Authors: Haifan Gong, Luoyao Kang, Yitao Wang, Xiang Wan, Haofeng Li

Published on: February 05, 2024

Impact Score: 8.3

Arxiv code: Arxiv:2402.03526

Summary

• What is new: Introduction of nnMamba, integrating CNNs with SSMs for biomedical imaging.
• Why this is important: Struggle of traditional CNNs with locality and Transformers’ heavy load in 3D medical imaging.
• What the research proposes: nnMamba, combining CNNs for local feature extraction and SSMs for long-range dependency modeling.
• Results: Superior performance in 3D image segmentation, classification, and landmark detection compared to existing methods.

Technical Details

Technological frameworks used: nnMamba architecture

Models used: CNNs, SSMs

Data used: High-dimensional medical images

Potential Impact

Healthcare imaging companies, medical diagnostics firms

Want to implement this idea in a business?

We have generated a startup concept here: MediVisonAI.