Artificial Intelligence supported spirometry in primary care: Clinical evaluation of the impact

01 Apr 2024
Introduction: Few primary care physicians feel comfortable performing spirometry tests or interpreting spirometry results. This results in an underestimation of chronic respiratory disease diagnoses, including asthma and COPD, and less time- and cost-effective secondary care visits. This study aims to investigate the perceived added value of providing ArtiQ.Spiro, i.e. AI-based software to support GPs in performing and interpreting spirometry, both in usefulness for diagnosis and in perceived support by primary care providers. Methods: 28 GPs from 6 different practices in Belgium used ArtiQ.Spiro during a 3-month period alongside existing referral pathways. Questionnaires and semi-structured interviews were used to assess the general experiences and believes on the use of spirometry in a primary care setting before and after the trial with participating GPs. The accuracy of the diagnostic hypothesis based on the clinical evaluation and interpretation of spirometry, supported by the AI software, was established by presenting the same data to a panel of pulmonologists and comparing diagnoses. This panel reviewed the same data and all relevant information from the patient record 3 months after the spirometry session. Results: Most GPs see spirometry as a primary care investigation and do see added value in diagnostic and follow-up purposes of spirometry. However, insufficient training, lack of experience and logistic concerns are the main barriers in current regular use of spirometry. Implementation of AI software increases confidence levels in diagnosis on a case-to-case basis and was perceived (very) useful by the GPs for quality control (75% of respondents) and interpretation support (69% of respondents). When comparing the AI-proposed diagnosis with the pulmonologists’ diagnosis, 82% of diagnoses were deemed accurate. Discussion: GPs do not perform spirometry consistently due to a lack of training, experience and logistic concerns. AI software can serve as a supportive tool to increase confidence to execute and interpret spirometry.

Resource information

Respiratory conditions
  • Chronic Respiratory Disease
Respiratory topics
  • Technology
  • Spirometry
Type of resource
Athens 2024
Sofie Willaert1, Julie Maes2, Elena Smets2, Maarten De Vos1,3, Marko Topalovic2, Jan Verbakel1,4 1Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium, 2ArtiQ NV, Leuven, Belgium, 3Department of Electrical Engineering, KU Leuven, Leuven, Belgium, 4NIHR Community Healthcare Medtech and IVD cooperative, Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, United Kingdom