{"id":64216,"date":"2026-01-14T10:52:29","date_gmt":"2026-01-14T10:52:29","guid":{"rendered":"https:\/\/youzum.net\/google-ai-releases-medgemma-1-5-the-latest-update-to-their-open-medical-ai-models-for-developers\/"},"modified":"2026-01-14T10:52:29","modified_gmt":"2026-01-14T10:52:29","slug":"google-ai-releases-medgemma-1-5-the-latest-update-to-their-open-medical-ai-models-for-developers","status":"publish","type":"post","link":"https:\/\/youzum.net\/ja\/google-ai-releases-medgemma-1-5-the-latest-update-to-their-open-medical-ai-models-for-developers\/","title":{"rendered":"Google AI Releases MedGemma-1.5: The Latest Update to their Open Medical AI Models for Developers"},"content":{"rendered":"

Google Research has expanded its Health AI Developer Foundations program (HAI-DEF)<\/a> with the release of MedGemma-1.5<\/a>. The model is released as open starting points for developers who want to build medical imaging, text and speech systems and then adapt them to local workflows and regulations.<\/p>\n

\n
\"\"
https:\/\/research.google\/blog\/next-generation-medical-image-interpretation-with-medgemma-15-and-medical-speech-to-text-with-medasr\/<\/figcaption><\/figure>\n<\/div>\n

MedGemma 1.5, small multimodal model for real clinical data<\/strong><\/h3>\n

MedGemma is a family of medical generative models built on Gemma. The new release, MedGemma-1.5-4B, targets developers who need a compact model that can still handle real clinical data. The previous MedGemma-1-27B model remains available for more demanding text heavy use cases.<\/p>\n

MedGemma-1.5-4B is multimodal. It accepts text, two dimensional images, high dimensional volumes and whole slide pathology images. The model is part of the Health AI Developer Foundations program so it is intended as a base to fine tune, not a ready made diagnostic device.<\/p>\n

\n
\"\"
https:\/\/research.google\/blog\/next-generation-medical-image-interpretation-with-medgemma-15-and-medical-speech-to-text-with-medasr\/<\/figcaption><\/figure>\n<\/div>\n

Support for high dimensional CT, MRI and pathology<\/strong><\/h3>\n

A major change in MedGemma-1.5 is support for high dimensional imaging. The model can process three dimensional CT and MRI volumes as sets of slices together with a natural language prompt. It can also process large histopathology slides by working over patches extracted from the slide.<\/p>\n

On internal benchmarks, MedGemma-1.5 improves disease related CT findings from 58% to 61% accuracy and MRI disease findings from 51% to 65% accuracy when averaged over findings. For histopathology, the ROUGE L score on single slide cases increases from 0.02 to 0.49. This matches the 0.498 ROUGE L score of the task specific PolyPath model.<\/p>\n

\n
\"\"
https:\/\/research.google\/blog\/next-generation-medical-image-interpretation-with-medgemma-15-and-medical-speech-to-text-with-medasr\/<\/figcaption><\/figure>\n<\/div>\n

Imaging and report extraction benchmarks<\/strong><\/h3>\n

MedGemma-1.5 also improves several benchmarks that are closer to production workflows.<\/p>\n

On the Chest ImaGenome benchmark for anatomical localization in chest X rays, it improves intersection over union from 3% to 38%. On the MS-CXR-T benchmark for longitudinal chest X-ray comparison, macro-accuracy increases from 61% to 66%.<\/p>\n

Across internal single image benchmarks that cover chest radiography, dermatology, histopathology and ophthalmology, average accuracy goes from 59% to 62%t. These are simple single image tasks, useful as sanity checks during domain adaptation.<\/p>\n

MedGemma-1.5 also targets document extraction. On medical laboratory reports, the model improves macro F1 from 60% to 78% when extracting lab type, value and units. For developers this means less custom rule based parsing for semi structured PDF or text reports.<\/p>\n

Applications deployed on Google Cloud can now work directly with DICOM, which is the standard file format used in radiology. This removes the need for a custom preprocessor for many hospital systems.<\/p>\n

\n
\"\"
https:\/\/research.google\/blog\/next-generation-medical-image-interpretation-with-medgemma-15-and-medical-speech-to-text-with-medasr\/<\/figcaption><\/figure>\n<\/div>\n

Medical text reasoning with MedQA and EHRQA<\/strong><\/h3>\n

MedGemma-1.5 is not only an imaging model. It also improves baseline performance on medical text tasks.<\/p>\n

On MedQA, a multiple choice benchmark for medical question answering, the 4B model improves accuracy from 64% to 69% relative to the previous MedGemma-1. On EHRQA, a text based electronic health record question answering benchmark, accuracy increases from 68% to 90%.<\/p>\n

These numbers matter if you plan to use MedGemma-1.5 as a backbone for tools such as chart summarization, guideline grounding or retrieval augmented generation over clinical notes. The 4B size keeps fine tuning and serving cost at a practical level.<\/p>\n

MedASR, a domain tuned speech recognition model<\/strong><\/h3>\n

Clinical workflows contain a large amount of dictated speech. MedASR is the new medical automated speech recognition model released together with MedGemma-1.5.<\/p>\n

MedASR uses a Conformer based architecture that is pre trained and fine tuned for clinical audio. It targets tasks such as chest X-ray dictation, radiology reports and general medical notes. The model is available through the same Health AI Developer Foundations channel on Vertex AI and on Hugging Face.<\/p>\n

In evaluations against Whisper-large-v3, a general ASR model, MedASR reduces word error rate for chest X-ray dictation from 12.5% to 5.2%. That corresponds to 58% fewer transcription errors. On a broader internal medical dictation benchmark, MedASR reaches 5.2% word error rate while Whisper-large-v3 has 28.2%, which corresponds to 82% fewer errors.<\/p>\n

Key Takeaways<\/strong><\/h3>\n