Deep Learning Approaches for Classification of Breast Cancer in Ultrasound (US) Images

Abstract

Breast cancer is one of the deadliest cancer types affecting women worldwide. As with all types of cancer, early detection of breast cancer is of vital importance. Early diagnosis plays an important role in reducing deaths and fighting cancer. Ultrasound (US) imaging is a painless and common technique used in the early detection of breast cancer. In this article, deep learning-based approaches for the classification of breast US images have been extensively reviewed. Classification performance of breast US images of architectures such as AlexNet, VGG, ResNet, GoogleNet and EfficientNet, which are among the most basic CNN architectures, has been compared. Then, transformer models, which are one of the most popular deep learning architectures these days and show similar performance to the performance of CNN' architectures in medical images, are examined. BUSI, the only publicly available dataset, was used in experimental studies. Experimental studies have shown that the transformer and CNN models successfully classify US images of the breast. It has been observed that vision transformer model outperforms other models with 88.6% accuracy, 90.1% precison, 87.4% recall and 88.7% F1-score. This study shows that deep learning architectures are successful in classification of US images and can be used in the clinic experiments in the near future.

CerviFormer: A pap smear‐based cervical cancer classification method using cross‐attention and latent transformer

“…The study Pacal and Kılıcarslan 45 provides a comprehensive analysis of 40 CNN‐based models and over 20 ViT‐based models on the Sipakmed pap smear dataset. The experimental findings indicate that the most recent ViT‐based models exhibit superior performance, while the CNN models currently in use demonstrate comparable effectiveness to the ViT models.…”

Enhancing Computer-Aided Cervical Cancer Detection Using a Novel Fuzzy Rank-Based Fusion

“…Haryanto et al [55] utilized the AlexNet architecture with a non-padding scheme and achieved an accuracy of 87.32%. Pacal et al [19] used ViT based Max-voting ensemble and reported 92.95% accuracy in the SIPaKMeD dataset. In [56], the authors applied segmentation followed by classification using five machine learning-based classifiers and reported an accuracy of 94.09%.…”

“…The authors in [18] used ViT and DenseNet161 to classify cervical cytology images and attained 68% accuracy. Pacal et al [19] addresses the issues of data quality and image variability by employing several advanced deep-learning techniques. They employed more than 40 CNN and 20 ViTbased models on the SIPaKMeD pap-smear dataset and reported the ViT models' superior performance with data augmentation and ensemble learning.…”

Detection of Monkeypox Among Different Pox Diseases with Different Pre-Trained Deep Learning Models

“…Deep learning methods are widely used in many medical classification problems, such as the classification of dermatological diseases (Zhou et al, 2022), cardiovascular diseases (Li et al, 2023), Alzheimer's disease (Hu et al, 2022), Parkinson's disease (Rezaee et al, 2022), chest diseases (Ibrahim et al, 2021), colon cancer and diseases (Pacal et al, 2020;Pacal and Karaboga, 2021) breast cancer (İ. Pacal, 2022), and brain tumors (Jia and Chen, 2020). Likewise, studies have been conducted on the classification and diagnosis of the above-mentioned diseases using ML methods (Aljaddouh and Malathi, 2022;Bhattacharjee et al, 2022;Ferreira et al, 2022;Shinde et al, 2022;Swathy and Saruladha, 2022;Vankdothu and Hameed, 2022;Vuidel et al, 2022).…”