Search papers, labs, and topics across Lattice.
VDv2 (Golhar et al., 2025), a high-definition phantom-based colonoscopy video dataset designed for
3
0
4
Uncertainty-aware reconstruction using a Gaussian process prior leads to a 21.9% improvement in endoscopic video restoration, enhancing clinical interpretability.
Automated segmentation of radiological Peritoneal Cancer Index (rPCI) regions from CT scans is now feasible, potentially replacing invasive surgical assessment for peritoneal metastases.
Current CADe systems excel at identifying early neoplasia in Barrett's esophagus on balanced datasets, but their clinical utility plummets in real-world, low-prevalence settings, exposing a critical gap in prevalence-aware design.
