Journals / Black Sea Journal of Engineering and Science / 2025 / Cilt: 8 Sayı: 6

Histopathologic Image Classification of Breast Cancer Using Deep Learning Algorithms

Derin Öğrenme Algoritmalarıyla Meme Kanserinin Histopatolojik Görüntü Sınıflandırması

Ejder Ertürkmen(KAHRAMANMARAŞ SÜTÇÜ İMAM ÜNİVERSİTESİ)0000-0001-5961-6326
Ali Öter(KAHRAMANMARAŞ SÜTÇÜ İMAM ÜNİVERSİTESİ, TEKNİK BİLİMLER MESLEK YÜKSEKOKULU)0000-0002-9546-0602
Pages
1697–1714
Publication Date
2025-11-15
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results show that the EfficientNet-Swin model achieves the highest success rate with 92% accuracy and 97% ROC-AUC.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This research aims to compare the performance of different deep learning models used in breast cancer diagnosis to determine the most successful model.
📊
Key Findings & Data:Breast cancer is one of the most common and deadly cancers worldwide, and early detection of breast cancer significantly increases the survival rate of patients.
💡
Conclusion & Impact:The results show that the EfficientNet-Swin model achieves the highest success rate with 92% accuracy and 97% ROC-AUC.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Derin öğrenme · Meme kanseri · Histopolojik görüntü

Abstract

This research aims to compare the performance of different deep learning models used in breast cancer diagnosis to determine the most successful model. Breast cancer is one of the most common and deadly cancers worldwide, and early detection of breast cancer significantly increases the survival rate of patients. In recent years, deep learning-based models have made great progress in the field of medical image analysis and have started to play an important role in the diagnosis process, especially by achieving high accuracy rates on histopathological images. In this research, EfficientNet-Swin, VGG16, ResNet50 and Hybrid CNN-LSTM-Quantum models are compared and their classification performances are evaluated using metrics such as Precision, Recall, F1-Score, Accuracy and AUC. The results show that the EfficientNet-Swin model achieves the highest success rate with 92% accuracy and 97% ROC-AUC. As a transformer-based model, EfficientNet-Swin has better generalization capacity and strong feature extraction capability compared to traditional CNN models. The Hybrid CNN-LSTM-Quantum model offers an innovative approach by combining deep learning and quantum computing techniques. This model has emerged as a promising method, especially in biomedical imaging applications that require time series analysis. The VGG16 model is notable for its low false positive rate, while the ResNet50 model requires additional optimization due to the risk of overlearning. The findings of the study show that transformer-based models have higher accuracy and generalization capacity compared to traditional CNN architectures. In particular, the EfficientNet-Swin model was found to be the most suitable model for clinical use for breast cancer diagnosis. Future work should focus on testing these models on larger and more diverse datasets to ensure their clinical integration. Furthermore, the development of quantum computing-enabled hybrid models can further improve the accuracy and efficiency of deep learning-based diagnostic systems.

