CSIRO and Singular Health AI-driven project successfully delivered

singular health segment
Segmented vertebrae registered to a patient using Singular's MedVR software. Image Supplied.

Recently ASX-listed healthtech Singular Health Group Ltd and Australia’s national science agency CSIRO have announced their successful collaboration of developing an automated segmentation of a spinal vertebra from CT scans.

So far, the automated segmentation has an accuracy rate above 95%. Once edited by radiologists and surgeons this increases to 100%.

This innovation allows surgeons and radiologists to not just visualise the spine in 3D using Singular’s software, but also to manipulate individual vertebrae, better plan surgeries and even allow customised surgical implants and guides.

It is more big news for Subiaco-based Singular Health which recently completed its Initial Price Offering (IPO). Additionally, the company was one of 36 recipients across WA for the New Industries Fund’s $20,000 Innovation Vouchers last year.

The project received funding and support from the CSIRO Kick-Start initiative, a program that gives Australian startups and small businesses access to CSIRO’s research and development expertise and capabilities.

Executive Director of Innovation at Singular Health, Dr Guan Tay, said he believes the use of AI will disrupt the medical imagery industry.

The use of artificial intelligence in medical imaging, and more specifically radiology, has the ability to profoundly change the workflow for radiologists.

With around 45,000 Australians undergoing spinal surgery every year, the rapid segmentation will save thousands of hours for radiologists and surgeons who will only have to make small mark-ups and/or validate the segmentation as opposed to manual segmentation slice-by-slice.

Dr Guan Tay, Executive Director of Innovation at Singular Health

The analysis and segmentation only takes about two minutes using Singular’s MedVR & 3DicomViewer software. This makes the process semi-automated as surgeons and radiologists can edit the output in order to ensure compliance with the imaging.

To put this into perspective, the manual segmentation process by medical practitioners usually takes hours.

Rather tediously, the manual process involves identifying and marking verterbrae on every single CT slice, of which thee can be hundreds. Unsurprisingly, this task can be mentally fatiguing.

An example of the manual segmentation process, which can take hours to complete. Image Supplied.

Dr Dadong Wang, the Quantitative Imaging Research Team Leader at CSIRO’s Data 61, praised the user-friendliness of the new innovation.

CSIRO has been able to apply data science and Artificial Intelligence to enable users to better visualise the spine and individual vertebrae. This semi-automated segmentation and labelling of the spinal vertebrae is both time saving and can be represented in 3D models, which can also be viewed in virtual reality, providing higher interactivity.

Dr Dadong Wang, CSIRO Research Lead

Singular Health was founded in 2017 and is a medical tech company that commercialised the proprietary Volumetric Rendering Platform (“VRP”) which is used for the 3D and VR visualisation of anatomy by using standard radiological imagery.

Prior to being released commercially, the spine segmentation function will be further tested and validated.


The New Industries Fund is a sponsor of Startup News