Blog

Radiology is suffering one of the biggest staff shortages in the NHS which is causing patient delays and increasing costs. Recent figures suggest Radiologists’ workload in reading and interpreting scans has increased by 30% 2012-2017 yet the UK has a consultant radiologist vacancy rate of 10%. Hospitals have often had to outsource scans to the private sector or increase overtime on already highly stretched staff costing approximately £116m per year.

However, could AI alleviate this problem by augmenting the specialist skills of radiologists?

Dr Shah Islam, Clinical Research Fellow at Imperial College has been creating an AI model to augment the skills and expertise of his colleagues. Working with L3C, a specialist AI Cloud provider he has used IBM’s AI Vision solution embedded in L3C AI Cloud to create and train an AI model that could dramatically reduce the number of scans and x-rays needing to be assessed by radiologists.

With approximately 70% of scans and x-rays being classified as ‘normal’ and not needing further investigation, Dr Shah has been working on the premise that a properly trained AI model could eliminate these from radiologists’ workload.  This would lead to quicker feedback to clinical teams where no issues had been found and allow radiologists to focus on those scans that needed further investigation.

AI Vision allows subject matter experts such as Dr Islam who have identified bottlenecks in radiology workflow  to create powerful AI models by labelling images according to their specialist domain. Using a publicly available set of anonymized x-rays and scans he has been able to create a model that can rapidly classify whether an x-ray is ‘normal’ or needs further investigation. Obvious initial applications could be triaging of abnormal chest x-rays and breast mammograms as well as skeletal fracture detection.

AI models will always need human expertise to create the initial validation of images but L3C and IBM have provided an environment that makes these easy to create without any prior programming knowledge, for the first time giving autonomy back to experienced clinicians who understand where AI can best augment workflow and are best served to use it. The unique platform provides the processing capacity based on IBM Power9 servers to enhance the time to accuracy.

For more information contact office@l3c.cloud