BATFormer: Towards Boundary-Aware Lightweight Transformer for Efficient Medical Image Segmentation

计算机科学 分割 图像分割 人工智能 变压器 医学影像学 尺度空间分割 计算机视觉 工程类 电气工程 电压
作者
Xian Lin,Li Yu,Kwang-Ting Tim Cheng,Zengqiang Yan
出处
期刊:IEEE Journal of Biomedical and Health Informatics [Institute of Electrical and Electronics Engineers]
卷期号:27 (7): 3501-3512 被引量:91
标识
DOI:10.1109/jbhi.2023.3266977
摘要

OBJECTIVE: Transformers, born to remedy the inadequate receptive fields of CNNs, have drawn explosive attention recently. However, the daunting computational complexity of global representation learning, together with rigid window partitioning, hinders their deployment in medical image segmentation. This work aims to address the above two issues in transformers for better medical image segmentation. METHODS: We propose a boundary-aware lightweight transformer (BATFormer) that can build cross-scale global interaction with lower computational complexity and generate windows flexibly under the guidance of entropy. Specifically, to fully explore the benefits of transformers in long-range dependency establishment, a cross-scale global transformer (CGT) module is introduced to jointly utilize multiple small-scale feature maps for richer global features with lower computational complexity. Given the importance of shape modeling in medical image segmentation, a boundary-aware local transformer (BLT) module is constructed. Different from rigid window partitioning in vanilla transformers which would produce boundary distortion, BLT adopts an adaptive window partitioning scheme under the guidance of entropy for both computational complexity reduction and shape preservation. RESULTS: BATFormer achieves the best performance in Dice of 92.84 %, 91.97 %, 90.26 %, and 96.30 % for the average, right ventricle, myocardium, and left ventricle respectively on the ACDC dataset and the best performance in Dice, IoU, and ACC of 90.76 %, 84.64 %, and 96.76 % respectively on the ISIC 2018 dataset. More importantly, BATFormer requires the least amount of model parameters and the lowest computational complexity compared to the state-of-the-art approaches. CONCLUSION AND SIGNIFICANCE: Our results demonstrate the necessity of developing customized transformers for efficient and better medical image segmentation. We believe the design of BATFormer is inspiring and extendable to other applications/frameworks.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
Lizhenhua完成签到,获得积分10
刚刚
南城发布了新的文献求助10
刚刚
1秒前
快乐行云发布了新的文献求助10
1秒前
初初初啊发布了新的文献求助10
1秒前
李爱国的应助被阿斯顿撒大学采纳,获得10
1秒前
FashionBoy的应助被LDY采纳,获得10
1秒前
2秒前
369ninja发布了新的文献求助10
2秒前
3秒前
3秒前
钱仍城完成签到,获得积分10
3秒前
4秒前
5秒前
junyang完成签到,获得积分10
6秒前
whj发布了新的文献求助10
7秒前
HLYY完成签到,获得积分10
8秒前
科研通AI6.2的应助被刘雨采纳,获得10
9秒前
9秒前
咖啡因重度依赖患者完成签到 ,获得积分10
9秒前
四天垂完成签到 ,获得积分10
9秒前
lx的应助被畅快怜容采纳,获得10
9秒前
爆米花的应助被Aries采纳,获得10
10秒前
科研通AI6.4的应助被舒心采纳,获得10
10秒前
晨阳完成签到,获得积分10
10秒前
11秒前
114514发布了新的文献求助10
11秒前
林深见雾完成签到 ,获得积分10
11秒前
12秒前
阿飞完成签到,获得积分10
14秒前
朴实的电灯胆完成签到,获得积分10
14秒前
西瓜完成签到 ,获得积分10
14秒前
深情安青的应助被loffm采纳,获得10
14秒前
General发布了新的文献求助10
15秒前
372925abc发布了新的文献求助10
16秒前
LDY发布了新的文献求助10
16秒前
16秒前
zww发布了新的文献求助10
16秒前
沈一给沈一的求助进行了留言
17秒前
科研要努力完成签到,获得积分10
17秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
What Else Can I Tell You?: A Pragmatic Study of English Rhetorical Questions as Discursive and Argumentative Acts 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8165074
求助须知:如何正确求助?哪些是违规求助? 9558738
关于积分的说明 21747291
捐赠科研通 7755624
什么是DOI,文献DOI怎么找? 3428780
关于科研通互助平台的介绍 2553162
邀请新用户注册赠送积分活动 2473361