A Deep Learning Approach for Lung Cancer Detection: AddressingOverfitting and Enhancing Model Generalization
محل انتشار: سومین کنفرانس ملی محاسبات نرم و علوم شناختی
سال انتشار: 1403
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 145
فایل این مقاله در 11 صفحه با فرمت PDF قابل دریافت می باشد
- صدور گواهی نمایه سازی
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
SCCS03_010
تاریخ نمایه سازی: 15 بهمن 1403
چکیده مقاله:
Lung cancer is one of the leading causes of cancer-related deaths worldwide,emphasizing the need for accurate and early detection methods. This study investigates adeep learning-based framework for lung cancer prediction, employing a ConvolutionalNeural Network (CNN) architecture for imaging data analysis. The model achieved atraining accuracy of ۹۵% but demonstrated a testing accuracy of only ۷۵.۶%, indicatingchallenges with generalization and overfitting. Early stopping, learning rate adjustments,and dropout layers were used during training, but further improvements are necessary. Thefindings highlight the potential of CNNs for lung cancer detection while underlining theimportance of addressing data diversity, model complexity, and regularization to enhanceperformance. Future work will focus on integrating advanced techniques like transferlearning, data augmentation, and explainable AI (XAI) for robust and interpretable cancerdetection solutions.
کلیدواژه ها:
نویسندگان
Mahsa Yaghoobi
Department of Computer Engineering, Ardabil Branch, Islamic Azad University, Ardabil, Iran;
Shiva Razzagzadeh
Department of Computer Engineering, Ardabil Branch, Islamic Azad University, Ardabil, Iran;
Zeynab Rezaei
Department of Computer Engineering, Ardabil Branch, Islamic Azad University, Ardabil, Iran;