Özet

Bu çalışma, meme kanseri teşhisinde kullanılan farklı DL modellerinin performanslarını karşılaştırarak en başarılı modelin belirlenmesini amaçlamaktadır. Meme kanseri, dünya genelinde en yaygın ve ölümcül kanser türlerinden biri olup, erken teşhis edilmesi hastaların sağkalım oranlarını önemli ölçüde artırmaktadır. Son yıllarda DL tabanlı modeller, tıbbi görüntü analizi alanında büyük ilerlemeler kaydetmiş ve özellikle histopatolojik görüntüler üzerinde yüksek doğruluk oranları elde edilerek teşhis sürecinde önemli bir rol oynamaya başlamıştır. Çalışmada, EfficientNet-Swin, VGG16, ResNet50 ve Hibrit CNN-LSTM-Quantum modelleri karşılaştırılmış ve bu modellerin sınıflandırma performansları kesinlik (Precision), duyarlılık (Recall), F1-Skoru, doğruluk (Accuracy) ve AUC gibi ölçütler kullanılarak değerlendirilmiştir. Elde edilen sonuçlar, EfficientNet-Swin modelinin %92 doğruluk ve %97 AUC değeri ile en yüksek başarı oranına ulaştığını göstermektedir. Transformer tabanlı bir model olan EfficientNet-Swin, geleneksel CNN modellerine kıyasla daha iyi genelleme kapasitesine ve güçlü öznitelik çıkarma yeteneğine sahiptir. Hibrit CNN-LSTM-Quantum modeli, DL ve kuantum hesaplama tekniklerini birleştirerek yenilikçi bir yaklaşım sunmaktadır. Bu model, özellikle zaman serisi analizi gerektiren biyomedikal görüntüleme uygulamalarında umut vadeden bir yöntem olarak öne çıkmıştır. VGG16 modeli, düşük yanlış pozitif oranı ile dikkat çekerken, ResNet50 modeli aşırı öğrenme riski nedeniyle ek optimizasyon gerektirmektedir. Çalışmadan elde edilen bulgular, transformer tabanlı modellerin geleneksel CNN mimarilerine kıyasla daha yüksek doğruluk ve genelleme kapasitesine sahip olduğunu göstermektedir. Özellikle EfficientNet-Swin modelinin, meme kanseri teşhisi için klinik kullanıma en uygun model olduğu belirlenmiştir. Gelecekteki çalışmalar, bu modellerin daha büyük ve çeşitli veri setleri üzerinde test edilerek klinik entegrasyonlarının sağlanmasına odaklanmalıdır. Ayrıca, kuantum hesaplama destekli hibrit modellerin geliştirilmesi, DL tabanlı teşhis sistemlerinin doğruluk ve verimliliğini daha da artırabilir.

🔖

Cite This Article

Ertürkmen, E., Öter, A. (2025). Histopathologic Image Classification of Breast Cancer Using Deep Learning Algorithms. Black Sea Journal of Engineering and Science, Cilt: 8 Sayı: 6, 1697–1714. https://doi.org/10.34248/bsengineering.1644212

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (60)

