MTTM: Memory-Augmented with Mamba for 3D Medical Images Analysis
Hongkai Wei, Yang Yang, Shijie Sun, Huansheng Song, Keyu Guo, Yongfeng Bu
Abstract
The rapid advancement of artificial intelligence has propelled the healthcare industry into a new era of diagnostic precision. A pivotal component of this evolution is the accurate classification of 3D medical images, which necessitates extracting robust feature representations capable of effectively modeling long-range dependencies within the data. This paper introduces the Mamba Token Turing Machine (MTTM), a novel architecture that integrates the efficiency of Mamba with the memory mechanisms of the Token Turing Machine (TTM), effectively addressing limitations of Transformers in long-range dependency modeling. MTTM’s Memory-Augmented Processing Unit (MAPU) employs four blending methods, achieving state-of-the-art accuracy and efficiency on the MedMNIST v2 dataset, thereby advancing diagnostic precision in 3D medical image analysis. The code is available at https://github.com/hongkai-wei/MTTM.
BibTeX
@inproceedings{icassp2025_mttmmemoryaugmen,
title = {MTTM: Memory-Augmented with Mamba for 3D Medical Images Analysis},
author = {Hongkai Wei and Yang Yang and Shijie Sun and Huansheng Song and Keyu Guo and Yongfeng Bu},
booktitle = {ICASSP 2025},
year = {2025}
}