Feyzi Ferat ATEŞ, Abidin ÇALIŞKAN, Mesut TOĞAÇAR

Ultrason Tabanlı Meme Kanseri Görüntülerinin Derin Öğrenme Yaklaşımları ile Sınıflandırılması

Meme kanseri bayanlar arasında en sık görülen kanser türlerinden biridir. Diğer kanser türlerinde olduğu gibi meme kanseri hastalarının tedavisinde erken tanı önemlidir. Son zamanlarda yapay zekâ birçok alanda adını duyurmuştur. Sağlık alanında da yapay zekâ tanı ve tedavi süreçlerinde teknolojik alt yapı olarak kullanılmaya başlamıştır. Bu çalışma da ultrason tabanlı görüntüler kullanılarak meme kanseri teşhisini gerçekleştirebilecek yapay zeka tabanlı bir yaklaşım önerilmektedir. Önerilen yaklaşım önceden eğitilmiş evrişimsel sinir ağlarından oluşmaktadır. Her bir evrişimsel sinir ağının son katmanına yeni bir tam bağlantılı katman eklenmiştir. Tam bağlantılı katmanı önceki tam bağlantılı katmanlardan ayırt eden özelliği girdi türü sayısı kadar öznitelik vermesidir. Ardından evrişimsel sinir ağlarının tam bağlantılı katmanları birleştirilerek sınıflandırma işlemi gerçekleşmiştir. Bu çalışmada iyi huylu, kötü huylu ve normal olmak üzere üçlü bir sınıflandırma işlemi gerçekleşmiştir. Deneysel analiz sonucunda önerilen yaklaşım ile %99,57 genel doğruluk başarısı elde edilmiştir. Önerilen yaklaşım deneyde kullanılan evrişimsel sinir ağı modellerinden daha iyi performans göstermiştir.

Anahtar Kelimeler:

Derin öğrenme, özellik seçimi, özellik birleştirme, meme kanseri, sınıflandırma

Classification of Ultrasound-Based Breast Cancer Images with Deep Learning Approaches

Breast cancer is one of the most common types of cancer among women. As with other types of cancer, early diagnosis is important in the treatment of breast cancer patients. Recently, artificial intelligence has made its name in many fields. In the field of health, artificial intelligence has started to be used as a technological infrastructure in diagnosis and treatment processes. In this study, an artificial intelligence-based approach that can diagnose breast cancer using ultrasound-based images is proposed. The proposed approach consists of pre-trained convolutional neural networks. A new fully connected layer is added to the last layer of each convolutional neural network. The feature that distinguishes the fully connected layer from the previous fully connected layers is that it gives as many features as the number of input types. Then, the classification process was carried out by combining the fully connected layers of the convolutional neural networks. In this study, a triple classification process was carried out as benign, malignant and normal. As a result of the experimental analysis, 99.57% overall accuracy was achieved with the proposed approach. The proposed approach outperformed the convolutional neural network models used in the experiment.

Keywords:

Deep learning, feature selection, feature merging, breast cancer, classification.,

