Author Login IJIRT Partner Network (IPN) Login

Breast Cancer Classification Based on 1D-Deep Resnet

  • Unique Paper ID: 168530
  • Volume: 11
  • Issue: 5
  • PageNo: 1214-1221
  • Abstract:
  • Infrequent changes in a woman's health in recent years can be a sign of breast cancer. Consequently, identifying and categorizing breast cancer is a major diagnostic problem. Because they are a quick, non-invasive, and inexpensive diagnostic method, electrocardiograms, also known as mammograms, are frequently used to diagnose cancer. However, misinterpretation of cancer occurs due to the unpredictable nature of cancer parts and the susceptibility of mammograms to noise. Additionally, it takes effort and is prone to error to manually identify breast cancer using data from mammograms. Deep learning (DL), which outperforms conventional machine learning models, is the recommended method for quickly and automatically classifying the signals from mammograms. For an automatic breast cancer categorization (ABC) system, the proposed study suggests a unique deep learning architecture called a 1-dimensional deep residual neural network (1D-deep ResNet). The model was trained and assessed using signals from the Wisconsin (Diagnostic) dataset of mammography images. This suggested model makes two distinct contributions. Initially, the up-sampling technique is used to rectify the uneven mammography image data in order to minimize noise and avoid biased prediction results. Next, 1D-Deep ResNet is used to automatically classify the signals from the imbalanced mammograms. The BC system attempts to address the problems with conventional electrocardiograms (mammograms) in the detection of cancer. These problems can be caused by noise and unpredictable occurrences, which can result in errors and misinterpretations. The model performance is assessed by computing the accuracy, AUC, precision, recall, and f1 score using the confusion matrix. With an AUC score of 95.9% in the training dataset and 95.94% in the testing dataset, the tests' findings demonstrate that the suggested approach worked effectively. The suggested model performs better than other deep learning techniques now in use, such as CNN, LSTM, and GRU, and it will greatly reduce the amount of time that physicians must spend classifying the signals from mammograms. In contrast to other deep learning algorithms already on the market, our research indicates that the 1D-deep ResNet is more suited for automated breast cancer classification in the future.
Download article email to a friend

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 5
  • PageNo: 1214-1221

Breast Cancer Classification Based on 1D-Deep Resnet

Related Articles

04 Oct, 2024

ANALYSIS OF CORONARY DISEASES USING MACHINE LEARNING MODELS (ACDMLM)

Read More
Impact Factor
8.01 (Year 2024)
UGC Approved
Journal no 47859

Join Our IPN

IJIRT Partner Network

Submit your research paper and those of your network (friends, colleagues, or peers) through your IPN account, and receive 800 INR for each paper that gets published.

Join Now

Latest Publication

Archive

View More

Recent Conferences

NCSEM 2024

National Conference on Sustainable Engineering and Management - 2024 Last Date: 15th March 2024

Submit inquiry