Using machine learning to improve the characterisation and quantification of fibrotic lung.

Pulmonary fibrosis is severe untreatable feature of some acute and chronic lung diseases, it has recently been recognised as a consequence of severe COVID-19-induced pneumonia. Fibrosis is essentially scar tissue that replaces functional lung tissue and is a feature of Idiopathic pulmonary fibrosis (IPF) where it is progressive in nature and always causes the demise of the patient. Fibrosis is also observed in Acute Respiratory Distress Syndrome (ARDS) where it develops rapidly, does not usually progress but can leave patients with permanent lung damage and lifelong reduced physical capacity.  The current pre-clinical models of pulmonary fibrosis have a poor record in translating successful treatments from the laboratory to clinic, in part, due to their inability to mimic molecular or cellular mechanisms that occur during tobacco smoking, viral infections and the development of fibrotic lung diseases such as IPF. We are interested in developing pre-clinical models that better represent  these molecular and cellular mechanisms to improve and develop fibrosis treatments that will be successful in the laboratory and the clinic.

The aim of this project is to combine ‘wet lab’ methods with machine learning to improve the characterisation and         quantification of fibrosis in mouse models with real world relevance to IPF and ARDS. You will use a number of laboratory techniques including in vivo disease modelling, tissue culture, QPCR, western blotting and FACs as well as machine learning methodologies using the facilities located within the Biological Optical Microscopy Platform (BOMP).