AI IN PULMONARY FUNCTION ANALYSIS: REVOLUTIONISING THE DIAGNOSIS OF OBSTRUCTIVE AND RESTRICTIVE LUNG DISEASES

Abstract

Background: Artificial Intelligence (AI) integration for pulmonary function analysis. AI as it foreshadows a new era of pulmonary function diagnosis of obstructive and restrictive lung disease. These AI-based assistants promise to deliver greater accuracy and efficiency, improved automation, and smoother workflow in lung function exploration, yet the acceptance level among the health workforce still needs comprehensive scrutiny. Background: This study investigates acceptance of AI as well as perceived usefulness, ease of use, and associated barriers/challenges in the clinical setting.

Methods: A quantitative, cross-sectional research design was used including data collection via a structured questionnaire based on the Technology Acceptance Model (TAM) for assessing AI use by 273 healthcare professionals. The survey measured major determinants including Perceived Usefulness (PU), Perceived Ease of Use (PEU), Attitude Toward AI (AT), Behavioral Intention to Use AI (BIU), and Perceived Risks. Data analysis involved descriptive statistics, reliability testing (Cronbach's Alpha), correlation analysis, and regression modelling.

Results: The fact that the responses were positively skewed indicates that overall, healthcare professionals have a positive perception of AI in terms of usefulness and ease of use. However, internal consistency was poor (Cronbach’s Alpha = -0.194) , indicating the need for survey instrument refinement. The R² value of the regression analysis (R² = 0.036) indicates PU, PEU, and AT predict only 3.6% of the variance explained in BIU, leading to future research directions such as institutional policies, ethical concerns, and prior AI experience that can affect the actual usage of AI. Normal distribution of data is ruled out by the Shapiro-Wilk normality test, indicating that non-parametric statistical tests may be needed in further studies.

Conclusion: The currently provided pulmonary function AI technology is widely accepted by healthcare professionals but healthcare practitioners are not aware of AI in pulmonary function analysis and only the perceived ease of use and perceived usefulness are not enough to drive AI adoption in the healthcare profession. The need to consider additional factors, including trust in AI, regulatory frameworks, and organizational support is important for future research, as highlighted in the study. By addressing these challenges, AI-powered pulmonary diagnostics can be effectively executed, ultimately enhancing disease detection and Uber, patient outcomes, and clinical decision-making.