Workshops Archive

Fine-Tuning Foundation Models for Micro-CT Image Segmentation Using HPC Platforms


Workshop: XLOOP 2026: The 8th Annual Workshop on Extreme-Scale Experiment-in-the-Loop Computing

Authors: Xiaoya Chong, David Abramov, Pradyumna Elavarthi, Elizabeth Clark, Petrus Zwart, Wiebke Koepp, Dilworth Parkinson, and Dylan McReynolds (Lawrence Berkeley National Laboratory); Daniel Allan (Brookhaven National Laboratory); Johanna Weker (SLAC National Accelerator Laboratory); Nicholas Schwarz and Francesco De Carlo (Argonne National Laboratory); Nghia Vo (Brookhaven National Laboratory); Sharif Ahmed and Tim Snow (Diamond Light Source); Tamas Varga (Pacific Northwest National Laboratory); Tyler Farr and Harry Lisabeth (Lawrence Berkeley National Laboratory); Andrew McElrone (University of California, Davis); and Tanny Chavez and Alexander Hexemer (Lawrence Berkeley National Laboratory)

Abstract: Foundation models have demonstrated exceptional performance in computer vision, but adapting them to scientific imaging requires domain-specific datasets, scalable training methodologies, and high-performance computing (HPC) resources. This work presents methodologies for fine-tuning and deploying foundation models for synchrotron X-ray micro-computed tomography (micro-CT) image segmentation. We describe the adaptation of SAM3 for instance segmentation and DINOv3-Seg for semantic segmentation. We present strategies for data preprocessing, model configuration, loss design, and hyperparameter optimization for fine-tuning these models. We further describe scalable fine-tuning on local GPU servers through a web-based training interface and on the National Energy Research Scientific Computing Center (NERSC) through Slurm-based job submission. Finally, we present distributed inference strategies on NERSC and the Argonne Leadership Computing Facility (ALCF), enabling near real-time segmentation of large synchrotron datasets across hundreds of GPUs. Together, these methods provide guidance for adapting, training, and deploying foundation models for large-scale scientific image segmentation at synchrotron facilities.


Back to XLOOP 2026: The 8th Annual Workshop on Extreme-Scale Experiment-in-the-Loop Computing Archive Listing

Back to Full Workshop Archive Listing