Unsupervised domain transfer: Overcoming signal degradation in sleep monitoring by increasing scoring realism
For researchers in mobile sleep monitoring, this work presents an incremental approach to handling signal degradation without labeled target data, though it requires further development for practical use.
The paper investigates using hypnogram realism to guide unsupervised domain transfer for handling signal degradation in mobile sleep monitoring. The method improves Cohen's kappa by 0.03 to 0.29 depending on distortion type, but does not reach theoretical optimal performance and shows insignificant benefit on real-life domain mismatch.
Objective: Investigate whether hypnogram 'realism' can be used to guide an unsupervised method for handling arbitrary types of signal degradation in mobile sleep monitoring. Approach: Combining a pretrained, state-of-the-art 'u-sleep' model with a 'discriminator' network, we align features from a target domain with a feature space learned during pretraining. To test the approach, we distort the source domain with realistic signal degradations, to see how well the method can adapt to different types of degradation. We compare the performance of the resulting model with best-case models designed in a supervised manner for each type of transfer. Main Results: Depending on the type of distortion, we find that the unsupervised approach can increase Cohen's kappa with as little as 0.03 and up to 0.29, and that for all transfers, the method does not decrease performance. However, the approach never quite reaches the estimated theoretical optimal performance, and when tested on a real-life domain mismatch between two sleep studies, the benefit was insignificant. Significance: 'Discriminator-guided fine tuning' is an interesting approach to handling signal degradation for 'in the wild' sleep monitoring, with some promise. In particular, what it says about sleep data in general is interesting. However, more development will be necessary before using it 'in production'.