LSTM analysis of infant tidal breathing detects bronchopulmonary dysplasia with 97% accuracy

Objective: To test whether machine learning (ML) models trained on tidal breathing flow time series can discriminate between individuals with and without respiratory disease and predict lung function indices obtained from conventional pulmonary function testing. Background: Accurate assessment of respiratory function in infants and young children is challenging because conventional pulmonary function testing requires sophisticated equipment and/or active patient cooperation. Tidal breathing measurements, in contrast, can be obtained non-invasively with little or no patient cooperation and at low cost, yet their clinical utility has been limited. We hypothesized that sufficiently long tidal breathing flow time series contain clinically relevant information that can be extracted using a recurrent neural network known as a long short-term memory (LSTM) network. Approach: We evaluated LSTM models in two scenarios within the Basel-Bern Infant Lung Development cohort. First, we assessed the ability of a model trained on flow and derived volume time series to detect bronchopulmonary dysplasia (BPD) in 329 infants. Second, we examined whether a model trained on tidal breathing flow alone could predict forced expiratory volume in one second (FEV1) in 135 school-age children. Signals were filtered and normalized prior to model training, and performance was evaluated on held-out test datasets. Main results: For BPD detection, the model achieved 97.0% accuracy, 100% specificity, 91.7% sensitivity, 100% precision, and 95.7% F1 score. For FEV1 prediction, Bland-Altman analysis showed a mean bias of -0.009 L (95% CI -0.091 to 0.074), with limits of agreement of -0.416 L and 0.399 L. The mean relative prediction error was 13.7%. Significance: These findings demonstrate that temporal patterns in tidal breathing flow signals contain diagnostically and functionally relevant information. ML applied to tidal breathing measurements may provide a low-burden, minimal-cooperation approach for early respiratory disease detection and functional assessment across early life stages.