1
|
Xu Z, Zhong S, Gao Y, Huo J, Xu W, Huang W, Huang X, Zhang C, Zhou J, Dan Q, Li L, Jiang Z, Lang T, Xu S, Lu J, Wen G, Zhang Y, Li Y. Optimizing breast lesions diagnosis and decision-making with a deep learning fusion model integrating ultrasound and mammography: a dual-center retrospective study. Breast Cancer Res 2025; 27:80. [PMID: 40369585 PMCID: PMC12080162 DOI: 10.1186/s13058-025-02033-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/02/2024] [Accepted: 04/22/2025] [Indexed: 05/16/2025] Open
Abstract
BACKGROUND This study aimed to develop a BI-RADS network (DL-UM) via integrating ultrasound (US) and mammography (MG) images and explore its performance in improving breast lesion diagnosis and management when collaborating with radiologists, particularly in cases with discordant US and MG Breast Imaging Reporting and Data System (BI-RADS) classifications. METHODS We retrospectively collected image data from 1283 women with breast lesions who underwent both US and MG within one month at two medical centres and categorised them into concordant and discordant BI-RADS classification subgroups. We developed a DL-UM network via integrating US and MG images, and DL networks using US (DL-U) or MG (DL-M) alone, respectively. The performance of DL-UM network for breast lesion diagnosis was evaluated using ROC curves and compared to DL-U and DL-M networks in the external testing dataset. The diagnostic performance of radiologists with different levels of experience under the assistance of DL-UM network was also evaluated. RESULTS In the external testing dataset, DL-UM outperformed DL-M in sensitivity (0.962 vs. 0.833, P = 0.016) and DL-U in specificity (0.667 vs. 0.526, P = 0.030), respectively. In the discordant BI-RADS classification subgroup, DL-UM achieved an AUC of 0.910. The diagnostic performance of four radiologists improved when collaborating with the DL-UM network, with AUCs increased from 0.674-0.772 to 0.889-0.910, specificities from 52.1%-75.0 to 81.3-87.5% and reducing unnecessary biopsies by 16.1%-24.6%, particularly for junior radiologists. Meanwhile, DL-UM outputs and heatmaps enhanced radiologists' trust and improved interobserver agreement between US and MG, with weighted kappa increased from 0.048 to 0.713 (P < 0.05). CONCLUSIONS The DL-UM network, integrating complementary US and MG features, assisted radiologists in improving breast lesion diagnosis and management, potentially reducing unnecessary biopsies.
Collapse
Affiliation(s)
- Ziting Xu
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Shengzhou Zhong
- School of Biomedical Engineering, Southern Medical University, Guangzhou, 510515, Guangdong, China
| | - Yang Gao
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Jiekun Huo
- Department of Imaging, Zengcheng Branch of Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Weimin Xu
- Department of Radiology, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Weijun Huang
- Department of Ultrasound, First People's Hospital of Foshan, Foshan, 510515, Guangdong, China
| | - Xiaomei Huang
- Department of Medical Imaging, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, Guangdong, China
| | - Chifa Zhang
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Jianqiao Zhou
- Department of Ultrasound, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, 200025, China
| | - Qing Dan
- Shenzhen Key Laboratory for Drug Addiction and Medication Safety, Department of Ultrasound, Institute of Ultrasonic Medicine, Peking University Shenzhen Hospital, Shenzhen Peking University-The Hong Kong University of Science and Technology Medical Center, Shenzhen, 518036, China
| | - Lian Li
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Zhouyue Jiang
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Ting Lang
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Shuying Xu
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Jiayin Lu
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China
| | - Ge Wen
- Department of Medical Imaging, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, Guangdong, China.
| | - Yu Zhang
- School of Biomedical Engineering, Southern Medical University, Guangzhou, 510515, Guangdong, China.
| | - Yingjia Li
- Department of Ultrasound, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China.
| |
Collapse
|
2
|
Larroza A, Pérez-Benito FJ, Tendero R, Perez-Cortes JC, Román M, Llobet R. Breast Delineation in Full-Field Digital Mammography Using the Segment Anything Model. Diagnostics (Basel) 2024; 14:1015. [PMID: 38786313 PMCID: PMC11120343 DOI: 10.3390/diagnostics14101015] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/17/2024] [Revised: 05/06/2024] [Accepted: 05/13/2024] [Indexed: 05/25/2024] Open
Abstract
Breast cancer is a major health concern worldwide. Mammography, a cost-effective and accurate tool, is crucial in combating this issue. However, low contrast, noise, and artifacts can limit the diagnostic capabilities of radiologists. Computer-Aided Diagnosis (CAD) systems have been developed to overcome these challenges, with the accurate outlining of the breast being a critical step for further analysis. This study introduces the SAM-breast model, an adaptation of the Segment Anything Model (SAM) for segmenting the breast region in mammograms. This method enhances the delineation of the breast and the exclusion of the pectoral muscle in both medio lateral-oblique (MLO) and cranio-caudal (CC) views. We trained the models using a large, multi-center proprietary dataset of 2492 mammograms. The proposed SAM-breast model achieved the highest overall Dice Similarity Coefficient (DSC) of 99.22% ± 1.13 and Intersection over Union (IoU) 98.48% ± 2.10 over independent test images from five different datasets (two proprietary and three publicly available). The results are consistent across the different datasets, regardless of the vendor or image resolution. Compared with other baseline and deep learning-based methods, the proposed method exhibits enhanced performance. The SAM-breast model demonstrates the power of the SAM to adapt when it is tailored to specific tasks, in this case, the delineation of the breast in mammograms. Comprehensive evaluations across diverse datasets-both private and public-attest to the method's robustness, flexibility, and generalization capabilities.
