AI’s Precision Lens Brings New Clarity to Chronic Lung Disease

OXFORD, England – August 25, 2026 – For the millions grappling with Interstitial Lung Disease (ILD), the journey is often one of uncertainty. This complex family of over 200 disorders, characterized by progressive scarring of lung tissue, has long challenged clinicians who rely on a combination of lung function tests, patient-reported symptoms, and visual interpretation of CT scans to diagnose and manage the condition. The subjective nature of reading these scans can lead to diagnostic delays and difficulties in tracking the slow, insidious progression of the disease.

Now, a new wave of artificial intelligence is promising to bring objective, data-driven clarity to this field. Brainomix, an AI medical imaging firm spun out of the University of Oxford, is poised to present five new studies at the upcoming European Respiratory Society (ERS) Congress in Barcelona. The research showcases how its AI platform can automatically quantify subtle changes in lung tissue, potentially transforming how ILD is detected, monitored, and treated. This move signals a critical shift from AI as a research concept to a tangible tool in the pulmonologist's arsenal.

The AI Revolution in Pulmonology

The integration of AI into radiology is not a new concept, but its application in chronic respiratory diseases like ILD is gaining significant momentum. The core problem AI aims to solve is variability. One radiologist might describe lung scarring as 'mild,' while another sees it as 'moderate.' Tracking changes across scans taken months or years apart is even more challenging. AI offers a solution by systematically analyzing every pixel of a CT scan to produce consistent, numerical data—or 'imaging biomarkers.'

Brainomix's FDA-cleared and CE-marked platform, Brainomix 360 e-Lung, is at the forefront of this movement. It automates the identification and quantification of key radiographic features associated with lung fibrosis. This isn't about replacing the clinician but augmenting their expertise with a powerful quantitative lens. As one expert in the field noted, AI-powered analysis has the potential to “expedite healthcare delivery and improve clinically meaningful outcomes” by providing a level of precision that is difficult to achieve by eye alone.

The field is becoming increasingly competitive, with companies like Qure.ai and Thirona also developing AI tools for lung imaging. However, Brainomix has carved out a niche with its deep focus on ILD and its integration into major clinical trials. Its strategic partnership with Boehringer Ingelheim, a pharmaceutical leader in pulmonary fibrosis, to use the e-Lung platform in drug development underscores the industry’s growing confidence in these AI-driven biomarkers as reliable endpoints for assessing treatment efficacy.

From Lab to Clinic: The Evidence for Precision

The five abstracts Brainomix will present at the ERS Congress represent a significant body of evidence supporting the technology's broad utility. The studies collectively demonstrate AI's potential across the entire patient journey, from initial assessment to long-term prognosis. One study, a post-hoc analysis of a Phase 2a trial for the drug taladegib, shows the AI was able to detect an increase in lung volume and a regression of ILD markers, providing quantitative proof of a treatment's effect.

Another presentation, led by Prof Ayodeji Adegunsoye from the University of Chicago, links AI-quantified lung scarring to telomere shortening—a genetic marker of aging and disease—and mortality risk. This bridges the gap between imaging data and the underlying biology of the disease.

"These five studies build on Brainomix's growing legacy of clinical evidence in quantitative imaging for fibrotic lung disease, further demonstrating the potential of quantitative CT to deliver objective insights across the Interstitial Lung Disease (ILD) patient journey," said Prof Peter George, a Consultant Pulmonologist and Senior Medical Director at Brainomix, in the company's announcement.

Crucially, another study based on the landmark INBUILD trial shows that early changes detected by the AI platform are associated with future declines in lung function, even in patients who appear clinically stable. This predictive power is a game-changer, allowing clinicians to identify high-risk individuals and intervene before irreversible lung damage occurs.

Beyond the Scan: A New Hope for Patients

For those living with ILD, these technological advancements translate into tangible hope. The ability to get an earlier, more confident diagnosis can alleviate months of anxiety and uncertainty. More importantly, it can enable treatment to begin sooner, which is critical in a progressive disease where preserving existing lung function is paramount.

By providing objective data on whether a therapy is working, AI tools like e-Lung empower doctors and patients to make more informed decisions. If a treatment isn't effective, a change can be made sooner. This personalized approach moves away from a one-size-fits-all model and toward care tailored to an individual's specific disease activity.

Brainomix and its partners are already working to validate this real-world impact. A prospective study known as PROGRESS-PPF is underway to evaluate whether using the e-Lung platform leads to earlier clinical diagnoses of Progressive Pulmonary Fibrosis (PPF), a particularly aggressive form of ILD. The goal is to prove that what works in controlled trials can deliver meaningful benefits in the messy reality of everyday clinical practice, ultimately improving patient quality of life and survival.

The Integration Challenge: Bridging AI and Clinical Reality

Despite the promise, the path from FDA clearance to widespread clinical adoption is fraught with challenges. The hype surrounding AI often glosses over the practical hurdles of implementation. Hospital IT environments are notoriously complex, and integrating a new software tool into established radiology workflows without causing disruption requires significant technical expertise and planning.

Furthermore, building trust is essential. Clinicians need to be confident in the AI's outputs, and that requires transparency, robust validation, and clear protocols for human oversight. Questions of accountability—who is responsible if an algorithm makes an error?—remain a topic of intense debate within the medical community. Another critical concern is data bias; AI models trained on limited demographic data may not perform equally well across diverse patient populations, potentially exacerbating health disparities.

Here, Brainomix may have an advantage. Its initial success was with Brainomix 360 Stroke, an AI platform for acute stroke assessment that is already deployed in over 300 hospitals worldwide. That experience has provided the company with invaluable lessons in navigating the complexities of hospital integration, regulatory compliance, and building clinical trust. By applying this knowledge to the lung disease space, the company is well-positioned to bridge the gap between technological innovation and meaningful clinical impact.