UDAMA: Unsupervised Domain Adaptation through Multi-discriminator Adversarial Training with Noisy Labels Improves Cardio-fitness Prediction
This work addresses the challenge of generalizing healthcare models to unseen scenarios by leveraging noisy labeled data, which is incremental as it builds on existing domain adaptation techniques.
The paper tackles the problem of predicting cardio-respiratory fitness using small-scale high-quality labeled data and large-scale noisy labeled data by introducing UDAMA, an unsupervised domain adaptation method with multi-discriminator adversarial training, achieving up to 12% performance improvement over state-of-the-art models.
Deep learning models have shown great promise in various healthcare monitoring applications. However, most healthcare datasets with high-quality (gold-standard) labels are small-scale, as directly collecting ground truth is often costly and time-consuming. As a result, models developed and validated on small-scale datasets often suffer from overfitting and do not generalize well to unseen scenarios. At the same time, large amounts of imprecise (silver-standard) labeled data, annotated by approximate methods with the help of modern wearables and in the absence of ground truth validation, are starting to emerge. However, due to measurement differences, this data displays significant label distribution shifts, which motivates the use of domain adaptation. To this end, we introduce UDAMA, a method with two key components: Unsupervised Domain Adaptation and Multidiscriminator Adversarial Training, where we pre-train on the silver-standard data and employ adversarial adaptation with the gold-standard data along with two domain discriminators. In particular, we showcase the practical potential of UDAMA by applying it to Cardio-respiratory fitness (CRF) prediction. CRF is a crucial determinant of metabolic disease and mortality, and it presents labels with various levels of noise (goldand silver-standard), making it challenging to establish an accurate prediction model. Our results show promising performance by alleviating distribution shifts in various label shift settings. Additionally, by using data from two free-living cohort studies (Fenland and BBVS), we show that UDAMA consistently outperforms up to 12% compared to competitive transfer learning and state-of-the-art domain adaptation models, paving the way for leveraging noisy labeled data to improve fitness estimation at scale.