  1. Abraham H, Akhalwaya IY, Alexander T, vd. 2019. Qiskit: an open-source framework for quantum computing. IBM Research. URL: https://qiskit.org
  2. Agliardi G, Grossi M, Pellen M, Prati E. 2022. Quantum integration of elementary particle processes. Phys Lett B. https://doi.org/10.1016/j.physletb.2022.137228
  3. Akın E, Şahin ME. 2024. DL ve yapay sinir ağı modelleri üzerine bir inceleme. EMO Bilimsel Dergi, 14(1): 27-38.
  4. Albarqouni S, Baur C, Achilles F, Belagiannis V, Demirci S, Navab N. 2016. AggNet: deep learning from crowds for mitosis detection in breast cancer histology images. IEEE Trans Med Imaging, 35(5): 1313-1321. https://doi.org/10.1109/TMI.2016.2528120
  5. Başara B, Soytutan Çağlar İ, Aygün A, Özdemir TA, Kulalı B, Uzun SB. 2019. Sağlık istatistikleri yıllığı. URL: https://sbsgm.saglik.gov.tr/Eklenti/40564/0/saglikistatistikleri-yilligi-2019.pdf (erişim tarihi: 10 Şubat 2025).
  6. Biçici Y. 2023. DL ile mammogram görüntülerindeki lezyonların tespiti ve sınıflandırılması. YL tezi. Bilecik Şeyh Edebali Üniversitesi, Fen Bilimleri Enstitüsü, Bilecik.
  7. Brown J, Lee K. 2019. Deep learning in medical image analysis: challenges and applications. Springer.
  8. Cengil E, Çınar A. 2020. Göğüs verileri metrikleri üzerinden kanser sınıflandırılması. DÜMF Müh Derg, 11(2): 513–519.
  9. Chang J, Yu J, Han T, Chang H, Park E. 2017. A method for classifying medical images using transfer learning: a pilot study on histopathology of breast cancer. In: IEEE 19th Int Conf e-Health Netw Appl Serv, Oct 1-4, Dalian, China.
  10. Dastagir M, Han D. 2024. Towards hybrid quantum-classical deep learning architecture for indoor-outdoor detection using QCNN-LSTM and cluster state signal processing. IEEE Signal Process Lett, 31: 2945-2949. https://doi.org/10.1109/LSP.2024.3480043
  11. Dixit A, Mani A, Gorbachev S. 2024. Hybrid framework for medical image classification: integrating quantum DE and deep learning. In: 2024 IEEE Int Conf Intell Signal Process Effective Commun Technol, 1-6. https://doi.org/10.1109/INSPECT63485.2024.10896054
  12. Dümen S, Kavalcı Yılmaz E, Adem K. 2024. Meyve suyu fabrikalarında limon kalite sınıflandırması için görüntü dönüştürücü ve swin dönüştürücü modellerinin performansı. Eur Food Res Technol, 250: 2291–2302. https://doi.org/10.1007/s00217-024-04537-5
  13. Erdem E, Aydın T. 2021. Göğüs kanseri histopatolojik görüntü sınıflandırması. Bilisim Teknol Derg, 14(1): 87-94. https://doi.org/10.17671/gazibtd.746673
  14. Gonzalez M, Patel R, Smith J. 2023. Performance metrics in deep learning: a comparative study. Int J Artif Intell, 45(2): 112-130.
  15. Guang Y, Su-Ya C, Min N, Yuan-Hua L, Mei-Ling Z. 2023. Research on the construction method of hybrid quantum LSTM neural network for image classification. Acta Phys Sin, 72(5): 058901. https://doi.org/10.7498/aps.72.20221924
  16. Gupta K, Chawla N. 2020. Analysis of histopathological images for prediction of breast cancer using traditional classifiers with pre-trained CNN. Procedia Comput Sci, 167: 878-889.
  17. Güler A. 2024. Quantum algorithms for image processing: enhancing computational efficiency and accuracy in high-dimensional visual data analysis. Hum Comput Interact. https://doi.org/10.62802/hxc0ag94
  18. Habek GC, Taşdemir S, Başçiftci F, Yılmaz A. 2024. Transfer DL teknikleri ile görüntü sınıflandırmada aktivasyon fonksiyonlarının performans üzerindeki etkisi. Afyon Kocatepe Univ Fen Muh Bilim Derg, 24(2): 294-307. https://doi.org/10.35414/akufemubid.1334098
  19. Henkel C. 2021. Efficient large-scale image retrieval with deep feature orthogonality and hybrid-swin-transformers. URL: https://arxiv.org/abs/2110.03786 (erişim tarihi: 10 Şubat 2025).
