Acute Respiratory Distress Syndrome (ARDS) is studied in the lab using preclinical models of Acute Lung Injury (ALI). Histological assessments play a pivotal role in understanding lung injury, yet current methods are time-consuming, prone to selection bias, and heavily reliant on human judgment. These limitations make it difficult to compare findings across studies and labs, hindering progress in the field.
Our LungInsight team is dedicated to transforming this process by developing advanced artificial intelligence (AI) tools to assess lung injury in histological images from laboratory models. Using cutting-edge machine learning techniques, including Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), our goal is to create standardized, automated, and reliable systems that streamline image analysis, reduce bias, and enhance accuracy.
We are also ensuring that our approach includes ‘explainable AI’. This means that our AI tools will offer clear and understandable explanations for decisions made and outputs produced. This will build confidence in the tools and also help researchers better interpret findings. Ultimately, LungInsight strives to advance and accelerate the study of preclinical ALI, paving the way for both the deeper understanding of lung diseases as well as the development of novel therapies.