PDF

___

S. Ortega-Martorell, P. Riley, I. Olier, R.G. Raidou, R. Casana-Eslava, M. Rea, L. Shen, P.J.G. Lisboa, C. Palmieri, Breast cancer patient characterisation and visualisation using deep learning and fisher information networks, Sci. Rep. 12 (2022) 14004. doi:10.1038/s41598-022-17894-6.
Y. Gao, B. Reig, L. Heacock, D.L. Bennett, S.L. Heller, L. Moy, Magnetic Resonance Imaging in Screening of Breast Cancer, Radiol. Clin. North Am. 59 (2021) 85–98. doi:10.1016/j.rcl.2020.09.004.
L. Balkenende, J. Teuwen, R.M. Mann, Application of Deep Learning in Breast Cancer Imaging, Semin. Nucl. Med. 52 (2022) 584–596. doi:10.1053/j.semnuclmed.2022.02.003.
C. Janiesch, P. Zschech, K. Heinrich, Machine learning and deep learning, Electron. Mark. 31 (2021) 685–695. doi:10.1007/s12525-021-00475-2.
F. Emmert-Streib, Z. Yang, H. Feng, S. Tripathi, M. Dehmer, An Introductory Review of Deep Learning for Prediction Models With Big Data, Front. Artif. Intell. 3 (2020). doi:10.3389/frai.2020.00004.
H. Aljuaid, N. Alturki, N. Alsubaie, L. Cavallaro, A. Liotta, Computer-aided diagnosis for breast cancer classification using deep neural networks and transfer learning, Comput. Methods Programs Biomed. 223 (2022) 106951. doi:https://doi.org/10.1016/j.cmpb.2022.106951.
S. Zahoor, U. Shoaib, I.U. Lali, Breast Cancer Mammograms Classification Using Deep Neural Network and Entropy-Controlled Whale Optimization Algorithm, Diagnostics. 12 (2022) 557. doi:10.3390/diagnostics12020557.
E.H. Houssein, M.M. Emam, A.A. Ali, An optimized deep learning architecture for breast cancer diagnosis based on improved marine predators algorithm, Neural Comput. Appl. (2022). doi:10.1007/s00521-022-07445-5.
M. Thilagaraj, N. Arunkumar, P. Govindan, Classification of Breast Cancer Images by Implementing Improved DCNN with Artificial Fish School Model, Comput. Intell. Neurosci. 2022 (2022) 6785707. doi:10.1155/2022/6785707.
W. Al-Dhabyani, M. Gomaa, H. Khaled, A. Fahmy, Dataset of breast ultrasound images, Data Br. 28 (2020) 104863. doi:10.1016/j.dib.2019.104863.
A. Shah, Breast Ultrasound Images Dataset, Kaggle Web. (2021). https://www.kaggle.com/datasets/aryashah2k/breast-ultrasound-images-dataset?resource=download.
I.H. Sarker, Deep Learning: A Comprehensive Overview on Techniques, Taxonomy, Applications and Research Directions, SN Comput. Sci. 2 (2021) 420. doi:10.1007/s42979-021-00815-1.
B. Ait Skourt, A. El Hassani, A. Majda, Mixed-pooling-dropout for convolutional neural network regularization, J. King Saud Univ. - Comput. Inf. Sci. 34 (2022) 4756–4762. doi:https://doi.org/10.1016/j.jksuci.2021.05.001.
N. Sharma, R. Sharma, N. Jindal, Machine Learning and Deep Learning Applications-A Vision, Glob. Transitions Proc. 2 (2021) 24–28. doi:10.1016/j.gltp.2021.01.004.
W. Alsaggaf, Z. Cömert, M. Nour, K. Polat, H. Brdesee, M. Toğaçar, Predicting fetal hypoxia using common spatial pattern and machine learning from cardiotocography signals, Appl. Acoust. 167 (2020) 107429. doi:10.1016/j.apacoust.2020.107429.
A. Diker, Z. Comert, E. Avci, M. Togacar, B. Ergen, A Novel Application based on Spectrogram and Convolutional Neural Network for ECG Classification, in: 2019 1st Int. Informatics Softw. Eng. Conf., IEEE, 2019: pp. 1–6. doi:10.1109/UBMYK48245.2019.8965506.
M. Toğaçar, B. Ergen, Z. Cömert, Tumor type detection in brain MR images of the deep model developed using hypercolumn technique, attention modules, and residual blocks, Med. Biol. Eng. Comput. 59 (2021) 57–70. doi:10.1007/s11517-020-02290-x.
C. Banerjee, T. Mukherjee, E. Pasiliao, An Empirical Study on Generalizations of the ReLU Activation Function, in: Proc. 2019 ACM Southeast Conf., Association for Computing Machinery, New York, NY, USA, 2019: pp. 164–167. doi:10.1145/3299815.3314450.
S. Li, L. Wang, J. Li, Y. Yao, Image Classification Algorithm Based on Improved AlexNet, J. Phys. Conf. Ser. 1813 (2021). doi:10.1088/1742-6596/1813/1/012051.
J. Redmon, A. Farhadi, YOLO9000: Better, faster, stronger, Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017. 2017-Janua (2017) 6517–6525. doi:10.1109/cvpr.2017.690.
P. Sowa, J. Izydorczyk, Darknet on OpenCL: a multi-platform tool for object detection and classification, (2020) 1–22. doi:10.20944/preprints202007.0506.v1.
A. Venkata, S. Abhishek, Resnet18 Model With Sequential Layer For Computing Accuracy On Image Classification Dataset, 10 (2022) 176–181.
M. El, A. Seddik, C. Louart, R. Couillet, M. Tamaazousti, The Unexpected Deterministic and Universal Behavior of Large Softmax Classifiers, Int. Conf. Artif. Intell. Stat. (2021). https://melaseddik.github.
M. Toğaçar, Z. Cömert, B. Ergen, Enhancing of dataset using DeepDream, fuzzy color image enhancement and hypercolumn techniques to detection of the Alzheimer’s disease stages by deep learning model, Neural Comput. Appl. 33 (2021) 9877–9889. doi:10.1007/s00521-021-05758-5.
E. Başaran, A new brain tumor diagnostic model: Selection of textural feature extraction algorithms and convolution neural network features with optimization algorithms, Comput. Biol. Med. 148 (2022) 105857. doi:https://doi.org/10.1016/j.compbiomed.2022.105857.