Salmon, C. E. G. | bioRxiv Science

bioRxiv2025

Tractography-based transcranial magnetic stimulation prediction using machine learning

BackgroundIdentifying the stimulation brain target is fundamental for transcranial magnetic stimulation (TMS). Currently, this process is time-consuming and heavily dependent on the operators expertise. ObjectiveThis study evaluated a deep learning-based approach to structural brain connectivity for improving stimulation site prediction and real-time cortical excitability mapping during neuronavigation. ResultsTractography-derived connectivity features and distal myographic responses were used to train four neural network models across five subjects. Neural network inputs were incrementally varied using either tractography alone, coil coordinates alone, or hybrid combinations of both using different concatenation regimes. The best performance was observed in two subjects out of 5, where hybrid models integrating coil coordinates and fiber information achieved higher F1 scores and accuracy. ConclusionDespite these promising results, substantial inter-subject variability was observed. This approach may also be applied to other brain regions to investigate connectivity, and the use of machine learning could be extended to functional domains beyond the motor cortex, including sensory and cognitive areas.

neuroscience↗

bioRxiv2024

Beyond functional connectivity: deep learning applied to resting-state fMRI time series in the prediction of 58 human traits in the HCP

Machine learning has made several inroads into the study of brain-behavior relations based on in vivo imaging. While the advent of deep neural networks was expected to further improve predictions, the current literature based on resting-state functional connectivity presents mixed results. We hypothesize that the representation of the data, i.e. in the form of functional connectivity, could restrict an advantage of deep learning techniques, namely that of learning complex representations directly from the data. Thus, we investigated if bypassing this feature extraction resulted in improved performance in the prediction of 58 widely studied behavioral traits from a large sample of Human Connectome Project subjects, using deep learning techniques. For this task, we adapted the InceptionTime architecture, which jointly predicts traits directly from regional time series through representation learning, and compared results with a strong kernel-based baseline. Results revealed that both models achieve comparable performance in most traits. Eleven significant differences in mean squared error were detected, however, with seven favoring the neural network approach, and this number increased when accounting for covariates. We additionally show that contrary to the expectation, the neural network approach was more robust to reductions in the training set size. On the other hand, it was more sensitive to reductions in the length of the time series at test time. Our results present a more nuanced view of the potential of deep learning for the prediction of behavior from neuroimaging, which allows learning features directly from the data.

neuroscience↗