Current work on neural registration architectures: published methods, active benchmarks, and ongoing extensions.
A new architecture that treats the relationship between two observations as something to be measured directly, not guessed at.
Task 1 — PSMAReg: registration of pre-therapy and follow-up whole-body PSMA PET/CT scans for therapy planning, dosimetry, and response assessment.
Task 2 — Learn2Breath: registration of inspiratory and expiratory chest CT scans to enable robust quantification of lung biomechanics and disease progression.
Multimode optical fibers offer ultra-thin, low-cost light delivery for biomedical and scientific imaging but scramble spatial information through mode mixing. Real-time calibration recovers the correspondence between input and output, enabling minimally-invasive endoscopy and deep-tissue imaging at fiber-diameter resolution.