  20. Hernandez D, Patel S. 2020. Evaluating false negatives in medical diagnosis using machine learning models. J Med Inform, 38(4): 215-229.
  21. Hochreiter S, Schmidhuber J. 1997. Long short-term memory. Neural Comput, 9(8): 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
  22. Hossin M, Sulaiman MN. 2015. A review on evaluation metrics for data classification evaluations. Int J Data Knowl Eng Process, 5(11): 1-11. https://doi.org/10.5121/ijdkp.2015.5201
  23. Hua L, Zhuang S, Li D, Zhao J, Ma Y. 2019. Benign and malignant classification of mammogram images based on deep learning. Biomed Signal Process Control, 51: 347–354. https://doi.org/10.1016/j.bspc.2019.03.005
  24. Hüseyin MM, Ali MŞ, Ahmed MM, Rakib MRH, Kona MA, Afrin S, İslam MK, Ahsan MM, Raj ŞMRH, Rahman MH. 2023. Explainable AI-based hybrid CNN-LSTM model for cardiovascular disease identification. Intell Med, 42: 1-15. https://doi.org/10.1016/j.imu.2023.101370
  25. Kaddes M, Ayid YM, Elshewey AM, vd. 2025. LSTM modeline sahip hibrit CNN’ye dayalı meme kanseri sınıflandırması. Sci Rep, 15: 4409. https://doi.org/10.1038/s41598-025-88459-6
  26. Kaggle. Breast histopathology images. URL: https://www.kaggle.com/datasets/paultimothymooney/breast-histopathology-images (erişim tarihi: 10 Ocak 2025).
  27. Karakurt M, İşeri İ. 2022. Patoloji görüntülerinin DL yöntemleri ile sınıflandırılması. Avrupa Bilim Teknol Derg, (33): 192-206. https://doi.org/10.31590/ejosat.1011091
  28. Kausar T, Wang M, Idrees M, Lu Y. 2019. HWDCNN: multi-class recognition in breast histopathology with Haar wavelet decomposed image based convolution neural network. Biocybern Biomed Eng, 39(4): 967-982.
  29. Kulkarni A, Gite S, Kulkarni A, Sonawani S. 2024. Bridging traditional and modern AI techniques for breast cancer histopathology. In: Proc 2024 Int Conf Intell Comput Emerg Commun Technol (ICEC), Nov 23-25, Guntur, India.
  30. Kumar R, Zhang Y, Li H. 2021. Transfer learning and deep neural networks for breast cancer detection. IEEE J Biomed Health Inform, 40(5): 1342-1355.
  31. Lakshmi Priya CV, Biju VG, Vinod BR, Ramachandran S. 2024. Deep learning approaches for breast cancer detection in histopathology images: a review. Cancer Biomark, 40(1): 1-25. https://doi.org/10.3233/CBM-230251
  32. LeCun Y, Bengio Y, Hinton G. 2015. Deep learning. Nature, 521(7553): 436–444. https://doi.org/10.1038/nature14539
  33. Li H, Zhuang S, Li D, Zhao J, Ma Y. 2019. Benign and malignant classification of mammogram images based on deep learning. Biomed Signal Process Control, 51: 347-354. https://doi.org/10.1016/j.bspc.2019.02.017
  34. Mahmud MI, Mamun M, Abdelgawad A. 2023. A deep analysis of transfer learning based breast cancer detection using histopathology images. Electr Eng Syst Sci Image Video Process. https://doi.org/10.48550/arXiv.2304.05022
  35. Mayfield J, El-Naqa I. 2024. Evaluation of VQC-LSTM for disability forecasting in multiple sclerosis using sequential multisequence MRI. Quantum Mach Intell, 6: 41. https://doi.org/10.1007/s42484-024-00171-2
  36. Meva. Meme kanseri teşhis yöntemleri. URL: https://www.memekanseri.org.tr/meme-sagligi/meme-kanseri-teshis-yontemleri (erişim tarihi: 10 Ocak 2025).
  37. Mridha MF, Hamid MA, Monowar MM, Keya AJ, Ohi AQ, Islam MR, Kim JM. 2021. Derin öğrenmeye dayalı meme kanseri teşhisine ilişkin kapsamlı bir araştırma. Cancers, 13(23): 1-20.
  38. Nahid AA, Mehrabi MA, Kong Y. 2018. Histopathological breast cancer image classification by deep neural network techniques guided by local clustering. Biomed Res Int, 2018: 45.
  39. Nahid AA, Mehrabi MA, Kong Y. 2018. Histopathological breast-cancer image classification using LSTM with transfer learning. In: Proc Int Conf Mach Learn Cybern, 1–5.
