Breast Cancer Detection Using Streamlit Frame Work

  • Unique Paper ID: 166866
  • Volume: 11
  • Issue: 2
  • PageNo: 2305-2310
  • Abstract:
  • Breast cancer is still a major global health concern, requiring sophisticated and effective diagnostic instruments. This research proposes an innovative method for automated breast cancer categorization using deep learning algorithms linked to the Streamlit framework. Three modalities are the subject of the study: histopathology, mammography, and ultrasound. Histopathology is given special attention for a thorough examination. To standardize input photos for feature extraction later on, the method starts with image preprocessing, which includes scaling and grayscale conversion. To capture important traits indicative of malignant tissues, key statistical variables like variance, mean, and median are computed. Reliable model training and assessment are guaranteed by dataset enrichment and splitting techniques. Convolutional neural networks (CNNs) and VGG-19, which are modified for binary classification tasks, are two different deep learning architectures that are examined. Several layers of convolution and pooling are incorporated into the CNN model, which then adds dense layers to optimize accuracy and efficiency. In contrast, the pre-trained VGG-19 model on ImageNet is refined and assessed for its implementation and learning performance in identifying features associated with breast cancer. The efficiency of the suggested method is demonstrated by the experimental results, which achieve a high degree of accuracy in differentiating between benign and malignant breast tissues. The applied models' performance indicators, including validation scores, loss, and accuracy, are carefully examined and contrasted, exposing the advantages and disadvantages of each technique. Additionally, by incorporating the Streamlit framework, an intuitive interface for submitting, processing, and displaying diagnostic information is provided, improving user engagement and visualization. The web-based tool, which serves both researchers and healthcare practitioners, guarantees usability and accessibility. In summary, our work adds to the current endeavors to improve breast cancer diagnostics by utilizing deep learning and approachable frameworks such as Streamlit. The results highlight the potential of cutting-edge machine learning methods to enhance healthcare outcomes and open the door for further advancements in computer-assisted diagnostic tool development.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 2
  • PageNo: 2305-2310

Breast Cancer Detection Using Streamlit Frame Work

Related Articles