Abstract
Image-to-image translation has emerged as a powerful technique in medicalimaging, enabling tasks such as image denoising and cross-modality conversion.However, it suffers from limitations in handling out-of-distribution sampleswithout causing performance degradation. To address this limitation, we proposea novel Test-Time Adaptation (TTA) framework that dynamically adjusts thetranslation process based on the characteristics of each test sample. Ourmethod introduces a Reconstruction Module to quantify the domain shift and aDynamic Adaptation Block that selectively modifies the internal features of apretrained translation model to mitigate the shift without compromising theperformance on in-distribution samples that do not require adaptation. Weevaluate our approach on two medical image-to-image translation tasks: low-doseCT denoising and T1 to T2 MRI translation, showing consistent improvements overboth the baseline translation model without TTA and prior TTA methods. Ouranalysis highlights the limitations of the state-of-the-art that uniformlyapply the adaptation to both out-of-distribution and in-distribution samples,demonstrating that dynamic, sample-specific adjustment offers a promising pathto improve model resilience in real-world scenarios. The code is available at:https://github.com/cosbidev/Sample-Aware_TTA.