Neural Network-Based Image Processing for Vitamin Deficiency Detection Using CNN

This project explores the use of Convolutional Neural Networks (CNNs) for detecting vitamin deficiencies through image processing. The process begins by executing a code that facilitates the selection of a body part—tongue, lips, nails, or eyes—based on the user’s choice. After selecting a specific image of the chosen body part, the image undergoes preprocessing steps to enhance quality and features. The CNN is then trained using these preprocessed images, employing various layers and training options tailored to detect specific deficiency indicators. For instance, if the tongue is selected, the CNN classifies symptoms such as smooth texture, red color, glossitis, or an unclear mouth, each corresponding to potential deficiencies. Similarly, if lips are chosen, classifications may include cracked lips, shiny red appearance, and other related symptoms. The final output displays the detected deficiency based on the image analysis, facilitating early diagnosis and intervention. This approach leverages deep learning to provide accurate and automated vitamin deficiency detection, showcasing the efficacy of CNNs in medical image analysis and preventive healthcare.

Keywords: Vitamin Deficiency Related Dataset, Deep Learning, Convolution Neural Network, Image Processing Techniques, accuracy.