Telefónica Tech Deploys AI to Screen Lung Cancer in Spain

Telefónica Tech has expanded deployment of clinical decision support software across regional health services to help doctors rapidly prioritise urgent cases.

It partnered with global healthcare technology company Harrison.ai, which has deployed an AI diagnostic tool across Spain's regional health services.

The software analyses chest X-rays to detect up to 124 clinical findings. These include pulmonary nodules that could indicate lung cancer.

Training and certification standards

According to Harrison.ai, the algorithm was trained on data from more than 780,000 chest X-ray studies, with at least three qualified radiologists independently labelled each study during development.

The system holds Conformité Européenne (CE) IIb medical device certification. This represents the highest level of assurance for diagnostic support tools in the European Union (EU), permitting use in potentially lethal health conditions. It confirms compliance with EU legislation governing safety, efficacy and quality for medical devices.

How the system works

Healthcare professionals can access AI-generated insights through a standard web browser. The technology provides a rapid second opinion on chest X-ray findings and then medical staff retain final authority on patient diagnoses and treatment decisions. 

The system functions as a clinical decision support tool rather than an autonomous diagnostic device.

Clinicians in accident and emergency departments or primary care units can also use the software to filter cases. It identifies which patients may require specialist referral.

Integration with existing infrastructure

The AI tool integrates directly into existing healthcare information technology (IT) systems and radiologists’ workflows. According to Telefónica Tech, this approach allows customisation to match technical and clinical requirements.

Radiologists can use the system to categorise cases by urgency, with the aim of prioritising time-critical patients within their existing patient lists.

“The incorporation of AI into radiology represents a significant step towards more effective and efficient healthcare models, in which technology becomes an ally of healthcare professionals to deliver faster, more accurate and personalised care,” says Carlos Martínez, Head of Data and AI at Telefónica Tech. 

“With this new solution, which we are already offering to several regional health services, we are continuing to expand the services we offer our clients and making progress towards our aim of becoming the best gateway for citizens, businesses and public administrations to access digital technologies.”

Clinical workflow benefits

The technology could reduce diagnostic response times by streamlining workflows and help detect diseases at earlier stages.

“We are proud to partner with Telefónica Tech to bring clinically validated and scalable AI to healthcare professionals across Spain,” says Dr Aengus Tran, CEO and Co-founder of Harrison.ai. 

“Telefónica’s reach in Spain gives us the opportunity to do what we set out to do: reduce the gap in healthcare capacity.”

Aengus adds the aim is to help clinicians detect diseases such as lung cancer earlier, prioritise urgent cases and ensure no findings are missed. The software limits unnecessary specialist referrals by providing an initial assessment layer.

Industry perspective on deployment

Dean Bubley, Founder and Director of Disruptive Analysis, commented on the partnership in a LinkedIn post. He noted the approach aligns with vertical AI solution strategies pursued by some telecommunications companies in Asia.

“This aligns with some other (mostly Asian) telcos that are pursuing ‘AI Factory’ strategies more geared towards vertical AI solutions, either for B2B or B2C sectors,” he wrote. He described the integration approach as exactly right.

“No enterprise is going to listen to an arriving telco on its doorstep and suggest a complete re-architecting of its IT (and especially AI) portfolio around network-based compute” he adds. “That doesn't mean that some workloads, in some places, couldn't also benefit from more connectivity. I'm sure that shifting big X-ray images and data around will indirectly drive network upgrades in some places as well.”​​​​​​​