Will AstraZeneca Prove the Potential of AI in Lung Cancer Diagnosis?

As part of AstraZeneca’s ongoing partnership with Qure.ai, they have launched a 6-month collaborative research project in the UK. The pilot study aims to determine the effectiveness that prognostic artificial intelligence technology can have in detecting lung cancer when used as an assistive tool for radiologists. 

Why it's notable:

• The 6-month study intends to assess the speed and accuracy of Qure.ai’s AI-powered technology in detecting lung cancer. The trial will involve the analysis of more than 250,000 x-rays, taken from patients within the greater Manchester area in the UK. 

• This clinical trial is part of an ongoing partnership between AstraZeneca and Qure.ai. The original partnership formed in 2020 was part of AZ’s Emerging Markets Health Innovation Hubs programme and was focused on employing AI technology to detect lung cancer in emerging markets, specifically in the APAC, Africa and Latin America regions. 

• Qure.ai is an Indian-based startup that has developed deep-learning solutions to aid physicians in routine detection and diagnosis. The AI-powered technology specific to chest x-rays, qXR, has demonstrated high sensitivity, accuracy and specificity as a radiologist-assisted tool to identify signs of lung cancer, TB and heart failure. 

Industry Implications:

• Lung cancer is a leading cause of mortality worldwide. On average, in the UK alone, 95 people die of lung cancer every day, yet studies have indicated that 79% of cases are preventable. Due to the huge economic and social burden of lung cancer, there is demand for novel technologies which can improve diagnosis and reduce the progression of the disease, and prognostic AI-powered technology may hold the key to this global burden.

• Pharma is interested in exploring the potential of AI-powered technology in medical imaging, and similar collaborations to this latest partnership have emerged in this space, including Roche’s partnership with PathAI, AstraZeneca’s partnership with Oxipit and Bayer’s recent acquisition of Blackford Analysis. 

• Evidently, AI in medical imaging has been gaining a lot of traction due to its ability to improve diagnosis and the potential to reduce the workload across understaffed hospitals. Unfortunately, for the most part, AI technologies currently remain in the research setting and are not yet routinely used in clinical practice. While, the recent launch of this, and similar, clinical pilot studies are an indicator of what is to come, it still begs the question of whether AI can manifest to become commonplace in the clinical setting.