Grant: Alliance Grants
McGill University
$44,732.00
Brilliance:Design and evaluation of vendor agnostic PET image processing tools for Ultra-high Resolution Brain PET imaging
EF 2024-25
Résumé
There is an unmet need for established in vivo biomarkers capable of detecting brain disease patterns as early as possible. Traditionally, positron emission tomography (PET) imaging using PET probes targeted to biological processes is the gold standard for sensitive modelling of fundamental biological parameters associated with brain diseases, in live humans. Ultra high-resolution (UHR) PET imaging systems that provide the highest possible spatial resolution with optimal sensitivity trade-offs are critical for visualizing focal and subtle physiological disease patterns, which usually appear before behavioural changes manifest. Recent innovations in PET detector design and technology have enabled the feasibility of imaging the brain at unprecedented spatial precision (near 1 x 1 x 1 mm voxel volume, reaching the theoretical limit for dedicated brain imaging). However, to leverage the full performance of next generation high resolution brain PET systems, including leading-edge UHR PET systems developed in Quebec and Manitoba, inherent image acquisition, reconstruction, processing, and analysis challenges of UHR PET imaging must be solved.
The small sized detector design of UHR brain PET systems set to maximize resolution, is at the expense of relatively lower photon detection efficiency (i.e., reduced sensitivity), meaning higher noise levels, compared to whole-body PET systems with larger detector sizes. This coupled with increased likelihood for inaccurate reconstruction of measured PET activity from poor correction of PET photons attenuation, noise, and head motion can significantly degrade image quality and affect parameter estimates of UHR brain scans. To establish UHR brain PET imaging as the true benchmark for in vivo molecular imaging, this project will create novel approaches to correct acquired photon attenuation and noise, minimize PET radiation dose, and eliminate the invasiveness of PET tracer kinetic modelling, so frequent repeated mapping of biological parameters in clinical populations, particularly in radiosensitive pediatric populations can be readily performed.
We will develop leading-edge intelligent PET imaging tools built on advancements in machine learning to optimize UHR brain PET imaging and obtain the best image quality at the lowest possible radiation dose to then provide a new way of imaging the brain gently and wisely. Building on collective expertise of the team in PET detector design and engineering, image analysis software development, and PET neuroimaging, we will develop the Brilliance platform to house a collection of high-fidelity UHR PET image analysis research software aimed at clinical translation to acquire, visualize, and quantify molecular processes in the brain. Over the next 3 years, prototype Brilliance software packages will be created and evaluated to 1) optimize ultra-low dose PET acquisition, 2) reconstruct high quality noise-free brain PET images, and 3) provide a convenient one-stop-shop clinical ready tool for neurological evaluations. Based on our recent experience with development of a clinic-ready PET analysis prototype (caliPER) and our ability to rapidly deploy this for brain PET imaging, with access to retrospective PET data and access to high-performance computing infrastructure, this timeline and project is feasible, if NSERC-FRQNT can be secured for recruitment and training of top talent in computational PET neuroimaging.
Programme
Alliance Grants
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
586861-2023
Name nom
Anazodo, Udunna CU
Department département
Montreal Neurological Institute
Organization id
61
Institution établissement
McGill University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2023
Award amount
44732
Program id
ALLRP
Program name en
Alliance Grants
Program name fr
Subventions Alliance
Group en
RESEARCH PARTNERSHIPS
Group fr
PARTENARIATS DE RECHERCHE
Committee code
96
Committee name en
RPP Internal Decision Cttee
Committee name fr
Cté de décision interne - PPR
Area of application code
1211
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Medical and health sciences
Area of application fr
Sciences médicales et de la santé
Research subject code
3114
Research subject group en
Physics
Research subject group fr
Physique
Research subject en
Biological and Medical Physics
Research subject fr
Physique biologique et médicale
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Brilliance:Design and evaluation of vendor agnostic PET image processing tools for Ultra-high Resolution Brain PET imaging
Keywords
attenuation correction; brain imaging; deep learning; magnetic resonance imaging; positron emission tomography (pet)
Application summary
There is an unmet need for established in vivo biomarkers capable of detecting brain disease patterns as early as possible. Traditionally, positron emission tomography (PET) imaging using PET probes targeted to biological processes is the gold standard for sensitive modelling of fundamental biological parameters associated with brain diseases, in live humans. Ultra high-resolution (UHR) PET imaging systems that provide the highest possible spatial resolution with optimal sensitivity trade-offs are critical for visualizing focal and subtle physiological disease patterns, which usually appear before behavioural changes manifest. Recent innovations in PET detector design and technology have enabled the feasibility of imaging the brain at unprecedented spatial precision (near 1 x 1 x 1 mm voxel volume, reaching the theoretical limit for dedicated brain imaging). However, to leverage the full performance of next generation high resolution brain PET systems, including leading-edge UHR PET systems developed in Quebec and Manitoba, inherent image acquisition, reconstruction, processing, and analysis challenges of UHR PET imaging must be solved.
The small sized detector design of UHR brain PET systems set to maximize resolution, is at the expense of relatively lower photon detection efficiency (i.e., reduced sensitivity), meaning higher noise levels, compared to whole-body PET systems with larger detector sizes. This coupled with increased likelihood for inaccurate reconstruction of measured PET activity from poor correction of PET photons attenuation, noise, and head motion can significantly degrade image quality and affect parameter estimates of UHR brain scans. To establish UHR brain PET imaging as the true benchmark for in vivo molecular imaging, this project will create novel approaches to correct acquired photon attenuation and noise, minimize PET radiation dose, and eliminate the invasiveness of PET tracer kinetic modelling, so frequent repeated mapping of biological parameters in clinical populations, particularly in radiosensitive pediatric populations can be readily performed.
We will develop leading-edge intelligent PET imaging tools built on advancements in machine learning to optimize UHR brain PET imaging and obtain the best image quality at the lowest possible radiation dose to then provide a new way of imaging the brain gently and wisely. Building on collective expertise of the team in PET detector design and engineering, image analysis software development, and PET neuroimaging, we will develop the Brilliance platform to house a collection of high-fidelity UHR PET image analysis research software aimed at clinical translation to acquire, visualize, and quantify molecular processes in the brain. Over the next 3 years, prototype Brilliance software packages will be created and evaluated to 1) optimize ultra-low dose PET acquisition, 2) reconstruct high quality noise-free brain PET images, and 3) provide a convenient one-stop-shop clinical ready tool for neurological evaluations. Based on our recent experience with development of a clinic-ready PET analysis prototype (caliPER) and our ability to rapidly deploy this for brain PET imaging, with access to retrospective PET data and access to high-performance computing infrastructure, this timeline and project is feasible, if NSERC-FRQNT can be secured for recruitment and training of top talent in computational PET neuroimaging.