Segment anything in medical images
Abstract Medical image segmentation is a critical component in clinical practice, facilitating accurate diagnosis, treatment planning, and disease monitoring. However, existing methods, often tailored to specific modalities or disease types, lack generalizability across the diverse spectrum of medic...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-44824-z |
| _version_ | 1797276488018427904 |
|---|---|
| author | Jun Ma Yuting He Feifei Li Lin Han Chenyu You Bo Wang |
| author_facet | Jun Ma Yuting He Feifei Li Lin Han Chenyu You Bo Wang |
| author_sort | Jun Ma |
| collection | DOAJ |
| description | Abstract Medical image segmentation is a critical component in clinical practice, facilitating accurate diagnosis, treatment planning, and disease monitoring. However, existing methods, often tailored to specific modalities or disease types, lack generalizability across the diverse spectrum of medical image segmentation tasks. Here we present MedSAM, a foundation model designed for bridging this gap by enabling universal medical image segmentation. The model is developed on a large-scale medical image dataset with 1,570,263 image-mask pairs, covering 10 imaging modalities and over 30 cancer types. We conduct a comprehensive evaluation on 86 internal validation tasks and 60 external validation tasks, demonstrating better accuracy and robustness than modality-wise specialist models. By delivering accurate and efficient segmentation across a wide spectrum of tasks, MedSAM holds significant potential to expedite the evolution of diagnostic tools and the personalization of treatment plans. |
| first_indexed | 2024-03-07T15:28:54Z |
| format | Article |
| id | doaj.art-93ceecdaaa7b4372b936d4b24a2466b7 |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-03-07T15:28:54Z |
| publishDate | 2024-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-93ceecdaaa7b4372b936d4b24a2466b72024-03-05T16:34:14ZengNature PortfolioNature Communications2041-17232024-01-011511910.1038/s41467-024-44824-zSegment anything in medical imagesJun Ma0Yuting He1Feifei Li2Lin Han3Chenyu You4Bo Wang5Peter Munk Cardiac Centre, University Health NetworkDepartment of Computer Science, Western UniversityPeter Munk Cardiac Centre, University Health NetworkTandon School of Engineering, New York UniversityDepartment of Electrical Engineering, Yale UniversityPeter Munk Cardiac Centre, University Health NetworkAbstract Medical image segmentation is a critical component in clinical practice, facilitating accurate diagnosis, treatment planning, and disease monitoring. However, existing methods, often tailored to specific modalities or disease types, lack generalizability across the diverse spectrum of medical image segmentation tasks. Here we present MedSAM, a foundation model designed for bridging this gap by enabling universal medical image segmentation. The model is developed on a large-scale medical image dataset with 1,570,263 image-mask pairs, covering 10 imaging modalities and over 30 cancer types. We conduct a comprehensive evaluation on 86 internal validation tasks and 60 external validation tasks, demonstrating better accuracy and robustness than modality-wise specialist models. By delivering accurate and efficient segmentation across a wide spectrum of tasks, MedSAM holds significant potential to expedite the evolution of diagnostic tools and the personalization of treatment plans.https://doi.org/10.1038/s41467-024-44824-z |
| spellingShingle | Jun Ma Yuting He Feifei Li Lin Han Chenyu You Bo Wang Segment anything in medical images Nature Communications |
| title | Segment anything in medical images |
| title_full | Segment anything in medical images |
| title_fullStr | Segment anything in medical images |
| title_full_unstemmed | Segment anything in medical images |
| title_short | Segment anything in medical images |
| title_sort | segment anything in medical images |
| url | https://doi.org/10.1038/s41467-024-44824-z |
| work_keys_str_mv | AT junma segmentanythinginmedicalimages AT yutinghe segmentanythinginmedicalimages AT feifeili segmentanythinginmedicalimages AT linhan segmentanythinginmedicalimages AT chenyuyou segmentanythinginmedicalimages AT bowang segmentanythinginmedicalimages |