The technique of deep learning was applied with the use of Convolutional Neural Network when analyzing skin disease using the DermaNet dataset on Kaggle so that it can ensure diagnosis at an earlier time in its development to treat the diseases. The model chosen is Incep- tionV3 that is renowned for its capability in recognizing images. The model was trained, vali- dated, and tested on the DermaNet dataset so it could diagnose patients with a high accuracy rate. It also contained major preprocessing steps such as flattening and normalization to improve its performance. For better accessibility, the model was combined with a Telegram bot. This bot allows users to upload images of skin conditions via the Telegram app which then is analyzed by the InceptionV3 model for a likely diagnosis. This solution speaks for itself by combining state- of-the-art AI and machine learning through a very intuitive interface and promotes early skin disease detection, improving healthcare accessibility and outcomes.

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

AI-Powered Skin Disease Diagnosis: Leveraging Convolutional Neural Networks and Real-Time Accessibility via Telegram Bot

  • Soja Rani,
  • S. Sivaramakrishnan,
  • Ishika Dayal,
  • Manisha Gupta,
  • N. C. Deepak

摘要

The technique of deep learning was applied with the use of Convolutional Neural Network when analyzing skin disease using the DermaNet dataset on Kaggle so that it can ensure diagnosis at an earlier time in its development to treat the diseases. The model chosen is Incep- tionV3 that is renowned for its capability in recognizing images. The model was trained, vali- dated, and tested on the DermaNet dataset so it could diagnose patients with a high accuracy rate. It also contained major preprocessing steps such as flattening and normalization to improve its performance. For better accessibility, the model was combined with a Telegram bot. This bot allows users to upload images of skin conditions via the Telegram app which then is analyzed by the InceptionV3 model for a likely diagnosis. This solution speaks for itself by combining state- of-the-art AI and machine learning through a very intuitive interface and promotes early skin disease detection, improving healthcare accessibility and outcomes.