In 2019, the use of artificial intelligence in medical imaging was still in its infancy.
The technology was plagued by error rates as high as 50 per cent, a figure that has barely changed across recent decades.
And then the onset of the pandemic the following year increased demand for imaging, creating huge backlogs and an increased strain on clinicians.
This motivated brothers Dr Aengus Tran and Dimitry Tran, both Vietnamese immigrants, to develop a tool that can use AI to improve the accuracy and efficiency of medical diagnostics, with an aim of improving outcomes for patients around the world.
In early 2020, the brothers’ company Harrison.ai formed a joint venture with I-MED, Australia’s largest radiology network, and Annalise.AI was formed, with an initial focus on chest x-rays.
Chest x-rays are one of the most commonly-performed patient exams, but can be difficult to accurately interpret.
The AI tool developed by the joint venture can detect 124 findings in chest x-rays in under 10 seconds, alleviating pressure on clinicians and improving health outcomes for patients. This means that conditions such as trauma, cancer and heart disease can be diagnosed faster, allowing for quicker treatment and freeing up the time of clinicians.
Annalise.AI Enterprise CXR is a finalist in the InnovationAus 2023 Awards for Excellence in the Health Tech category. You can secure your tickets to the black-tie event here.
The Annalise AI model was trained on a dataset of 520,000 chest x-ray studies, comprising 820,000 individual x-rays, taken from Australia and around the world. The company then conducted the largest radiology research study ever performed to test and analyse its algorithm.
This showed that radiologists were significantly better at detecting diseases when they were using the AI tool, which was acting as a second set of eyes for them.
Further research has found that 90 per cent of radiologists said the tool increased the accuracy of their diagnoses of health conditions, and reduced the time they were spending on each case by 12 percent.
The joint venture secured a $129 million Series B round in 2019, led by Sonic Healthcare and I-MED Radiology Network, alongside existing investors Blackbird Ventures and Skip Capital.
The Annalise.AI Enterprise CXR tool automatically detects the suspected locations of x-ray findings on multiplanar reformatted images with pathology-appropriate window and level settings.
The tool is now in use around the world, and one in four Australian radiologists have access to the chest x-ray platform. Last year, the tool processed nearly 2 million x-rays in Australia alone.
The InnovationAus 2023 Awards for Excellence are proudly supported by Investment NSW, AusIndustry, Australian Computer Society, Technology Council of Australia, Agile Digital, CSIRO, TechnologyOne, IP Australia, METS Ignited and Q-CTRL. You can book your seat – or book a table – for these awards right here.
The InnovationAus 2023 Awards for Excellence are proudly supported by Investment NSW, AusIndustry, Australian Computer Society, Technology Council of Australia, Agile Digital, CSIRO, TechnologyOne, IP Australia, METS Ignited and Q-CTRL.
Protecting your great ideas with intellectual property (IP) rights can lead to lasting benefits for your growing business. IP refers to creations of the mind, such as a brand, logo, invention, design or artistic work. Head to the IP Australia website to find out more about IP, and how it might help your business.
Reserve your place at the InnovationAus Awards for Excellence black-tie dinner by clicking here.
Do you know more? Contact James Riley via Email.