Classification of Lung Cancer using Gene Expressions based on Deep Learning

الموضوعات

علوم تطبيقية وتكنولوجية

الملخص

Lung cancer is one of the most prevalent diseases affecting people and is a main factor in the rising mortality rate. Recently, Machine Learning (ML) and Deep Learning (DL) techniques have been utilized in detecting and classifying several types of cancer,including lung cancer. These methods can assist in enhancing the accuracy of lung cancer classification. In this thesis,aDL model isapplied in combination with feature selection techniques to enhance the efficiency of lung cancer classification using gene expression datasets. A feature selection method examines severalfeature(gene)percentagesto achieve the most reasonable accuracy using the least number of genes. A detailed studyisimplemented to identify the most effective DL model for detecting lung cancer subtypesand their severity by applying multi-label binary classification.In addition, this thesis addresses challenges within the gene expression dataset, such as imbalanced classes and overfitting due to limited sample size and a high featurecount. This study primarily classifies two distinct subtypes of lung cancer, Lung Adenocarcinoma (LUAD) and Lung Squamous Cell Carcinoma (LUSC), and then categorizes them as severe or non-severe cases.Furthermore, an optimization technique is implemented totune the hyperparameters of the DL model to identify the optimal parameter values for achieving the highest classification accuracy.Based on the findings, the optimized CNN model for lung cancer classification resulted in significant improvements in both accuracy and efficiency. The optimized CNN model exhibits remarkable accuracy across varying gene percentages, with notable achievements of0.80 accuracy for lung cancer subtype classification using only five genes and 0.87 accuracy for severity classification.

روابط وملفات

التعريف والنوع

رقم الوثيقة

0779399f-9eba-4864-90cb-380da2bc7573

رقم العقد

0

نوع الوسائط

Crawler

نوع المحتوى

الرسائل العلمية

صيغة المصدر

رسائل ماجيستير

نوع الملف

pdf text

أسماء الملفات

0779399f-9eba-4864-90cb-380da2bc7573_1.pdf

بيانات النشر

ألقاب المؤلفين

[{"name_ar":"Ali Bou Nassif","title_ar":"اشراف","title_en":"Supervision"},{"name_ar":"Jibran Sualeh Muhammad","title_ar":"اشراف","title_en":"Supervision"},{"name_ar":"Nour A M Abujabal ","title_ar":"اعداد","title_en":"Preparation"}]

اللغة

English

المصدر والدورية

اسم المصدر

Classification of Lung Cancer using Gene Expressions based on Deep Learning

المحتوى والصفحات

عدد الصفحات

0

كلمات الباحثين

Diagnostic Imaging Deep learning Medical Informatics Applications Neoplasms -- diagnosis Early Detection Cancer Image Processing, Computer-Assisted --

إشراف وإعداد

الإشراف

Ali Bou Nassif, Jibran Sualeh Muhammad

الإعداد

Nour A M Abujabal

الاقتباسات الببليوغرافية