相关实验视频

Updated: Jan 10, 2026

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

基于监督分支和不确定性估计的膀瘤的半监督医疗图像细分

Menghan Li^1,2, Zhengping Li^3,4, Chao Xu^1,2

¹School of Integrated Circuits, Anhui University, Hefei, 230601, Anhui, China.

Scientific reports

|November 25, 2025

概括

本研究介绍了UDS-MT,这是一种使用有限数据进行膀瘤细分的半监督方法. 它通过将平均教师网络与指导分支和新浪损失术语相结合来实现高准确度.

科学领域:

医疗成像医学成像
人工智能的人工智能
在瘤学瘤学.

背景情况:

监督学习在膀瘤细分方面表现出色,但需要大量的标记数据.
有限的标记数据对实现准确的细分结果构成重大挑战.

研究的目的:

使用有限的标记数据,开发用于膀瘤的半监督细分方法 (UDS-MT).
为了提高膀瘤细分的准确性和可靠性.

主要方法:

对于参数更新,UDS-MT采用了一个具有指数移动平均 (EMA) 机制的平均教师网络.
一个指导分支利用不确定性估计来过高可信度像素,生成可靠的面具并减轻过拟合.
一个新的防御损失术语只计算高信心预测的损失,提高伪标签可靠性.

主要成果:

通过UDS-MT方法,在只有15%的标记数据下,膀瘤细分的Dice系数达到80.04%.
在有限的数据条件下,这种性能至少比现有方法高2.81%.

结论:

拟议的UDS-MT方法有效地解决了膀瘤细分中的有限标记数据的挑战.
与其他方法相比,它表现出卓越的性能和可靠性,为临床应用提供了有前途的解决方案.

更多相关视频

Magnetic Resonance Imaging Assessment of Carcinogen-induced Murine Bladder Tumors

Magnetic Resonance Imaging Assessment of Carcinogen-induced Murine Bladder Tumors

Published on: March 29, 2019

A Murine Orthotopic Bladder Tumor Model and Tumor Detection System

A Murine Orthotopic Bladder Tumor Model and Tumor Detection System

Published on: January 12, 2017

相关概念视频

Imaging Studies VI: Voiding Cystourethrography and Cystography

Imaging Studies VI: Voiding Cystourethrography and Cystography

Voiding Cystourethrography (VCUG) and Cystography are specialized radiographic procedures used to examine the structure and function of the bladder and urethra.Voiding Cystourethrography (VCUG)A Voiding Cystourethrogram (VCUG) is a diagnostic imaging procedure that assesses the anatomy and function of the lower urinary tract. It focuses on the bladder, bladder neck, and urethra, helping detect abnormalities such as vesicoureteral reflux (VUR)—the backward or reverse flow of urine into the...