Real-Time Typical Urodynamic Signal Recognition System Using Deep Learning.

IF 1.8 3区医学 Q3 UROLOGY & NEPHROLOGY

International Neurourology Journal Pub Date : 2025-03-01 Epub Date: 2025-03-31 DOI:10.5213/inj.2448430.215

Xin Liu, Ping Zhong, Di Chen, Limin Liao

{"title":"Real-Time Typical Urodynamic Signal Recognition System Using Deep Learning.","authors":"Xin Liu, Ping Zhong, Di Chen, Limin Liao","doi":"10.5213/inj.2448430.215","DOIUrl":null,"url":null,"abstract":"Purpose: Gold-standard urodynamic examination is widely used in the diagnosis and treatment of lower urinary tract dysfunction. The purpose of urodynamic quality control is to standardize urodynamic examination and ensure its clinical reference value. In our study, we attempted to use a deep learning (DL) algorithm model, mainly for the recognition of typical urodynamic signal, to help physicians complete high-quality urodynamic examinations.Methods: Urodynamic image data from 2 cohorts of adult patients with neurogenic bladder were used: (1) 300 patients with neurogenic bladder in our center from 2012 to 2018 (1,960 images used to train and validate the DL model); and (2) 100 patients with neurogenic bladder from 2020 to 2021 (695 images used to test the performance of the DL model). This resulted in a total of 2,655 images to train, validate and test the DL algorithm to predict the urdynamic signals.Results: Yolov5l had the best detection performance and the highest comprehensive index score (F1, 0.81; mean average precision, 0.83). Our study is a retrospective single-center study, and the generalization ability of the model has not been verified.Conclusion: DL algorithms can help operators identify typical urodynamic signals in real time, improve the interpretation and quality of urodynamic examination, and benefit patients.","PeriodicalId":14466,"journal":{"name":"International Neurourology Journal","volume":"29 1","pages":"40-47"},"PeriodicalIF":1.8000,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12010905/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Neurourology Journal","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.5213/inj.2448430.215","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/3/31 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"UROLOGY & NEPHROLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Purpose: Gold-standard urodynamic examination is widely used in the diagnosis and treatment of lower urinary tract dysfunction. The purpose of urodynamic quality control is to standardize urodynamic examination and ensure its clinical reference value. In our study, we attempted to use a deep learning (DL) algorithm model, mainly for the recognition of typical urodynamic signal, to help physicians complete high-quality urodynamic examinations.

Methods: Urodynamic image data from 2 cohorts of adult patients with neurogenic bladder were used: (1) 300 patients with neurogenic bladder in our center from 2012 to 2018 (1,960 images used to train and validate the DL model); and (2) 100 patients with neurogenic bladder from 2020 to 2021 (695 images used to test the performance of the DL model). This resulted in a total of 2,655 images to train, validate and test the DL algorithm to predict the urdynamic signals.

Results: Yolov5l had the best detection performance and the highest comprehensive index score (F1, 0.81; mean average precision, 0.83). Our study is a retrospective single-center study, and the generalization ability of the model has not been verified.

Conclusion: DL algorithms can help operators identify typical urodynamic signals in real time, improve the interpretation and quality of urodynamic examination, and benefit patients.

查看原文本刊更多论文

基于深度学习的实时典型尿动信号识别系统。

目的：金标准尿动力学检查广泛应用于下尿路功能障碍的诊断和治疗。尿动力学质量控制的目的是规范尿动力学检查，保证尿动力学检查的临床参考价值。在我们的研究中，我们尝试使用深度学习（DL）算法模型，主要用于识别典型的尿动力学信号，以帮助医生完成高质量的尿动力学检查。方法：采用两组成年神经源性膀胱患者尿动力学图像数据：(1)2012 - 2018年我院收治的300例神经源性膀胱患者（1960张图像用于DL模型的训练和验证）；(2) 2020 - 2021年100例神经源性膀胱患者（695张图像用于测试DL模型的性能）。这导致总共2655张图像需要训练、验证和测试深度学习算法来预测非动态信号。结果：Yolov5l的检测性能最好，综合指标得分最高(F1, 0.81；平均精密度0.83)。本研究为回顾性单中心研究，模型的泛化能力尚未得到验证。结论：深度学习算法可以帮助操作者实时识别典型尿动力学信号，提高尿动力学检查的解释和质量，使患者受益。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

International Neurourology Journal UROLOGY & NEPHROLOGY-

CiteScore

4.40

自引率

21.70%

发文量

审稿时长

4 weeks

期刊介绍： The International Neurourology Journal (Int Neurourol J, INJ) is a quarterly international journal that publishes high-quality research papers that provide the most significant and promising achievements in the fields of clinical neurourology and fundamental science. Specifically, fundamental science includes the most influential research papers from all fields of science and technology, revolutionizing what physicians and researchers practicing the art of neurourology worldwide know. Thus, we welcome valuable basic research articles to introduce cutting-edge translational research of fundamental sciences to clinical neurourology. In the editorials, urologists will present their perspectives on these articles. The original mission statement of the INJ was published on October 12, 1997. INJ provides authors a fast review of their work and makes a decision in an average of three to four weeks of receiving submissions. If accepted, articles are posted online in fully citable form. Supplementary issues will be published interim to quarterlies, as necessary, to fully allow berth to accept and publish relevant articles.