FS-YOLOv9: A Frequency and Spatial Feature-Based YOLOv9 for Real-time Breast Cancer Detection

IF 3.8 2区 医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING Academic Radiology Pub Date : 2025-03-01 DOI:10.1016/j.acra.2024.09.048
Haitian Gui , Tao Su , Xinhua Jiang , Li Li , Lang Xiong , Ji Zhou , Zhiyong Pang
{"title":"FS-YOLOv9: A Frequency and Spatial Feature-Based YOLOv9 for Real-time Breast Cancer Detection","authors":"Haitian Gui ,&nbsp;Tao Su ,&nbsp;Xinhua Jiang ,&nbsp;Li Li ,&nbsp;Lang Xiong ,&nbsp;Ji Zhou ,&nbsp;Zhiyong Pang","doi":"10.1016/j.acra.2024.09.048","DOIUrl":null,"url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>Breast cancer screening is critical for reducing mortality rates. YOLOv9, a new real-time object-detection model, is ideal for cancer screening. A customized YOLOv9 model with enhancements for detecting breast cancer on the basis of species and morphological diversity has potential clinical significance.</div></div><div><h3>Materials and Methods</h3><div>The internal dataset consisted of 687 cases split 3:1 for cross-validation. Additionally, 98 cases from external datasets were used for testing. We developed an FS-YOLOv9 model customized for breast cancer detection that incorporated an extra max-pooling layer before the Conv1 of the Adown to enhance high-brightness features. The Adown of the P3 in the backbone was replaced with a high-frequency Haar wavelet convolution kernel, which ignored the low-frequency components during down-sampling to enhance morphology and texture features. The reliability and robustness of our model was determined by measuring the F1 score, the area under curve of free-response receiver operating characteristic (FAUC), mean average precision (mAP), recall, and precision, and comparing them with the findings for the official YOLOv9, YOLOv8, YOLOv5 models.</div></div><div><h3>Results</h3><div>In comparison with the official YOLOv9 model, FS-YOLOv9 showed a higher average F1 score (0.700 vs. 0.669), FAUC (0.695 vs. 0.662), and mAP50 (0.713 vs. 0.679) in the internal dataset; in the external testing dataset, the FS-YOLOv9 improved the average F1 score, FAUC, and mAP50 by 4.58%, 5.78%, and 4.41% respectively.</div></div><div><h3>Conclusion</h3><div>Our FS-YOLOv9 model showed significantly improved performance in detecting breast cancer, making it more practical for high-risk breast cancer diagnosis.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 3","pages":"Pages 1228-1240"},"PeriodicalIF":3.8000,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Academic Radiology","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1076633224006974","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0

Abstract

Rationale and Objectives

Breast cancer screening is critical for reducing mortality rates. YOLOv9, a new real-time object-detection model, is ideal for cancer screening. A customized YOLOv9 model with enhancements for detecting breast cancer on the basis of species and morphological diversity has potential clinical significance.

Materials and Methods

The internal dataset consisted of 687 cases split 3:1 for cross-validation. Additionally, 98 cases from external datasets were used for testing. We developed an FS-YOLOv9 model customized for breast cancer detection that incorporated an extra max-pooling layer before the Conv1 of the Adown to enhance high-brightness features. The Adown of the P3 in the backbone was replaced with a high-frequency Haar wavelet convolution kernel, which ignored the low-frequency components during down-sampling to enhance morphology and texture features. The reliability and robustness of our model was determined by measuring the F1 score, the area under curve of free-response receiver operating characteristic (FAUC), mean average precision (mAP), recall, and precision, and comparing them with the findings for the official YOLOv9, YOLOv8, YOLOv5 models.

Results

In comparison with the official YOLOv9 model, FS-YOLOv9 showed a higher average F1 score (0.700 vs. 0.669), FAUC (0.695 vs. 0.662), and mAP50 (0.713 vs. 0.679) in the internal dataset; in the external testing dataset, the FS-YOLOv9 improved the average F1 score, FAUC, and mAP50 by 4.58%, 5.78%, and 4.41% respectively.

Conclusion

Our FS-YOLOv9 model showed significantly improved performance in detecting breast cancer, making it more practical for high-risk breast cancer diagnosis.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
FS-YOLOv9:基于频率和空间特征的实时乳腺癌检测 YOLOv9。
理由和目标:乳腺癌筛查对于降低死亡率至关重要。新型实时物体检测模型 YOLOv9 是癌症筛查的理想选择。定制的 YOLOv9 模型在物种和形态多样性的基础上增强了检测乳腺癌的功能,具有潜在的临床意义:内部数据集由 687 个病例组成,以 3:1 的比例进行交叉验证。此外,我们还使用了外部数据集中的 98 个病例进行测试。我们开发了一个专为乳腺癌检测定制的 FS-YOLOv9 模型,该模型在 Adown 的 Conv1 之前加入了一个额外的最大池化层,以增强高亮度特征。骨干 P3 的 Adown 由高频 Haar 小波卷积核代替,在下采样过程中忽略了低频成分,以增强形态和纹理特征。通过测量 F1 分数、自由响应接收器操作特征曲线下面积(FAUC)、平均精度(mAP)、召回率和精度,并与官方 YOLOv9、YOLOv8 和 YOLOv5 模型的结果进行比较,确定了我们的模型的可靠性和鲁棒性:与官方 YOLOv9 模型相比,FS-YOLOv9 在内部数据集中的平均 F1 得分(0.700 vs. 0.669)、FAUC(0.695 vs. 0.662)和 mAP50(0.713 vs. 0.679)更高;在外部测试数据集中,FS-YOLOv9 的平均 F1 得分、FAUC 和 mAP50 分别提高了 4.58%、5.78% 和 4.41%:我们的 FS-YOLOv9 模型在检测乳腺癌方面的性能有了显著提高,使其更适用于高危乳腺癌诊断。
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 去求助
来源期刊
Academic Radiology
Academic Radiology 医学-核医学
CiteScore
7.60
自引率
10.40%
发文量
432
审稿时长
18 days
期刊介绍: Academic Radiology publishes original reports of clinical and laboratory investigations in diagnostic imaging, the diagnostic use of radioactive isotopes, computed tomography, positron emission tomography, magnetic resonance imaging, ultrasound, digital subtraction angiography, image-guided interventions and related techniques. It also includes brief technical reports describing original observations, techniques, and instrumental developments; state-of-the-art reports on clinical issues, new technology and other topics of current medical importance; meta-analyses; scientific studies and opinions on radiologic education; and letters to the Editor.
期刊最新文献
How to Succeed on Your Residency Interview Abnormal Functional Attributes of Central Executive Network in Patients with Mild Cognitive Impairment Associated with End-Stage Renal Disease Enhancing Outcome Prediction in Intracerebral Hemorrhage Through Deep Learning: A Retrospective Multicenter Study Advances in spatial resolution and radiation dose reduction using super-resolution deep learning–based reconstruction for abdominal computed tomography: A phantom study Diffusion Weighted Imaging for the Assessment of Lymph Node Metastases in Women with Cervical Cancer: A Meta-analysis of the Apparent Diffusion Coefficient Values
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1