Collapse
Affiliation(s)
- Andrés Larroza
- Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera s/n, 46022 València, Spain; (A.L.); (F.J.P.-B.); (R.T.); (J.C.P.-C.)
| | - Francisco Javier Pérez-Benito
- Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera s/n, 46022 València, Spain; (A.L.); (F.J.P.-B.); (R.T.); (J.C.P.-C.)
| | - Raquel Tendero
- Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera s/n, 46022 València, Spain; (A.L.); (F.J.P.-B.); (R.T.); (J.C.P.-C.)
| | - Juan Carlos Perez-Cortes
- Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera s/n, 46022 València, Spain; (A.L.); (F.J.P.-B.); (R.T.); (J.C.P.-C.)
| | - Marta Román
- Department of Epidemiology and Evaluation, IMIM (Hospital del Mar Research Institute), Passeig Marítim 25-29, 08003 Barcelona, Spain;
| | - Rafael Llobet
- Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera s/n, 46022 València, Spain; (A.L.); (F.J.P.-B.); (R.T.); (J.C.P.-C.)
| |
Collapse
|
3
|
Han G, Guo W, Zhang H, Jin J, Gan X, Zhao X. Sample self-selection using dual teacher networks for pathological image classification with noisy labels. Comput Biol Med 2024; 174:108489. [PMID: 38640633 DOI: 10.1016/j.compbiomed.2024.108489] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/15/2023] [Revised: 04/02/2024] [Accepted: 04/15/2024] [Indexed: 04/21/2024]
Abstract
Deep neural networks (DNNs) involve advanced image processing but depend on large quantities of high-quality labeled data. The presence of noisy data significantly degrades the DNN model performance. In the medical field, where model accuracy is crucial and labels for pathological images are scarce and expensive to obtain, the need to handle noisy data is even more urgent. Deep networks exhibit a memorization effect, they tend to prioritize remembering clean labels initially. Therefore, early stopping is highly effective in managing learning with noisy labels. Previous research has often concentrated on developing robust loss functions or implementing training constraints to mitigate the impact of noisy labels; however, such approaches have frequently resulted in underfitting. We propose using knowledge distillation to slow the learning process of the target network rather than preventing late-stage training from being affected by noisy labels. In this paper, we introduce a data sample self-selection strategy based on early stopping to filter out most of the noisy data. Additionally, we employ the distillation training method with dual teacher networks to ensure the steady learning of the student network. The experimental results show that our method outperforms current state-of-the-art methods for handling noisy labels on both synthetic and real-world noisy datasets. In particular, on the real-world pathological image dataset Chaoyang, the highest classification accuracy increased by 2.39 %. Our method leverages the model's predictions based on training history to select cleaner datasets and retrains them using these cleaner datasets, significantly mitigating the impact of noisy labels on model performance.
Collapse
Affiliation(s)
- Gang Han
- School of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, China; School of Electronic and Information Engineering, Taizhou University, Taizhou 318000, China
| | - Wenping Guo
- School of Electronic and Information Engineering, Taizhou University, Taizhou 318000, China.
| | - Haibo Zhang
- School of Electronic and Information Engineering, Taizhou University, Taizhou 318000, China
| | - Jie Jin
- School of Electronic and Information Engineering, Taizhou University, Taizhou 318000, China
| | - Xingli Gan
- School of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, China
| | - Xiaoming Zhao
- School of Electronic and Information Engineering, Taizhou University, Taizhou 318000, China
| |
Collapse
|
4
|
Montoya-del-Angel R, Sam-Millan K, Vilanova JC, Martí R. MAM-E: Mammographic Synthetic Image Generation with Diffusion Models. SENSORS (BASEL, SWITZERLAND) 2024; 24:2076. [PMID: 38610288 PMCID: PMC11014323 DOI: 10.3390/s24072076] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 12/13/2023] [Revised: 03/12/2024] [Accepted: 03/22/2024] [Indexed: 04/14/2024]
Abstract
Generative models are used as an alternative data augmentation technique to alleviate the data scarcity problem faced in the medical imaging field. Diffusion models have gathered special attention due to their innovative generation approach, the high quality of the generated images, and their relatively less complex training process compared with Generative Adversarial Networks. Still, the implementation of such models in the medical domain remains at an early stage. In this work, we propose exploring the use of diffusion models for the generation of high-quality, full-field digital mammograms using state-of-the-art conditional diffusion pipelines. Additionally, we propose using stable diffusion models for the inpainting of synthetic mass-like lesions on healthy mammograms. We introduce MAM-E, a pipeline of generative models for high-quality mammography synthesis controlled by a text prompt and capable of generating synthetic mass-like lesions on specific regions of the breast. Finally, we provide quantitative and qualitative assessment of the generated images and easy-to-use graphical user interfaces for mammography synthesis.
Collapse
Affiliation(s)
- Ricardo Montoya-del-Angel
- Computer Vision and Robotics Institute (ViCOROB), University of Girona, 17004 Girona, Spain; (K.S.-M.); (R.M.)
| | - Karla Sam-Millan
- Computer Vision and Robotics Institute (ViCOROB), University of Girona, 17004 Girona, Spain; (K.S.-M.); (R.M.)
| | - Joan C. Vilanova
- Department of Radiology, Clínica Girona, Institute of Diagnostic Imaging (IDI) Girona, University of Girona, 17004 Girona, Spain;
| | - Robert Martí
- Computer Vision and Robotics Institute (ViCOROB), University of Girona, 17004 Girona, Spain; (K.S.-M.); (R.M.)
| |
Collapse
|