Semi-Supervised Learning for Sparsely-Labeled Sequential Data: Application to Healthcare Video Processing

Labeled data is a critical resource for training and evaluating machine learning models. However, many real-life datasets are only partially labeled. We propose a semi-supervised machine learning training strategy to improve event detection performance on sequential data, such as video recordings, when only sparse labels are available, such as event start times without their corresponding end times. Our method uses noisy guesses of the events' end times to train event detection models. Depending on how conservative these guesses are, mislabeled false positives may be introduced into the training set (i.e., negative sequences mislabeled as positives). We further propose a mathematical model for estimating how many inaccurate labels a model is exposed to, based on how noisy the end time guesses are. Finally, we show that neural networks can improve their detection performance by leveraging more training data with less conservative approximations despite the higher proportion of incorrect labels. We adapt sequential versions of MNIST and CIFAR-10 to empirically evaluate our method, and find that our risk-tolerant strategy outperforms conservative estimates by 12 points of mean average precision for MNIST, and 3.5 points for CIFAR. Then, we leverage the proposed training strategy to tackle a real-life application: processing continuous video recordings of epilepsy patients to improve seizure detection, and show that our method outperforms baseline labeling methods by 10 points of average precision.