Grant: Doctoral Research Award: Canada Graduate Scholarships
McGill University
$105,000.00
Development and validation of a multi-input neural network to predict outcomes following radiation therapy of soft tissue sarcoma
FY 2022-23
Summary
Soft tissue sarcoma (STS) accounts for 1% of adult malignancies but comprises over 100 distinct subtypes with varying genetic translocations, clinical characteristics, and disease prognoses. Outcomes have plateaued over the past 3 decades and new innovations in matching patients to optimal treatments are needed to further improve outcomes. Neither chemotherapy nor immunotherapy have a clear role in first-line adjuvant treatment of STS. It is thus imperative to more precisely define the patient population who will benefit most from radiation therapy (RT), which remains the adjuvant treatment option most suitable for STS.
We propose to use a multi-input deep neural network model for generating predictions regarding the effectiveness of RT. We will include STS patients treated with surgery and radiation therapy at MSKCC from 2002-2020. Our neural network will analyze patient demographic and clinical variables as well as preoperative MRI data, to predict three important outcomes: 1) overall survival, 2) risk of local recurrence, 3) risk of adverse events. We will collaborate closely and be supported by MSKCC's AI/ML and NextGen analytics team to develop and scale the neural network model.
Advances in accurately predicting any of these outcomes would provide valuable information to both physcians and patients, and help move the care of soft tissue sarcoma patients into an era of individualized precision medicine. Once demonstrated to be feasible, we will leverage the results and insights from this study to commence a prospective multi-institutional sarcoma database.
Department
Program
Doctoral Research Award: Canada Graduate Scholarships
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
475836
Name
Bozzo Anthony
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Cancer Research
Theme2
Clinical
Prcname2
Doctoral Research Awards - B
Projecttitle
Development and validation of a multi-input neural network to predict outcomes following radiation therapy of soft tissue sarcoma
Programtype2
Studentship Programs
Orgnameinp2
McGill University
Cihramount2
$105,000
Cihrequipment2
$0
Programname2
Doctoral Research Award: Canada Graduate Scholarships
Keyworddelim
Machine Learning; Neural Network; Orthopedic Oncology; Radiation Therapy; Sarcoma; Soft Tissue Sarcoma
Abstract
Soft tissue sarcoma (STS) accounts for 1% of adult malignancies but comprises over 100 distinct subtypes with varying genetic translocations, clinical characteristics, and disease prognoses. Outcomes have plateaued over the past 3 decades and new innovations in matching patients to optimal treatments are needed to further improve outcomes. Neither chemotherapy nor immunotherapy have a clear role in first-line adjuvant treatment of STS. It is thus imperative to more precisely define the patient population who will benefit most from radiation therapy (RT), which remains the adjuvant treatment option most suitable for STS.
We propose to use a multi-input deep neural network model for generating predictions regarding the effectiveness of RT. We will include STS patients treated with surgery and radiation therapy at MSKCC from 2002-2020. Our neural network will analyze patient demographic and clinical variables as well as preoperative MRI data, to predict three important outcomes: 1) overall survival, 2) risk of local recurrence, 3) risk of adverse events. We will collaborate closely and be supported by MSKCC's AI/ML and NextGen analytics team to develop and scale the neural network model.
Advances in accurately predicting any of these outcomes would provide valuable information to both physcians and patients, and help move the care of soft tissue sarcoma patients into an era of individualized precision medicine. Once demonstrated to be feasible, we will leverage the results and insights from this study to commence a prospective multi-institutional sarcoma database.
Competitiondate
202211
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Bozzo, Anthony
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $105,000~ $0~