Temporal Label Smoothing for Early Event Prediction

29 Aug 2022  ·  Hugo Yèche, Alizée Pace, Gunnar Rätsch, Rita Kuznetsova ·

Models that can predict the occurrence of events ahead of time with low false-alarm rates are critical to the acceptance of decision support systems in the medical community. This challenging task is typically treated as a simple binary classification, ignoring temporal dependencies between samples, whereas we propose to exploit this structure. We first introduce a common theoretical framework unifying dynamic survival analysis and early event prediction. Following an analysis of objectives from both fields, we propose Temporal Label Smoothing (TLS), a simpler, yet best-performing method that preserves prediction monotonicity over time. By focusing the objective on areas with a stronger predictive signal, TLS improves performance over all baselines on two large-scale benchmark tasks. Gains are particularly notable along clinically relevant measures, such as event recall at low false-alarm rates. TLS reduces the number of missed events by up to a factor of two over previously used approaches in early event prediction.

PDF Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Respiratory Failure HiRID Temporal Label Smoothing AUPRC 0.604±0.002 # 1
Recall@50 77.0 # 1
Circulatory Failure HiRID Temporal Label Smoothing AUPRC 0.406±0.003 # 1
Recall@50 32.3 # 1
Decompensation MIMIC-III Benchmark Temporal Label Smoothing AUPRC 35.5 # 1
Recall@50 29.3 # 1