From X-rays to CT scans: AI’s next leap in radiology now launched

Harrison.ai has launched its CT Chest AI platform designed to assist clinicians in detecting 167 radiological features, including those indicative of life-threatening conditions, tumours, and chronic diseases.

Claiming one of the broadest coverages in chest imaging, the Harrison.ai CT Chest identifies “close to” nine critical findings such as pulmonary embolism, acute aortic syndrome, pneumothorax, and acute rib fractures. The algorithm prioritises urgent cases on the worklist, highlights the slices containing the findings, overlays them for visual clarity, and provides interpretations with confidence scores, tools intended to help clinicians act faster on time-sensitive cases.

Cancer screening is another area where the solution aims to make a tangible difference. With up to 70% of lung cancers still diagnosed at later stages, Harrison.ai CT Chest supports earlier identification of suspected tumours and lesions across lung tissue, soft tissue, mediastinum, lymph nodes, and even the upper abdomen. The solution can assist radiologists in detecting, staging, and monitoring features suggestive of lung, gastric, and pancreatic cancers, while also streamlining lung screening workflows through reporting tools designed to save valuable time.

This launch comes roughly a year after Harrison.ai unveiled Annalise.ai, its AI-powered chest X-ray diagnostic tool, which leverages large language models to analyse images. Annalise.ai was trained using de-identified patient data from the IMED Radiology Network, spanning around 250 clinics across Australia. Following that release, concerns were raised over whether patient consent had been secured for the use of their x-rays in AI training.

An investigation by the Office of the Australian Privacy Commissioner found no breach, concluding that adequately de-identified data likely ensured compliance under the Privacy Act, as sufficiently de-identified information does not qualify as “personal information.”

However, experts caution against complacency. “There is, however, still a significant risk of reidentification based on the combination of characteristics taken from an individual patient’s data that may be used in training the AI. This risk is increased where multiple datasets containing data from the same patient are used together,” explained Andrew Lowe, Principal at Spruson & Ferguson.

Dr Aengus Tran, co-founder and CEO of Harrison.ai., said “We’ve seen firsthand how a comprehensive AI solution for a single modality can transform radiology workflows. With CT Chest, we’ve taken the same comprehensive approach [as the chest X-ray AI solution] to empower clinicians with deeper insights and faster, more confident decision-making.”

In reimagining healthcare across the entire patient journey, Health Industry Hub^TM is the only one-stop-hub uniting the diversity of the Pharma, MedTech, Diagnostics & Biotech sectors to inspire meaningful change.

The Health Industry Hub^TM content is copyright protected. Access is available under individual user licenses. Please click here to subscribe and visit T&Cs here.