Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Jun 11;12(6):640.
doi: 10.3390/bioengineering12060640.

A Lightweight Breast Cancer Mass Classification Model Utilizing Simplified Swarm Optimization and Knowledge Distillation

Wei-Chang Yeh  1   2 , Wei-Chung Shia  3 , Yun-Ting Hsu  1 , Chun-Hui Huang  4 , Yong-Shiuan Lee  5
Affiliations

A Lightweight Breast Cancer Mass Classification Model Utilizing Simplified Swarm Optimization and Knowledge Distillation

Wei-Chang Yeh et al. Bioengineering (Basel). .

Abstract

In recent years, an increasing number of women worldwide have been affected by breast cancer. Early detection is crucial, as it is the only way to identify abnormalities at an early stage. However, most deep learning models developed for classifying breast cancer abnormalities tend to be large-scale and computationally intensive, often overlooking the constraints of cost and limited computational resources. This research addresses these challenges by utilizing the CBIS-DDSM dataset and introducing a novel concatenated classification architecture and a two-stage strategy to develop an optimized, lightweight model for breast mass abnormality classification. Through data augmentation and image preprocessing, the proposed model demonstrates a superior performance compared to standalone CNN and DNN models. The two-stage strategy involves first constructing a compact model using knowledge distillation and then refining its structure with a heuristic approach known as Simplified Swarm Optimization (SSO). The experimental results confirm that knowledge distillation significantly enhances the model's performance. Furthermore, by applying SSO's full-variable update mechanism, the final model-SSO-Concatenated NASNetMobile (SSO-CNNM)-achieves outstanding performance metrics. It attains a compression rate of 96.17%, along with accuracy, precision, recall, and AUC scores of 96.47%, 97.4%, 94.94%, and 98.23%, respectively, outperforming other existing methods.

Keywords: convolutional neural networks; knowledge distillation; lightweight breast cancer mass classification model; simplified swarm optimization.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure 1
Figure 1
Comparison of an ROI image with and without white borders. (a) White borders in the ROI image (image with black frame). (b) ROI image after the removal of white borders.
Figure 2
Figure 2
Comparison of ROI images before and after applying the NLM algorithm. (a) Original ROI image. (b) ROI image after applying the NLM algorithm.
Figure 3
Figure 3
Comparison of ROI images before and after applying CLAHE. (a) Original ROI image. (b) ROI image after applying CLAHE.
Figure 4
Figure 4
A two-stage framework diagram illustrating the model architecture.
Figure 5
Figure 5
The integrated classification model architecture proposed in this study.
Figure 6
Figure 6
Knowledge distillation process.
Figure 7
Figure 7
SSO-optimized architecture of the lightweight ensemble classification model.
Figure 8
Figure 8
Illustration of the encoding and decoding scheme used in this study.
Figure 9
Figure 9
Performance of DNN and CNN models (accuracy, Before Data Augmentation).
Figure 10
Figure 10
Performance of DNN and integrated classification models (accuracy, Before Data Augmentation).
Figure 11
Figure 11
Performance of the DNN and CNN models (accuracy, After Data Augmentation).
Figure 12
Figure 12
Performance of the DNN and integrated classification models (accuracy, After Data Augmentation).
Figure 13
Figure 13
Lightweight integrated classification model.
Figure 14
Figure 14
Box plot of accuracy across the four experimental levels.
Figure 15
Figure 15
ROC curve of SSO-CNNM.
Figure 16
Figure 16
Convergence process of SSO iterative optimization.
Figure 17
Figure 17
Optimized lightweight concatenated classification model (SSO-CNNM).
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Sharma G.N., Dave R., Sanadya J., Sharma P., Sharma K. Various types and management of breast cancer: An overview. J. Adv. Pharm. Technol. Res. 2010;1:109. doi: 10.4103/2231-4040.72251. - DOI - PMC - PubMed
    1. World Health Organization. International Agency for Research on Cancer . GLOBOCAN 2012: Estimated Cancer Incidence, Mortality and Prevalence Worldwide in 2012. IARC Publications; Lyon, France: 2012.
    1. Rangayyan R.M., Ayres F.J., Desautels J.L. A review of computer-aided diagnosis of breast cancer: Toward the detection of subtle signs. J. Frankl. Inst. 2007;344:312–348. doi: 10.1016/j.jfranklin.2006.09.003. - DOI
    1. Sardanelli F., Aase H.S., Álvarez M., Azavedo E., Baarslag H.J., Balleyguier C., Baltzer P.A., Beslagic V., Bick U., Bogdanovic-Stojanovic D., et al. Position paper on screening for breast cancer by the European Society of Breast Imaging (EUSOBI) and 30 national breast radiology bodies from Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Israel, Lithuania, Moldova, The Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Spain, Sweden, Switzerland and Turkey. Eur. Radiol. 2017;27:2737–2743. - PMC - PubMed
    1. Debelee T.G., Schwenker F., Ibenthal A., Yohannes D. Survey of deep learning in breast cancer image analysis. Evol. Syst. 2020;11:143–163. doi: 10.1007/s12530-019-09297-2. - DOI

LinkOut - more resources