose (FDG) positron emission tomography/computed tomography (PET/CT), with patients having a worse prognosis if they do not demonstrate complete metabolic response (CMR). However, approximately 15% of patients will relapse even after CMR. Therefore, being able to identify patients who are likely to relapse it may be possible to stratify treatment early to improve patient outcomes. The aim of this project is to develop and test image derived predictive models based on the baseline PET/CT to risk stratify patients pre-treatment.

Original language	English
Supervisors/Advisors	Scarsbrook, Andrew, Supervisor, External person Frangi, Alejandro F, Supervisor Tsoumpas, Charalampos, Supervisor, External person Gleeson, Fergus V, Supervisor, External person
Place of Publication	[Great Britain]
Publisher	University of Leeds
Publication status	Published - 2022
Externally published	Yes

Keywords

lymphoma
positron emission tomography/computed tomography
outcome prediction
radiomics
machine learning

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

https://etheses.whiterose.ac.uk/32361/

Cite this

@phdthesis{3127ab1e4d424c5a92449a8608bfe5fe,

title = "The use of machine learning/deep learning in PET/CT interpretation to aid in outcome prediction in lymphoma",

abstract = "Lymphoma is a haematopoietic malignancy consisting of two broad categories: Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL). These categories can be further split into subtypes with classical HL (cHL) and diffuse large B cell lymphoma (DLBCL) being the commonest subtypes. The gold standard imaging modality for staging and response assessment for cHL and DLBCL is 2-deoxy-2-[fluorine-18]fluoro-D-glucose (FDG) positron emission tomography/computed tomography (PET/CT), with patients having a worse prognosis if they do not demonstrate complete metabolic response (CMR). However, approximately 15% of patients will relapse even after CMR. Therefore, being able to identify patients who are likely to relapse it may be possible to stratify treatment early to improve patient outcomes. The aim of this project is to develop and test image derived predictive models based on the baseline PET/CT to risk stratify patients pre-treatment.",

keywords = "lymphoma, positron emission tomography/computed tomography, outcome prediction, radiomics, machine learning",

author = "Frood, {Russell Thomas}",

note = "Includes bibliographical references",

year = "2022",

language = "English",

publisher = "University of Leeds",

address = "United Kingdom",

}

TY - BOOK

T1 - The use of machine learning/deep learning in PET/CT interpretation to aid in outcome prediction in lymphoma

AU - Frood, Russell Thomas

N1 - Includes bibliographical references

PY - 2022

Y1 - 2022

N2 - Lymphoma is a haematopoietic malignancy consisting of two broad categories: Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL). These categories can be further split into subtypes with classical HL (cHL) and diffuse large B cell lymphoma (DLBCL) being the commonest subtypes. The gold standard imaging modality for staging and response assessment for cHL and DLBCL is 2-deoxy-2-[fluorine-18]fluoro-D-glucose (FDG) positron emission tomography/computed tomography (PET/CT), with patients having a worse prognosis if they do not demonstrate complete metabolic response (CMR). However, approximately 15% of patients will relapse even after CMR. Therefore, being able to identify patients who are likely to relapse it may be possible to stratify treatment early to improve patient outcomes. The aim of this project is to develop and test image derived predictive models based on the baseline PET/CT to risk stratify patients pre-treatment.

AB - Lymphoma is a haematopoietic malignancy consisting of two broad categories: Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL). These categories can be further split into subtypes with classical HL (cHL) and diffuse large B cell lymphoma (DLBCL) being the commonest subtypes. The gold standard imaging modality for staging and response assessment for cHL and DLBCL is 2-deoxy-2-[fluorine-18]fluoro-D-glucose (FDG) positron emission tomography/computed tomography (PET/CT), with patients having a worse prognosis if they do not demonstrate complete metabolic response (CMR). However, approximately 15% of patients will relapse even after CMR. Therefore, being able to identify patients who are likely to relapse it may be possible to stratify treatment early to improve patient outcomes. The aim of this project is to develop and test image derived predictive models based on the baseline PET/CT to risk stratify patients pre-treatment.

KW - lymphoma

KW - positron emission tomography/computed tomography

KW - outcome prediction

KW - radiomics

KW - machine learning

M3 - Doctoral Thesis

PB - University of Leeds

CY - [Great Britain]

ER -

The use of machine learning/deep learning in PET/CT interpretation to aid in outcome prediction in lymphoma

Abstract

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this