Jeremiah D. Deng

This study uses domain adaptation methods to reduce cross-ethnic prediction errors in neuroimaging-based cognitive prediction using the ABCD dataset.

The paper benchmarks ethnic bias in MRI-based cognitive prediction models and finds that balanced sampling reduces disparities without accuracy loss, with structural MRI showing greater bias than task-based fMRI.

This paper introduces a stacking machine learning approach to improve predictability, test-retest reliability, and generalisability of brain-wide associations for cognitive abilities using multimodal MRI data.

This paper shows that integrating task-based fMRI across tasks improves prediction and test-retest reliability of cognitive abilities compared to non-task MRI modalities.