A brain-based AI test could point to the best antidepressant for you - Silicon Canals

"Before treatment began, participants underwent neuroimaging. Instead of relying on a single modality, the researchers fused structural connectivity (how regions are physically wired) with functional connectivity (how regions co-activate at rest). The goal was not to throw every possible feature at a black box, but to learn a constrained pattern-what the authors call structure-function "covariation"-that carries the most predictive signal for outcome. In other words, the model tries to find the smallest set of connections that meaningfully forecasts symptom change."

"A new line of research suggests that baseline brain scans-read by a transparent, carefully trained AI model-may help doctors predict which common antidepressants are most likely to work for a given person, while also separating true drug effects from placebo lift. The study drawing attention to this possibility appears in Nature Mental Health and centers on a multimodal "brain signature" of response built from both brain structure and resting connectivity. To sum up the clinical promise plainly: machine learning can forecast individual responses to two widely prescribed SSRIs and to placebo, and it does so with a model designed to be interpretable by clinicians. If this sounds like an antidote to trial-and-error care, that's precisely why the study matters."

Baseline neuroimaging that combines structural connectivity and resting-state functional connectivity yields a compact structure-function covariation signature predictive of antidepressant response. Adults with major depressive disorder randomized to sertraline, escitalopram, or placebo underwent pre-treatment scans. A constrained, interpretable machine learning model identified a minimal set of connections carrying the most predictive signal for symptom change. The model achieved individual-level forecasts for sertraline and for placebo in the primary dataset and separates specific drug effects from placebo-related improvement. The approach emphasizes clinician interpretability and aims to reduce trial-and-error prescribing.

Read at Silicon Canals