Nine artificial intelligence (AI) technologies have been recommended for use by NHS clinicians to help plan treatment for patients undergoing external beam radiotherapy for cancers.
The draft guidance from the National Institute for Health and Care Excellence (NICE) was drawn up after clinical experts estimated that using AI could speed up the time taken to produce contours or outlines of healthy organs so that cancers, such as lung, prostate, or colorectal can be targeted.
NICE said it was the first time it had recommended the use of AI to help healthcare professionals in their roles. Sarah Byron, programme director for health technologies at NICE, said radiotherapy departments were under "severe pressure", and that "recommending the use of AI technologies to help support treatment planning alongside clinical oversight by a trained healthcare professional could save both time and money".
Speeds Up Treatment Planning Process
Current practice following a CT or MRI scan involves a radiographer marking up, or contouring, an image by hand to highlight organs at risk of radiation damage, lymph nodes, and the site of the cancer. The dose of radiotherapy is calculated to target the tumour site but also to prevent organs and healthy tissue being damaged.
It's predicted that using AI platforms will make it quicker for radiographers to produce these contours or outlines. Clinical experts advising the appraisal committee estimated a time saving of between 10 and 30 minutes per plan, while clinical evidence suggested a range of between three to 80 minutes. The draft guidance stressed that all contours created by AI must still be reviewed by a trained healthcare professional and edited as needed before being used in radiotherapy treatment planning.
Evidence seen by the committee also suggested that AI technologies generally produce similar quality contours of organs at risk as those carried out manually, with most only needing minor edits.
In terms of clinicians' hours saved, NICE said the lowest potential time saving of three minutes could save doctors 3750 hours per 75,000 treatment plans. At 80 minutes, 100,000 hours would be saved. The cost of using the technologies ranged from £4 to £50 per plan and included software and costs associated with training healthcare workers to use the platforms.
Stretched Radiotherapy Departments
A backlog caused by the COVID-19 pandemic measures, plus staff shortages, has led to long waits for scans. Ms Bryon said that "the recommendations made by our independent committee can help to bring waiting lists down for those needing radiotherapy treatment".
England's Health and Social Care Secretary Steve Barclay said: "It’s hugely encouraging to see the first positive recommendation for AI technologies from a NICE committee, as I’ve been clear the NHS must embrace innovation to keep fit for the future."
Commenting on the draft guidance, Dr Sam Godfrey, research information lead at Cancer Research UK, said: "AI has enormous potential to improve how we detect cancer. Harnessing the technology could reduce doctors’ workload, helping them deliver more crucial tests and improving the chances of an earlier cancer diagnosis."
Matt Hobbs, director of research at Prostate Cancer UK, said: "Artificial intelligence has huge potential to boost a man’s chances of living well and longer with prostate cancer. The evolving technology could make testing quicker and better, and could give us ground-breaking insights into how we can treat each man’s cancer in a smarter, more personalised way."
Nine AI Platforms
Due to a lack of robust data on current practice — and other variables such as the costs and time involved — more evidence needs to be collected over the next 3 years to generate a full cost-benefit analysis, NICE said.
The nine platforms given the green light are:
- AI-Rad Companion Organs RT (Siemens Healthineers)
- ART-Plan (TheraPanacea, Oncology Systems)
- DLCExpert (Mirada Medical)
- INTContour (Carina Medical)
- Limbus Contour (Limbus AI, AMG Medtech)
- MIM Contour ProtégéAI (MIM Software)
- MRCAT Prostate plus Auto-contouring (Philips)
- MVision Segmentation Service (MVision AI Oy, Xiel)
- RayStation (RaySearch)
NICE’s draft guidance is open for consultation until Friday 25 August 2023.