DTBV: A Deep Transfer-Based Bone Cancer Diagnosis System Using VGG16 Feature Extraction

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Bone cancer is a lethal and rare disease requiring early diagnosis for improved outcomes.
• Manual bone cancer detection is complex and time-consuming.
• Existing diagnostic methods lack sufficient accuracy and efficiency.

Purpose of the Study:

• To develop an automated deep transfer-based bone cancer diagnosis (DTBV) system.
• To enhance the accuracy and efficiency of bone cancer detection using artificial intelligence.
• To address the limitations of manual bone cancer diagnosis.

Main Methods:

• A deep transfer learning (TL) approach was employed, utilizing the VGG16 model for feature extraction from X-ray images.
• Mutual information was used for selecting the most relevant features.
• A support vector machine (SVM) classifier was trained on the selected features to differentiate between cancerous and healthy bone tissue.

Main Results:

• The proposed DTBV system achieved a high accuracy of 93.9% in detecting bone cancer.
• The VGG16 model effectively extracted discriminative features from bone X-ray images.
• The method demonstrated superior performance compared to existing bone cancer detection systems.

Conclusions:

• The DTBV system offers a highly efficient and accurate solution for bone cancer diagnosis.
• This novel application of transfer learning and mutual information for bone cancer detection shows significant promise.
• The developed system can aid clinicians in early and reliable diagnosis, potentially reducing mortality rates.