Scholarlink

<p>Among the cancers that pose the greatest threat to life worldwide is lung cancer. According to estimates from the World Cancer Research Fund International, there will be 1.8 million new instances of this disease diagnosed in 2022. When medical personnel diagnose and classify patients’ conditions proactively, they may treat them safely and efficiently. The advent of the microarray method has made it possible to examine the connections between genes and various diseases, including lung malignancies. Numerous methods have been developed to forecast gene-based diseases, but they still have problems with high computational cost, time consumption, complex data, and inaccurate prediction. Therefore, create an efficient lung cancer detection system in this research by designing an Improved Convolutional Neural Network with Honey Bee Mating Optimization (ICNN-HBMO). First, the system is trained using Omix data, and the dataset is normalized using min–max normalization. Then Kernel Principal Component Analysis (KPCA) technique is employed for feature reduction. Furthermore, an enhanced CNN is employed to classify lung cancer using HBMO. The HBMO algorithm optimizes the weight and bias parameters of the ICNN to improve prediction performance. The developed method is implemented in the Matlab tool, and the improved performance is compared to other existing methods. The developed technique attains high accuracy and high precision of 99.2% and 99%.</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

Gene-based lung cancer detection system through omix data and optimized convolutional neural network

  • 期刊论文

  • 发表日期：

    2026年3月3日

查看全文
  • M. Vasanthi,
  • Nouf Saad Aldahwan

摘要

Among the cancers that pose the greatest threat to life worldwide is lung cancer. According to estimates from the World Cancer Research Fund International, there will be 1.8 million new instances of this disease diagnosed in 2022. When medical personnel diagnose and classify patients’ conditions proactively, they may treat them safely and efficiently. The advent of the microarray method has made it possible to examine the connections between genes and various diseases, including lung malignancies. Numerous methods have been developed to forecast gene-based diseases, but they still have problems with high computational cost, time consumption, complex data, and inaccurate prediction. Therefore, create an efficient lung cancer detection system in this research by designing an Improved Convolutional Neural Network with Honey Bee Mating Optimization (ICNN-HBMO). First, the system is trained using Omix data, and the dataset is normalized using min–max normalization. Then Kernel Principal Component Analysis (KPCA) technique is employed for feature reduction. Furthermore, an enhanced CNN is employed to classify lung cancer using HBMO. The HBMO algorithm optimizes the weight and bias parameters of the ICNN to improve prediction performance. The developed method is implemented in the Matlab tool, and the improved performance is compared to other existing methods. The developed technique attains high accuracy and high precision of 99.2% and 99%.