Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a treatment-conditioned diffusion framework to forecast neurodegenerative disease progression by generating high-fidelity future brain states. The diffusion process is conditioned on initial DaTscan images and longitudinal levodopa dosage using a Transformer encoder to model pharmacological dynamics. Results demonstrate improved clinical fidelity and anatomical detail compared to baselines, evidenced by a 14% reduction in MSE and a 4.9% increase in SSIM.
Generating future brain scans conditioned on treatment could drastically improve personalized therapeutic interventions for neurodegenerative diseases.
Forecasting the progression of neurodegenerative diseases, such as Parkinson's disease, is essential for effective long-term planning and personalized therapeutic intervention. Existing systems typically produce scalar clinical scores that ignore the rich structure of longitudinal neuroimaging, while traditional generative approaches suffer from a loss of anatomical details and blurring subtle progression patterns. To address this, we introduce a novel treatment-conditioned diffusion framework that predicts high-fidelity future brain states by conditioning the generative process on patients'screening DaTscan images and levodopa equivalent daily dose over one year. The pipeline uses a Transformer-based encoder to represent non-linear, time-dependent pharmacological dynamics and optimizes generation through a multi-weight region-of-interest mask that focuses on biologically critical areas. Experimental evaluation shows that our framework maintains sharp anatomical boundaries and significantly improves clinical fidelity relative to the baseline, achieving 14.0% lower MSE, 7.2% lower MAE, and 4.9% higher SSIM.