Image analysis technologies can automatically identify individual cells and measure them
Advances in medical imaging
Advances in medical imaging are assisting with the earlier detection of a range of diseases including cardiovascular disease, stroke, cancers and neurodegenerative diseases such as Alzheimer’s disease.
19 January 2009 | Updated 14 October 2011
At the cellular level, new imaging technologies can identify changes associated with diseases faster and more reliably than before.
Early diagnosis provides the opportunity to deliver medical interventions to either prevent the development of the disease or prevent the progression of the condition and emergence of symptoms.
Disorders, such as cancers and Alzheimer’s disease, are major contributors to increased health expenditure and are well suited to diagnosis through imaging. Earlier detection of such diseases is expected to realise significant economic and social benefits.
In particular, medical imaging can provide the quantitative tools to help industry develop cheaper and more efficient screening tests.
Advanced medical imaging investigates new medical imaging modalities, including novel molecular agents, to advance health sciences. We also seek to develop new image analysis algorithms to automatically extract relevant information from the large quantities of information produced from digital images.
Within CSIRO we have three major research activities that address these areas:
Medical imaging technologies
Medical imaging research investigates new and improved modalities for clinical diagnosis, using minimally invasive techniques for the whole body, tissues, organs or cells. It encompasses various methods including:
X-ray and computer tomography (CT)
magnetic resonance imaging (MRI)
single-photon-emission tomography (SPECT)
positron-emission tomography (PET)
Our primary research goals in medical imaging include the development of:
innovative targeted contrast reagents for various imaging modalities including MRI, X-Ray and ultrasound
novel X-ray imaging methods and instruments, especially involving phase contrast imaging and synchrotron systems
novel image processing systems.
Our research aims to enable:
the assessment of the molecular basis of cell function leading to diagnosis of disease states at very early stages
improved monitoring of disease progression and response to treatment
the application of targeted therapy to locally identified disease states
increased resolution and sensitivity of modalities that measure anatomical features (e.g. CT or MRI).
CSIRO has established research and industry partnerships to ensure the successful development of medical imaging agents and techniques.
CSIRO’s image analysis technologies allow computers to do many tasks, normally carried out by eye, but much faster and more accurately.
One of the key issues addressed at CSIRO is to develop computer vision tools that can recognise and discriminate between objects and make detailed, accurate measurements much better and faster than people can.
Digital images contain a wealth of information and the medical and biotech industries produce massive amounts of image data. Technologies such as microscopy, microplate readers and microarray scanners are used for purposes like disease diagnostics, drug discovery and quality control.
CSIRO is developing computer vision tools that, like the human eye, can recognise and discriminate between objects but, with increased detail, accuracy of measurements and faster than a human operator.
CSIRO’s Biotech Imaging team is applying a growing portfolio of image analysis technologies to problems like:
measuring and classifying the shapes of nerve cells exposed to drug treatments
understanding cell processes such as ‘vesicle fusion’
monitoring cell movements
counting and measuring linear features such as hairs and cell membranes.
We are currently developing cellular screening tools for location proteomics, neurobiology and microbiology.
As part of CSIRO's Computational and Simulation Science initiative, we are also exploring applications of graphics processing units (GPUs) to speed up image analysis tasks.
Biomedical imaging analysis
The aim of this research is to develop new methods and software for characterising diseases from medical images, and for the training and planning of computer assisted interventions.
On-going projects cover a wide range of imaging modalities (MRI, PET, CT, ultrasound, optical imaging) and aim at fusing anatomical, functional and molecular information.
Our approach is to embed into image processing algorithms expert knowledge from physiology, biology, pathology and the physics of the imaging acquisition.
We are developing novel shape, signal, and anatomical models at different levels: body, organ and tissue.
Main applications include:
knee cartilage thickness estimation from MRI for osteoarthritis diagnostic and treatment planning
Alzheimer’s disease early detection using imaging biomarkers from MRI and PET
brain tumour characterisation from MRI and PET
improve prostate radiotherapy using MRI and CT
automatic identification of organs in small rodents using micro-PET and micro-CT
advanced colonoscopy simulation, developed for use as a training tool, using soft tissue modelling and ultra-realistic visualisation
ultrasound simulation for training of prostate biopsy intervention.
CSIRO has implemented a large number of medical imaging algorithms under a common development platform.
The Cross Platform Medical Imaging Library (MILX™) expands on algorithms provided by the Insight Tool Kit (ITK), the Visualisation Tool Kit (VTK), and other open source libraries, and we are sharing these resources with our collaborators.
Read more about CSIRO Molecular & Health Technologies.