Intra-Session recurrence in intradialytic hypotension prediction: evaluation implications and recurrence-aware...

Real-time prediction of intradialytic hypotension (IDH) using deep learning has shown high accuracy; however, existing models typically treat all IDH events equally, without distinguishing between initial and recurrent occurrences within a dialysis session. This conventional approach neglects the distinct underlying physiological mechanisms and clinical intervention requirements of each occurrence type. This study aimed to systematically examine IDH recurrence patterns to evaluate their impact on model performance and to identify methodological improvements.

We retrospectively analyzed 12,767 hemodialysis sessions from 66 patients. Recurrent IDH was defined as an event occurring ≥30 min after the initial IDH. Deep learning models, including ConvMixer, temporal convolutional network, and long short-term memory with attention, were compared with a rule-based naïve baseline that predicted IDH solely from prior occurrence. Modeling strategies explicitly incorporating recurrence information were implemented. Model robustness across systolic blood pressure subgroups was evaluated and enhanced using adversarial training.

The probability of IDH increased markedly from 0.7-10.4% before initial events to 11.7-65.7% thereafter. Conventional evaluation that aggregated all IDH events overestimated performance, with particularly large gaps in F1 score and AUPRC between initial and recurrent IDH predictions. The naïve baseline achieved an area under the receiver operating characteristic (AUROC) curve of 0.798 without training, highlighting the strong influence of recurrence patterns on predictive performance. Incorporating recurrence-aware features and loss weighting improved AUROC by up to 8.0 percentage points across architectures. Adversarial training further reduced subgroup disparities while preserving overall model performance.

Incorporating recurrence patterns into IDH prediction models demonstrated improvements in accuracy, robustness, and comparability across studies. We recommend standardized evaluation protocols that explicitly account for recurrence to enhance clinical applicability and reliability.