相关实验视频

Updated: Jun 17, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

PolypNextLSTM:使用ConvNext和ConvLSTM的轻量级和快速的多片视频细分网络

Debayan Bhattacharya¹, Konrad Reuter², Finn Behrendt¹

¹Institute of Medical Technology and Intelligent Systems, Technische Universitaet Hamburg, Hamburg, Germany.

International journal of computer assisted radiology and surgery

|August 8, 2024

概括

一个新的深度学习模型PolypNextLSTM通过使用视频数据来获得时间洞察力来增强聚细分. 它以最小的参数实现卓越的性能,在具有挑战性的场景中超越现有方法.

科学领域:

医疗成像医学成像
人工智能的人工智能
计算机视觉计算机视觉

背景情况:

当前的多片细分模型经常使用单个图像,错过了视频中可用的关键时间信息.
临床诊断的息肉受益于观察时间动态,单图像模型不能复制.

研究的目的:

开发 PolypNextLSTM,一种基于视频的深度学习模型,用于聚合物细分.
利用时间信息来提高细分精度和效率.
为了创建一个最小参数的模型,适合边缘设备.

主要方法:

波利普NextLSTM使用了一个类似于UNet的架构,采用了经过修剪的ConvNext-Tiny骨干.
为了有效的时间特征融合,采用了卷积长短期记忆 (ConvLSTM) 模块.
该模型在SUN-SEG数据集上进行了评估,包括具有艺术品的具有挑战性的案例.

主要成果:

在难以检测的息肉上,PolypNextLSTM获得了0.7898的子得分,超过了最先进的模型.
该模型在视频中表现出卓越的表现,其中包含诸如幽灵和遮蔽等文物.
波利普NextLSTM显示了每秒最高的,表明了高效的处理.

结论:

PolypNextLSTM有效地集成时间信息,用于优质的多片细分.
该模型提供了高性能和计算效率的平衡.
它的精益架构使其成为实时应用程序和边缘设备的有希望的候选人.

关键词:

在美国,CNN是CNN.一个多的多.分段化分段化分段化分段化视频视频视频视频视频

更多相关视频

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

相关概念视频

Insensitive Nuclei Enhanced by Polarization Transfer (INEPT)

Insensitive Nuclei Enhanced by Polarization Transfer (INEPT)

Insensitive Nuclei Enhanced by Polarization Transfer (INEPT) is an advanced Nuclear Magnetic Resonance (NMR) technique specifically designed to detect and enhance the signals of low-abundance nuclei, such as carbon-13 and nitrogen-15, in small molecules. The fundamental principle behind INEPT is the transfer of polarization from a more abundant and highly polarizable nucleus, typically hydrogen-1, to the low-abundance nucleus of interest. This process effectively boosts the NMR signal of the...