BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Optimising the definition of MR-based lung imaging biomarkers - Ju
 an Delgado\, GSK
DTSTART:20180206T130000Z
DTEND:20180206T140000Z
UID:TALK100822@talks.cam.ac.uk
CONTACT:Dr Vivien Gruar
DESCRIPTION:*Introduction*\n\nMagnetic Resonance Imaging (MRI) is of inter
 est in pharmaceutical development for its translational potential from ani
 mal research into the clinical setting. MRI of the lung is a relatively no
 vel approach for the assessment of the amount of cellularity and oedema (f
 luid) as indicators/biomarkers of pulmonary toxicity and/or efficacy. The 
 quantification of these lesions is typically assessed by delineating regio
 ns-of-interest (ROIs) on the images and applying several filtering techniq
 ues to measure the final levels of these biomarkers. The selection of thes
 e ROIs (a process known as segmentation) can be achieved automatically usi
 ng various computational approaches. However\, the great variation in nois
 e and image quality in MR data results in automated segmentation often bei
 ng challenging. In fact\, the greatest difficulty with automation is the d
 iseased lung which appears closer in aspect to adjacent tissue (muscle/liv
 er/heart) than to healthy lung. Therefore\, manual segmentation remains th
 e principal mechanism for analysis. \n\n*Opportunity*\n\nWe have at our di
 sposal a large historic animal MR dataset (>1000 tomographic images) on mo
 re than seven models of disease/toxicity of the lung\, from various strain
 s of rats and mice. Furthermore\, although the images were acquired on a s
 ingle preclinical MR scanner\, these were made using various multiparametr
 ic MR sequences that result in different contrasts and signal-to-noise rat
 ios. The values of six combinations of segmentations and filtering methods
  will be provided for testing. These methods include: manual\, intensity t
 hresholding\, and multi-atlas segmentation\; combined with: high-pass filt
 ering for blood vessel and noise elimination\, and holistic quantification
 . Animal species\, strains\, disease model\, and MR parameters are availab
 le for use as covariates in the analysis.\n\n*Aim & Impact of the Project*
 \n\nIn this project\, we aim to understand the following questions:\n* to 
 what extent can automated segmentation be used?\n* which methods are best 
 suited for which covariates?\n* when is manual segmentation unavoidable? \
 n* what are the gaps in analysis that will require new segmentation method
 s?\n\nThe understanding of these statements is crucial to improve the effi
 cacy of MRI investigations\, introduce common approaches that can be appli
 ed across multiple centres\, and establish the precise criteria needed to 
 define respiratory conditions such as COPD (Chronic Obstructive Pulmonary 
 Disease)\, asthma\, IPF (Idiopathic Pulmonary Fibrosis) or drug-induced lu
 ng injury (DILI) via MRI. A clearer understanding of the use of MRI in thi
 s context will help support improvements to the lives of millions of patie
 nts affected by respiratory disease.\n\n*The student*\n\nYou will have a c
 lear understanding of machine learning techniques and cloud point analysis
  of complex datasets. You should have excellent theoretical understanding 
 of variational optimisation and Bayesian statistics\, as well as hands-on 
 experience with unsupervised machine learning techniques.\nInterest in pha
 rmaceutical development\, human and animal physiology\, medical imaging\, 
 and/or computer vision is beneficial. \n\nIf you join us\, you will be dir
 ectly contributing to impact the use of automated methods to extract decis
 ion-making biomarkers from image supporting medicinal development. Further
 \, you will be able to liaise with a diverse range of scientists from in-v
 ivo biologists to data scientists\, and be exposed to cutting-edge science
  and equipment."\n
LOCATION:MR3 Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR