A physics- and spatially informed diffusion model was developed to achieve rapid Monte Carlo–consistent dose prediction for CyberKnife radiotherapy. The model integrates finite-size pencil beam dose distributions, planning CT images, and signed distance maps to incorporate physical, anatomical, and spatial prior information. PSIDMViT improved agreement with Monte Carlo reference doses across head-and-neck, lung, and liver cancer cases. The model reduced the time required for Monte Carlo–consistent dose estimation from approximately one hour to about 18 minutes, supporting more efficient high-precision radiotherapy planning and verification.

- A joint team from KAIST, Gangnam Severance Hospital, and Asan Medical Center is the first to reveal how colorectal cancer progression reshapes the blood’s “metabolic map”

- The new method showed higher predictive power than CEA, a conventional blood biomarker, raising hopes for personalized treatment strategies

- An interdisciplinary idea born from the KAIST Master’s and PhD Venture Research Program led to a world-class research achievement

Biological sex factors into cancer in ways you might not expect. In lung cancer, a form called squamous cell carcinoma disproportionately affects males while adenocarcinoma occurs more frequently in females. Researchers want to know what is happening in our cells to cause these divergent outcomes in lung and other cancers, and how this knowledge can benefit physicians and patients.

AMHERST, Mass. — Bladder cancer, one of the 10 most common cancers, is notoriously difficult to treat, with up to 80% of patients experiencing a post-treatment recurrence. A common practice is to surgically remove the entire bladder, but this leaves the patient incontinent. Now, a research effort led by the University of Massachusetts Amherst has been awarded a five-year grant providing up to $3.45 million from the National Institutes of Health to develop a new way to prevent the cancer’s recurrence, without removing the organ or aggressive repeated treatments.