  40. Nguyen P, Tran L, Wang T. 2021. The impact of imbalanced datasets on classification performance: a case study in medical imaging. Bioinform Comput Biol J, 29(3): 56-72.
  41. Özgür SN, Bozkurt Keser S. 2021. Meme kanseri tümörlerinin DL algoritmaları ile sınıflandırılması. Turk Doga Fen Derg, 10(2): 212-222. https://doi.org/10.46810/tdfd.957618
  42. Perez-Salinas A, Cervera-Lierta A, Gil-Fuster E, Latorre JI. 2020. Data re-uploading for a universal quantum classifier. Quantum, 4: 226. https://doi.org/10.22331/q-2020-02-06-226
  43. Phyo P, Byun Y. 2021. Hybrid ensemble deep learning-based approach for time series energy prediction. Symmetry, 13(10): 19-42.
  44. Ragab DA, Sharkas M, Marshall S, Ren J. 2021. A framework for breast cancer classification using deep learning CNNs. Inform Med Unlock, 23: 100551. https://doi.org/10.1016/j.imu.2021.100551
  45. Rashmi R, Prasad K, Udupa CBK. 2021. Breast histopathological image analysis using image processing techniques for diagnostic purposes: a methodological review. J Med Syst, 46(1): 7.
  46. Sağlık Bakanlığı Halk Sağlığı Genel Müdürlüğü. Mamografide görüntü kalitesi. URL: https://hsgm.saglik.gov.tr/tr/kanser (erişim tarihi: 25 Ocak 2025).
  47. Sağlık Bakanlığı Halk Sağlığı Genel Müdürlüğü. Türkiye kanser istatistikleri. URL: https://hsgm.saglik.gov.tr/tr/kanser-istatistikleri (erişim tarihi: 25 Ocak 2025).
  48. Saha M, Chakraborty C, Racoceanu D. 2018. Efficient deep learning model for mitosis detection using breast histopathology images. Comput Med Imaging Graph, 64: 29-40.
  49. Schuld M, Sinayskiy I, Petruccione F. 2019. Machine learning with quantum computers. Springer. https://doi.org/10.1007/978-3-030-18545-4
  50. Smith A, Johnson B, Wang C. 2020. Metrics for evaluating deep learning models in medical image classification. Comput Biol Med, 115: 56-67.
  51. Spanhol FA, Oliveira LS, Cavalin PR, Petitjean C, Heutte L. 2017. Deep features for breast cancer histopathological image classification. In: Proc IEEE Int Conf Syst Man Cybern (SMC), Oct 5-8, Banff, Canada.
  52. Spanhol FA, Oliveira LS, Petitjean C, Heutte L. 2016. A dataset for breast cancer histopathological image classification. IEEE Trans Biomed Eng, 63(7): 1455-1462.
  53. Subasish M, Sarmistha M, Subhadarshini M, Ravindra JVR, Sachi NM. 2022. Evaluation of deep learning models for detecting breast cancer using histopathological mammogram images. Sustain Oper Comput, 3: 296-302.
  54. Utku A. 2024. Hybrid CNN-LSTM model for accurate long-term and short-term temperature prediction: a case study for Bingöl and Tunceli. Int J Pure Appl Sci, 10(2): 550-562. https://doi.org/10.29132/ijpas.1548698
  55. Utku F. 2024. Kuantum bilişim ve enerji uygulamaları. İstanbul: TeknoYay.
  56. Wahab N, Khan A. 2020. Multifaceted fused-CNN based scoring of breast cancer whole-slide histopathology images. Appl Soft Comput, 97: 106121.
  57. WHO. The global breast cancer initiative. URL: https://www.who.int/initiatives/global-breast-cancer-initiative (erişim tarihi: 10 Ocak 2025).
  58. Xu J, Xiang L, Li Q, Gilmore H, Wu J, Tang J, Madabhushi A. 2016. Stacked sparse autoencoder (SSAE) for nuclei detection on breast cancer histopathology images. IEEE Transactions on Medical Imaging, 35(1): 119–130. https://doi.org/10.1109/TMI.2015.2458702.
  59. Yang Z, Ran L, Zhang S, Xia Y, Zhang Y. 2019. EMS-Net: ensemble of multiscale convolutional neural networks for classification of breast cancer histology images. Neurocomputing, 366: 46-53.
  60. Zhang X, Chen L, Zhao M. 2022. Precision and recall trade-offs in deep learning-based medical diagnosis. Neural Netw, 56(7): 99-115.