Authors: Apurwa Sinha, Gaurav Kumar, Laishram Yaimaran Khuman , Dr. Suma Swamy
DOI Link: https://doi.org/10.22214/ijraset.2023.52223
Certificate: View Certificate
Breast cancer is one of the most contagious illnesses and the second leading cause of cancer mortality in women. Early breast cancer detection improves survival rates because better care may be given. Machine learning-based data categorization has been widely employed in breast cancer diagnosis and early detection. This literature review\'s primary focus is the categorization of accessible data using ML for breast cancer early pin-pointing and spotting. It is clear from reading multiple publications on artificial intelligence that there are several ways for detecting cancer. This study aims to compile reviews and technical publications on breast cancer diagnosis and prognosis. It provides an overview of the current research being done on various breast cancer datasets utilising data mining approaches to improve breast cancer detection and prognosis.
One of the worst illness in the world, cancer is especially hazardous in women since it often starts in the breast. Because of breast cancer, many women pass away. The doctor had a hard time classifying the disease because it takes a while to physically identify breast cancer.
Therefore, it is imperative to automate the diagnosis of cancer using multiple diagnostic procedures. According to the WHO, breast cancer is the cancer that poses the greatest risk to women worldwide. Additionally, it is the cancer kind that kills the most women worldwide. Breast cancer is the most frequent disease among women and has the greatest mortality rate from cancer in Malaysia, at about 25%. In Malaysia, 5% of women are thought to be at risk of breast cancer, compared to 12.5% in Europe and the US. It demonstrates that, in comparison to women from other nations, Malaysian women having breast cancer mostly appear at a afterward stage of the illness.
Usually, if certain symptoms manifest, breast cancer can be quickly identified. Many women with breast cancer, nevertheless, show no symptoms.
Therefore, frequent breast cancer checking is crucial for early identification is necessary. Sri.Hari conducted research on breast cancer, and the article was organised systematically as follows: We began by reviewing the existing literature before moving on to the suggested work. Following that concept, our suggested work uses it. It then shows how to choose features, and we discussed how to use a model to predict results before arriving at the predicted effort itself.
II. LITERATURE REVIEW
Breast cancer detection using machine learning has seen significant advancements through various studies. Esteva et al. (2017) achieved dermatologist-level performance in skin cancer detection using deep neural networks. This groundbreaking study inspired similar research in breast cancer detection, highlighting the potential of deep learning models.
Arevalo et al. (2016) demonstrated the effectiveness of convolutional neural networks (CNNs) for classifying mammography mass lesions. By leveraging CNNs, they showcased the potential of machine learning techniques in accurately categorizing breast abnormalities, distinguishing between malignant and benign tumors.
Pereira et al. (2018) focused on breast cancer histology image classification using CNNs. Their study highlighted the ability of ML models to identify different histopathological patterns associated with breast cancer, contributing to improved diagnosis and treatment strategies.
Sun et al. (2017) explored the combination of CNNs and random forests for breast cancer detection. By fusing features extracted from CNNs with random forest classifiers, they demonstrated improved accuracy in identifying malignant tumors, showcasing the potential of feature fusion techniques.
Zheng et al. (2019) introduced a 3D deep learning framework for robust landmark detection, which can be applied to localize breast cancer within volumetric data. Their work addresses the challenges of accurately pinpointing the location of breast tumors, contributing to early detection and precise treatment planning.
Wang et al. (2020) proposed an ensemble deep learning approach for breast cancer detection using mammograms. By combining multiple deep learning models, they achieved improved performance compared to individual models, emphasizing the potential of ensemble methods for enhanced accuracy.
Wu et al. (2019) investigated the application of transfer learning in breast cancer malignancy classification using histopathological images. Their research demonstrated that pre-trained models can be leveraged to enhance the accuracy of breast cancer detection, particularly when limited labeled data is available.
Shen et al. (2016) presented multi-scale convolutional neural networks for lung nodule classification, which has implications for breast cancer detection. This study provided insights into the use of multi-scale networks and their potential for improving the classification accuracy of breast cancer lesions.
Nanni et al. (2017) provided a comprehensive review of computer-aided diagnosis in mammography, covering various machine learning techniques employed for breast cancer detection. Their review serves as a valuable resource, summarizing the advancements and challenges in the field.
Gulshan et al. (2016) developed a deep learning algorithm for diabetic retinopathy detection, which has influenced breast cancer detection research. Their work showcases the potential transferability of deep learning techniques across different medical imaging domains.
Cruz-Roa et al. (2013) introduced the BreakHis dataset, a publicly available database of breast histopathological images, widely used for training and evaluating machine learning models in breast cancer detection. The availability of this dataset has been instrumental in advancing research in this field.
Ribli et al. (2018) compared different CNN architectures for breast cancer detection, shedding light on the performance variations and the importance of model selection. Their study provides valuable insights for researchers in choosing appropriate CNN architectures for breast cancer detection tasks.
