Test-Time Calibration: A Framework for Personalized Test-Time Adaptation in Real-World Biosignals

Yong-Yeon Jo, Byeong Tak Lee, Jeong-Ho Hong, Hak Seung Lee, Joon-myoung Kwon, Beom Joon Kim
Proceedings of the sixth Conference on Health, Inference, and Learning, PMLR 287:381-394, 2025.

Abstract

Test-Time Adaptation (TTA) methods have been widely used to enhance model robustness by continuously updating pre-trained models with unlabeled target data. However, in real-world biosignal applications-where factors such as age, lifestyle, and comorbidities induce significant variability–traditional TTA often falls short in addressing personalization needs. To satisfy such needs, we introduce a novel Test-Time Calibration (TTC) framework that integrates continuous self-supervised adaptation on unlabeled samples with periodic supervised calibration using the sporadically available ground-truth labels. Our approach leverages a model equipped with dual heads for supervised learning (SL) and self-supervised learning (SSL), and further incorporates a dual buffer along with a weighted batch sampling strategy to effectively manage and utilize both data types during the test phase. We evaluate our framework on two distinct datasets: the publicly available PulseDB, a benchmark for cuff-less blood pressure estimation, and a private ICU dataset collected from critically ill patients. Experimental results demonstrate that our approach improves blood pressure prediction accuracy and robustness, highlighting its suitability for dynamic, personalized biosignal applications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v287-jo25a, title = {Test-Time Calibration: A Framework for Personalized Test-Time Adaptation in Real-World Biosignals}, author = {Jo, Yong-Yeon and Lee, Byeong Tak and Hong, Jeong-Ho and Lee, Hak Seung and Kwon, Joon-myoung and Kim, Beom Joon}, booktitle = {Proceedings of the sixth Conference on Health, Inference, and Learning}, pages = {381--394}, year = {2025}, editor = {Xu, Xuhai Orson and Choi, Edward and Singhal, Pankhuri and Gerych, Walter and Tang, Shengpu and Agrawal, Monica and Subbaswamy, Adarsh and Sizikova, Elena and Dunn, Jessilyn and Daneshjou, Roxana and Sarker, Tasmie and McDermott, Matthew and Chen, Irene}, volume = {287}, series = {Proceedings of Machine Learning Research}, month = {25--27 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v287/main/assets/jo25a/jo25a.pdf}, url = {https://proceedings.mlr.press/v287/jo25a.html}, abstract = {Test-Time Adaptation (TTA) methods have been widely used to enhance model robustness by continuously updating pre-trained models with unlabeled target data. However, in real-world biosignal applications-where factors such as age, lifestyle, and comorbidities induce significant variability–traditional TTA often falls short in addressing personalization needs. To satisfy such needs, we introduce a novel Test-Time Calibration (TTC) framework that integrates continuous self-supervised adaptation on unlabeled samples with periodic supervised calibration using the sporadically available ground-truth labels. Our approach leverages a model equipped with dual heads for supervised learning (SL) and self-supervised learning (SSL), and further incorporates a dual buffer along with a weighted batch sampling strategy to effectively manage and utilize both data types during the test phase. We evaluate our framework on two distinct datasets: the publicly available PulseDB, a benchmark for cuff-less blood pressure estimation, and a private ICU dataset collected from critically ill patients. Experimental results demonstrate that our approach improves blood pressure prediction accuracy and robustness, highlighting its suitability for dynamic, personalized biosignal applications.} }
Endnote
%0 Conference Paper %T Test-Time Calibration: A Framework for Personalized Test-Time Adaptation in Real-World Biosignals %A Yong-Yeon Jo %A Byeong Tak Lee %A Jeong-Ho Hong %A Hak Seung Lee %A Joon-myoung Kwon %A Beom Joon Kim %B Proceedings of the sixth Conference on Health, Inference, and Learning %C Proceedings of Machine Learning Research %D 2025 %E Xuhai Orson Xu %E Edward Choi %E Pankhuri Singhal %E Walter Gerych %E Shengpu Tang %E Monica Agrawal %E Adarsh Subbaswamy %E Elena Sizikova %E Jessilyn Dunn %E Roxana Daneshjou %E Tasmie Sarker %E Matthew McDermott %E Irene Chen %F pmlr-v287-jo25a %I PMLR %P 381--394 %U https://proceedings.mlr.press/v287/jo25a.html %V 287 %X Test-Time Adaptation (TTA) methods have been widely used to enhance model robustness by continuously updating pre-trained models with unlabeled target data. However, in real-world biosignal applications-where factors such as age, lifestyle, and comorbidities induce significant variability–traditional TTA often falls short in addressing personalization needs. To satisfy such needs, we introduce a novel Test-Time Calibration (TTC) framework that integrates continuous self-supervised adaptation on unlabeled samples with periodic supervised calibration using the sporadically available ground-truth labels. Our approach leverages a model equipped with dual heads for supervised learning (SL) and self-supervised learning (SSL), and further incorporates a dual buffer along with a weighted batch sampling strategy to effectively manage and utilize both data types during the test phase. We evaluate our framework on two distinct datasets: the publicly available PulseDB, a benchmark for cuff-less blood pressure estimation, and a private ICU dataset collected from critically ill patients. Experimental results demonstrate that our approach improves blood pressure prediction accuracy and robustness, highlighting its suitability for dynamic, personalized biosignal applications.
APA
Jo, Y., Lee, B.T., Hong, J., Lee, H.S., Kwon, J. & Kim, B.J.. (2025). Test-Time Calibration: A Framework for Personalized Test-Time Adaptation in Real-World Biosignals. Proceedings of the sixth Conference on Health, Inference, and Learning, in Proceedings of Machine Learning Research 287:381-394 Available from https://proceedings.mlr.press/v287/jo25a.html.

Related Material