Advances in Medical Image Processing Using Convolutional Neural Networks

Strategic Insights -

Researchers at Universitas Trisakti have conducted a systematic literature review on the implementation of Convolutional Neural Networks (CNNs) in medical image processing, specifically in lung X-ray imaging. The study found that CNN architectures have significantly enhanced the accuracy of lung disease detection and support both segmentation and classification tasks. However, challenges such as dataset variability, model generalization, and ethical implications remain.

Key Takeaways:

  • The study evaluated 15 relevant studies published between 2019 and 2023, focusing on CNN architectures such as VGG, ResNet, AlexNet, and GoogLeNet.
  • CNN-based models demonstrated high effectiveness in segmentation and classification tasks, particularly for chest X-ray images used in diagnosing lung-related diseases.
  • The research highlights the potential of CNNs in medical imaging, emphasizing their accuracy and supporting both segmentation and classification tasks.
  • However, challenges such as dataset variability, model generalization, and ethical implications require further research and attention.
  • The study emphasizes the need for further research in CNN-based models, specifically in addressing dataset variability and model generalization.
  • The findings of the study suggest that CNN architectures have the potential to revolutionize medical image processing and diagnosis.

Statistics:

  • 15 studies were selected for the systematic literature review, published between 2019 and 2023.
  • The studies focused on CNN architectures such as VGG, ResNet, AlexNet, and GoogLeNet.
  • The research found that CNN-based models demonstrated high effectiveness in segmentation and classification tasks, with an accuracy rate of (statistic not provided).
  • The study emphasizes the need for further research in addressing dataset variability and model generalization.

Sources:

  • Literacy Review Study on the Implementation of Convolutional Neural Network Architecture in Segmentation and Classification of Lung Medical Images. JISA (Jurnal Informatika dan Sains), 2025,8(1):1-7.