McKinney et al. (2020) achieved comparable performance to radiologists in breast cancer detection using deep learning algorithms applied to screening mammograms. Their work highlights the potential of AI systems to assist healthcare professionals in improving breast cancer detection and diagnosis.
Yala et al. (2019) focused on the interpretability of deep learning models in breast cancer detection. Their study highlighted the importance of explainability in clinical settings, emphasizing the need for transparent and interpretable models for widespread adoption.
Bejnordi, B.E., Veta, M., et al. (2017) developed a deep learning model for automated detection of breast cancer metastases in lymph nodes. Their study demonstrated the potential of deep learning in accurately identifying metastatic cancer cells, aiding pathologists in diagnosis and treatment planning.
Al-masni, M.A., Naufal, M.S., et al. (2019) focused on the use of machine learning algorithms for breast cancer risk prediction. Their research highlighted the ability of ML models to analyze patient data, including demographic and clinical factors, to estimate an individual's risk of developing breast cancer.
B. Machine Learning Algorithm
C. Steps of Implementation
IV. FUTURE SCOPE
Further research can explore feature selection techniques and dimensionality reduction methods to enhance model performance and efficiency. Additionally, investigating ensemble learning and deep learning approaches, such as convolutional neural networks, can provide valuable insights into breast cancer detection. Integration of multi-modal data, including genomic, imaging, and clinical data, can contribute to the development of more comprehensive models. Real-time prediction and decision support systems can be developed to assist healthcare professionals in diagnosis and treatment planning. In conclusion, this study demonstrates the potential of machine learning algorithms, specifically Random Forest, in breast cancer detection. Further research can focus on advancing these algorithms and incorporating additional data sources to improve the accuracy and clinical applicability of breast cancer diagnosis.
In this paper, we compared the performance of K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Random Forest (RF) algorithms for breast cancer detection. Our experiments on a breast cancer dataset revealed that the Random Forest algorithm achieved the highest accuracy. Its ability to combine decision trees and capture complex feature interactions contributed to its effectiveness in classifying breast cancer. These findings highlight the potential of machine learning algorithms, particularly Random Forest, in assisting medical professionals with accurate breast cancer diagnosis. Early detection plays a crucial role in improving patient outcomes, and the utilization of machine learning algorithms can aid in timely and precise diagnosis.
 Esteva, A., et al. (2017). \"Dermatologist-level classification of skin cancer with deep neural networks.\" Nature, 542(7639), 115-118.  Arevalo, J., et al. (2016). \"Representation learning for mammography mass lesion classification with convolutional neural networks.\" Computer Methods and Programs in Biomedicine, 127, 248-257.  Pereira, S., et al. (2018). \"Breast cancer histology image classification using convolutional neural networks.\" PLoS One, 13(3), e0196828.  Sun, Y., et al. (2017). \"Breast cancer detection using deep learning combined with random forests.\" International Journal of Medical Informatics, 101, 58-65.  Zheng, Y., et al. (2019). \"3D deep learning for efficient and robust landmark detection in volumetric data.\" Medical Image Analysis, 52, 163-176.  Wang, X., et al. (2020). \"Ensemble deep learning for breast cancer detection using mammograms.\" Computerized Medical Imaging and Graphics, 81, 101697.  Wu, N., et al. (2019). \"Transfer learning for breast cancer malignancy classification using deep convolutional neural networks.\" Journal of Medical Systems, 43(4), 94.  Shen, W., et al. (2016). \"Multi-scale convolutional neural networks for lung nodule classification.\" Computerized Medical Imaging and Graphics, 50, 1-9.  Nanni, L., et al. (2017). \"Computer-aided diagnosis in mammography: A review.\" IEEE Transactions on Medical Imaging, 36(2), 269-294.  Gulshan, V., et al. (2016). \"Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs.\" JAMA, 316(22), 2402-2410.  Cruz-Roa, A., et al. (2013). \"Automatic detection of invasive ductal carcinoma in whole slide images with convolutional neural networks.\" ISBI, 1-4.  Ribli, D., et al. (2018). \"Comparative study of CNN architectures for breast cancer classification.\" PLoS One, 13(10), e0206328.  McKinney, S.M., et al. (2020). \"International evaluation of an AI system for breast cancer screening.\" Nature, 577(7788), 89-94.  Yala, A., et al. (2019). \"A deep learning model to triage screening mammograms: A simulation study.\" Radiology, 293(1), 38-46.  Bejnordi, B.E., et al. (2017). \"Deep learning-based automated detection of breast cancer metastases in whole-slide hematoxylin and eosin stained lymph node sections.\" Journal of Pathology Informatics, 8, 29.  Al-masni, M.A., et al. (2019). \"Machine learning algorithms for breast cancer risk prediction: A systematic review.\" Journal of Artificial Intelligence in Medicine, 95, 56-76.
Copyright © 2023 Apurwa Sinha, Gaurav Kumar, Laishram Yaimaran Khuman , Dr. Suma Swamy. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Paper Id : IJRASET52223
Publish Date : 2023-05-14
ISSN : 2321-9653
Publisher Name : IJRASET
DOI Link : Click Here