AI Could Help Find Undiagnosed COPD on Lung Cancer Screening CT

Jean Bourbeau, MD, says AI analysis of CT scans is not ready for routine use but could reshape early COPD diagnosis as GOLD expands its AI guidance. Artificial intelligence (AI) analysis of lung cancer screening CT scans could help clinicians find undiagnosed COPD but still requires validation before routine use, according to an expert who spoke on AI and emerging technologies in COPD at CHEST 2026 .

The Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2026 report introduced Chapter 6, the first chapter in the report dedicated to AI and emerging technologies in COPD.1,2 The update also revised the report’s screening and case-finding section, and GOLD leadership has pointed to lung cancer screening as an opportunity to identify patients with undiagnosed, undertreated disease.2,3 “COPD is the Alzheimer of the lung,” said Jean Bourbeau, MD, MPH, professor in the Division of Respiratory Medicine at McGill University and director of the COPD clinic and pulmonary rehabilitation at the McGill University Health Centre.

“When people come to our attention, there is already a lot of disease.” Bourbeau spoke with HCPLive at CHEST 2026, held October 18-21 in Phoenix, Arizona. AI Analysis of Lung Cancer Screening CT for COPD Case Finding Bourbeau said GOLD 2026 addresses underdiagnosis, and the 2027 report will emphasize it even further.

Earlier diagnosis would allow clinicians to treat patients sooner and prevent emergency department visits and hospitalizations, he explained. At present, clinicians miss opportunities to recognize at-risk patients through electronic health records and through imaging. Patients who smoke and qualify for lung cancer screening are the same patients at risk for COPD, according to Bourbeau.

Screening CT scans generate a large volume of information requiring an analytic process, and he said AI can fill this gap. The goal is to turn imaging obtained for early lung cancer detection into a way to flag patients at risk for COPD. Bourbeau does not consider the technology ready for prime-time use, though he expects it to arrive in the near future.

In his view, patients flagged by AI would proceed to spirometry for diagnostic confirmation. He said such a pathway would completely transform early COPD diagnosis, noting many of these patients are not early in their disease course but remain undiagnosed. Validation and Clinician Oversight in GOLD’s AI Guidance Bourbeau described 2 roles for GOLD: guiding physicians toward evidence-based practice and expert recommendations and highlighting where the field is headed.

The 2 major areas GOLD has promoted are an approach to COPD built around disease activity and stability, and an introduction to AI. Chapter 6 was developed under the leadership of Alvar Agusti, MD, of Hospital Clinic, University of Barcelona, whom Bourbeau called visionary for scoping both what AI currently does in COPD and its limitations.

A manuscript now under review will cover the topic in greater detail ahead of the GOLD 2027 report, according to Bourbeau.2 He expects the chapter to grow each year as the field tracks progress in AI for COPD diagnosis and management, helping physicians decide which tools are ready for practice.

Telehealth efficiency is another area where AI may contribute, he noted. “We have to know what we can do now,” said Bourbeau. “We have to know where the research is, but we still need validation.” Integrating Self-Management Research Into AI Tools for COPD Frequently Asked Questions What does GOLD 2026 Chapter 6 cover?

Chapter 6 is the first GOLD chapter dedicated to AI and emerging technologies in COPD. It outlines current applications of AI in COPD along with their limitations. Is AI analysis of lung cancer screening CT ready to detect COPD?

Not yet, according to Bourbeau. He expects the technology to reach practice in the near future once validation is complete. How would AI-based CT case finding for COPD work in practice?

AI would analyze lung cancer screening CT scans to flag patients at risk for COPD. Flagged patients would undergo spirometry to confirm the diagnosis. Bourbeau’s team is launching a new AI program to identify the needs of patients leaving the hospital after a COPD admission, with heart failure as a planned next step.

The program is designed to bring the most urgent daily tasks for each patient to the nurse’s dashboard, with the aim of preventing readmission. It builds on years of research from Living Well with COPD, a self-management program studied by Bourbeau’s group. He said academic groups with 20 years of data and validated approaches to self-management need to be included in commercial AI development, an element he described as currently lacking.

Safety nets must also be built, with healthcare professionals remaining behind the technology. “We want the healthcare professional to get better at what they’re doing the best, and we want AI to help the healthcare professional at what AI should be doing the best,” Bourbeau said.

Bourbeau cautioned clinicians against being excited by the technology alone and urged the field to contribute content and validation. “The future is starting now,” he said. Watch part 1 of our interview with Bourbeau here: AI and Emerging Technologies in COPD Care, With Jean Bourbeau, MD Bourbeau has no reported disclosures.

References Global strategy for prevention, diagnosis and management of COPD: 2026 report. Global Initiative for Chronic Obstructive Lung Disease. Published November 10, 2025.

Accessed October 23, 2026. https://goldcopd.org/wp-content/uploads/2026/01/GOLD-REPORT-2026-v1.3-8Dec2025_WMV2.pdf Key changes GOLD 2026. Global Initiative for Chronic Obstructive Lung Disease. Published November 10, 2025.

Accessed October 23, 2026. https://goldcopd.org/wp-content/uploads/2025/11/KEY-CHANGES-GOLD-2026-10Nov2025.pdf 2026 GOLD report provides key updates on COPD exacerbations, case finding. CHEST Physician. Accessed October 23, 2026. https://www.chestphysician.org/2026-gold-report-provides-key-updates-on-copd-exacerbations-case-finding/ Related to this article