Shanghai, China – July 25, 2025 – Luca Healthcare, a leader in developing predictive healthcare solutions using AI algorithms and machine learning, today announced that its groundbreaking research on an innovative AI algorithm for Chronic Obstructive Pulmonary Disease (COPD) diagnosis has been accepted for presentation at the prestigious European Respiratory Congress (ERS Congress) 2025. The presentation, titled “Multimodal deep learning in COPD diagnosis: a comprehensive evaluation of cough acoustics and clinical symptom integration,” will take place during the ERS annual congress from September 27 to October 1, 2025 in Amsterdam, Netherlands.
COPD remains one of the leading causes of death globally and presents a significant burden on healthcare systems and economies worldwide. In 2021, COPD caused an estimated 3.5 million deaths globally, making it the fourth leading cause of death worldwide and projected to remain the third leading cause of death through 2050. The economic burden is equally staggering, with direct and indirect costs estimated to be $36 billion annually in the United States alone.
Current diagnostic methods for COPD often require costly and inconvenient lung function measurements using specialized clinical equipment, leading to a substantial fraction of undiagnosed patients, particularly in low- and middle-income countries where nearly 90% of COPD deaths in those under 70 years of age occur. Luca Healthcare’s innovative algorithm leverages advanced deep learning techniques to analyze cough acoustics, integrating them with clinical symptoms to provide a reliable and accessible diagnostic tool. This novel approach addresses the critical need for early and efficient COPD diagnosis, moving beyond traditional, often prohibitive, diagnostic pathways.
“We are incredibly proud to have our work recognized and accepted by the European Respiratory Congress, a testament to the scientific rigor and potential impact of our research,” said Echo Chen, Ph.D., CEO of Luca Healthcare. “Our innovative AI algorithm, ‘Cough Search,’ offers a promising solution to the challenges of COPD diagnosis by utilizing acoustic biomarkers. This breakthrough has the potential to transform how COPD is identified, making early detection more affordable and accessible for millions globally, and ultimately improving patient outcomes in the face of this widespread and debilitating disease.”
The presented research details a comprehensive evaluation of various models, including vision transformers, for respiratory disease classification based on cough sounds. Luca Healthcare’s proprietary “Luca Cough” dataset, which includes diverse cough data from both COPD and non-COPD subjects, played a crucial role in validating the algorithm’s high accuracy, demonstrating an AUROC (Area Under the Receiver Operating Characteristic curve) of 92.5% for COPD/non-COPD classification. This innovative approach signifies a major step forward in leveraging readily available data, like cough sounds, for crucial medical diagnoses.
About Luca Healthcare
Luca Healthcare is a leader in using AI algorithms and machine learning to develop predictive healthcare solutions. The company is dedicated to providing affordable, accessible, and clinically validated software-based screening, tracking, management, and treatment solutions to a wider patient population through digital innovation and global partnerships.
For additional inquiries, please contact:
Garry Wang
Luca Healthcare
Email: garry.wang@lucahealthcare.com