Scholarlink

<p>Across the world, the women were affected by critical types of cancer which was named Cervical Cancer (CC) in their cervix part (which is connected to uterus with vagina) ninety percentage of women cervical cancer disease were linked to Human Papillomavirus (HPV) infection. In most of the residential countries the routine HPV testing has significantly reduced mortality rates but in developing countries the lack of affordable healthcare services continues to pose challenges for providing cost-effective solutions. This highlights the need for accurate algorithms to predict cervical cancer and identify women at risk. In recent years, Deep Learning (DL) architectures include leveraged to build precise predictive models to identify the accuracy in cervical cancer. This proposed work emphasis a novel and simplified transfer learning framework, incorporating VGGNet-19, ResNet -50, DenseNet-201, EfficientNet-B7 and InceptionresNet-V3, to categorize cervical images by the SIPaKMeD dataset. The Advanced DenseNet-201 method is out performed through Various metrics and effective accuracy was achieved compare with existing models this proposed system Resnet-50 which yielded improved result after augmentation and also used to attain the most effective accuracy 99.71% along with Precision, Recall, and F1-scores of 0.99. These outcomes validate the effectiveness of our advance in enabling affordable along with well-organized first-level screening for cervical cancer.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An intelligent system analysis of cervical cancer cells prediction using convolution neural network

  • 期刊论文

  • 发表日期：

    2025年9月8日

查看全文
  • M. Sharon,
  • Vijayakumar Adaickalam

摘要

Across the world, the women were affected by critical types of cancer which was named Cervical Cancer (CC) in their cervix part (which is connected to uterus with vagina) ninety percentage of women cervical cancer disease were linked to Human Papillomavirus (HPV) infection. In most of the residential countries the routine HPV testing has significantly reduced mortality rates but in developing countries the lack of affordable healthcare services continues to pose challenges for providing cost-effective solutions. This highlights the need for accurate algorithms to predict cervical cancer and identify women at risk. In recent years, Deep Learning (DL) architectures include leveraged to build precise predictive models to identify the accuracy in cervical cancer. This proposed work emphasis a novel and simplified transfer learning framework, incorporating VGGNet-19, ResNet -50, DenseNet-201, EfficientNet-B7 and InceptionresNet-V3, to categorize cervical images by the SIPaKMeD dataset. The Advanced DenseNet-201 method is out performed through Various metrics and effective accuracy was achieved compare with existing models this proposed system Resnet-50 which yielded improved result after augmentation and also used to attain the most effective accuracy 99.71% along with Precision, Recall, and F1-scores of 0.99. These outcomes validate the effectiveness of our advance in enabling affordable along with well-organized first-level screening for cervical cancer.