By research.google
Publication Date: 2026-01-13 12:00:00
Improved performance for medical imaging use cases
MedGemma was designed from the ground up as a multimodal model that reflects the multimodal nature of medicine. MedGemma 1 included support for the interpretation of two-dimensional medical images, including chest radiographs, dermatology images, fundus images, and histopathology patches.
With MedGemma 1.5, we expand support for high-dimensional medical imaging, starting with three-dimensional volume representations of CT imaging and MRI, as well as whole-body histopathological imaging. Developers can create applications in which multiple slices (for CT or MRI) or multiple patches (for histopathology) are provided as input along with a prompt describing the task.
In internal benchmarks, the absolute baseline accuracy of MedGemma 1.5 improved by 3% (61% vs. 58%) over MedGemma 1 in classifying disease-related CT findings and by 14% (65% vs. 51%)…
