Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Project Grant - PA: Pandemic Preparedness and Health Emergencies Research

University of Ottawa

$100,000.00

Accelerating and standardizing assessment of acute lung injury using an artificial intelligence based approach

EF 2023-24

Résumé
Acute respiratory distress syndrome is a life-threatening lung injury that results from various causes including trauma, infections (like COVID-19), or burns. In the laboratory, animal models are used to investigate lung injuries. The most important outcome used in these animals is 'histology'. Histology is a method where we examine lung samples under a microscope to see the injury. These images are then analyzed by experts to quantify how much lung injury has occurred. Such analysis is time-consuming, expensive, and prone to mistakes. This can contribute to errors in our understanding of lung injury. An automated method is needed to overcome these limitations. In preparation for this proposal, we performed a comprehensive literature review that revealed there is currently no artificial intelligence (AI) solution available for this purpose. Using AI, we aim to create and validate a novel platform that can accurately analyze histology images of lung injury in animal models. We have assembled an international team of lung injury experts to support this project. They will provide histology samples from their labs for us to develop and validate the platform. We will train an AI model using Python, a widely used programming language. We have collected histopathology images and have annotated them for lung injury. The AI model will learn to identify and measure the severity of lung injury from these images. To ensure precision and dependability, we will evaluate the model's performance and make adjustments. We will also validate our AI model using a separate set of images that have been independently quantified by researchers. Validation will provide an assessment of our AI model's performance and identify areas for refinement. Our work will establish a standardized and reliable AI platform for quantifying lung injury in animal models. Automation of histological assessment using artificial intelligence has the potential to accelerate lung injury studies significantly.
Programme
Project Grant - PA: Pandemic Preparedness and Health Emergencies Research
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database

Tel que publié

Id
504825
Name
Mer Arvind
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Circulatory and Respiratory Health
Theme2
Biomedical
Prcname2
Respiratory System
Projecttitle
Accelerating and standardizing assessment of acute lung injury using an artificial intelligence based approach
Programtype2
Operating Grants
Orgnameinp2
University of Ottawa
Cihramount2
$100,000
Cihrequipment2
$0
Programname2
Project Grant - PA: Pandemic Preparedness and Health Emergencies Research
Keyworddelim
Acute Respiratory Distress Syndrome (Ards); Animal Acute Lung Injury (Ali) Models; Artificial Intelligence (Ai); Histological Image Assessment; Lung Injury Score
Abstract
Acute respiratory distress syndrome is a life-threatening lung injury that results from various causes including trauma, infections (like COVID-19), or burns. In the laboratory, animal models are used to investigate lung injuries. The most important outcome used in these animals is 'histology'. Histology is a method where we examine lung samples under a microscope to see the injury. These images are then analyzed by experts to quantify how much lung injury has occurred. Such analysis is time-consuming, expensive, and prone to mistakes. This can contribute to errors in our understanding of lung injury. An automated method is needed to overcome these limitations. In preparation for this proposal, we performed a comprehensive literature review that revealed there is currently no artificial intelligence (AI) solution available for this purpose. Using AI, we aim to create and validate a novel platform that can accurately analyze histology images of lung injury in animal models. We have assembled an international team of lung injury experts to support this project. They will provide histology samples from their labs for us to develop and validate the platform. We will train an AI model using Python, a widely used programming language. We have collected histopathology images and have annotated them for lung injury. The AI model will learn to identify and measure the severity of lung injury from these images. To ensure precision and dependability, we will evaluate the model's performance and make adjustments. We will also validate our AI model using a separate set of images that have been independently quantified by researchers. Validation will provide an assessment of our AI model's performance and identify areas for refinement. Our work will establish a standardized and reliable AI platform for quantifying lung injury in animal models. Automation of histological assessment using artificial intelligence has the potential to accelerate lung injury studies significantly.
Competitiondate
202403
Approvedterm2
1 yr 0 mth
Pinamesdelim
Mer, Arvind; Lalu, Manoj M; Liaw, Patricia C
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $100,000~ $0~