• Puneet Misra
• V. K. Saraswat
• Mohd. Usman Siddharth
• Rakesh Srivastava

This study emphasizes the pressing need for improved laryngeal cancer detection. Current diagnostic methods, primarily endoscopic imaging, are hindered by the increasing volume of patient data and the potential for invasive biopsies to cause lasting damage. To mitigate these challenges, an artificial intelligence-driven framework is proposed, leveraging deep learning to enhance diagnostic accuracy and efficiency. This framework integrates VGG16 for robust feature extraction with an attention mechanism and an LSTM decoder, enabling precise identification of laryngeal cancer. By incorporating channel and spatial attention layers, the model focuses on relevant image features, improving detection accuracy. The suggested strategy is to address issues with prediction accuracy, resource use, and real-time performance to provide a more effective and assessable tool for early laryngeal cancer detection and better patient outcomes.

Laryngeal cancer, endoscopic images, artificial intelligence, deep learning, VGG16, encoder, decoder, CBAM, LSTM

"Improving Laryngeal Cancer Detection Accuracy Through Deep Learning and Attention Mechanisms", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 4, page no.c607-c613, April-2025, Available :http://www.jetir.org/papers/JETIR2504278.pdf

"Improving Laryngeal Cancer Detection Accuracy Through Deep Learning and Attention Mechanisms", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 4, page no. ppc607-c613, April-2025, Available at : http://www.jetir.org/papers/JETIR2504278.pdf