Scholarlink

This paper is focused on deep learning as well as machine learning techniques for brain tumors, diabetes, and heart disease detection system. For these disease predictions, the study uses a hybrid model that incorporates convolutional neural network from DL and support vector machine models from ML. Users of this suggested detection technique can obtain precise estimations of their risk of contracting certain diseases by simply entering pertinent medical data. The dataset is taken from the Kaggle for these diseases, which consisted of a variety of CT scan images. Deep learning uses a support vector machine with linear kernel in conjunction with CNN as the foundational model for image recognition. Feature selection and hyper-tuning is performed of CNN model before going to training of model. This suggested model shows the potential of machine learning in the early identification and prevention of these three severe diseases. This proposed model that is applied on the image’s dataset of various diseases, giving accuracy around 98%.

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

Brain Tumors, Diabetes, and Heart Disease Detection System Using Its Medical Dataset and Ensemble of CNN and SVM Model

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Vishan Kumar Gupta,
  • Paras Jain,
  • Sarvesh Vishwakarma,
  • Harish Tiwari,
  • Vidisha,
  • Nilakshi Goel

摘要

This paper is focused on deep learning as well as machine learning techniques for brain tumors, diabetes, and heart disease detection system. For these disease predictions, the study uses a hybrid model that incorporates convolutional neural network from DL and support vector machine models from ML. Users of this suggested detection technique can obtain precise estimations of their risk of contracting certain diseases by simply entering pertinent medical data. The dataset is taken from the Kaggle for these diseases, which consisted of a variety of CT scan images. Deep learning uses a support vector machine with linear kernel in conjunction with CNN as the foundational model for image recognition. Feature selection and hyper-tuning is performed of CNN model before going to training of model. This suggested model shows the potential of machine learning in the early identification and prevention of these three severe diseases. This proposed model that is applied on the image’s dataset of various diseases, giving accuracy around 98%.