Deep hashing and attention mechanism-based image retrieval of osteosarcoma scans for diagnosis of bone cancer

• Published in: Journal of bone oncology 2024-12, Vol.49, p.100645, Article 100645
• Main Authors: Zeng, Taisheng, Ye, Yuguang, Chen, Yusi, Zhu, Daxin, Huang, Yifeng, Huang, Ying, Chen, Yijie, Shi, Jianshe, Ding, Bijiao, Huang, Jianlong, Ling, Mengde
• Format: Article
• Language: English
• Subjects: Attention network; Deep hashing; Medical image retrieval; Osteosarcoma; Research Paper
• ISSN: 2212-1374; 2212-1366; 2212-1374
• DOI: 10.1016/j.jbo.2024.100645

•The algorithm uses deep hashing and attention mechanisms to improve efficiency and accuracy in osteosarcoma image retrieval.•Adaptive histogram equalization and data augmentation enhance image quality and diversity for better feature extraction.•The WRN-AM model maps high-dimensional features to low-dimensional hash codes, enabling fast, precise similarity matching. Due to its intricate nature and substantial data size, microscopic image data of osteosarcoma often present a significant obstacle to the effectiveness of conventional image retrieval methods. Therefore, this study investigates a new approach for medical image retrieval using advanced deep hashing techniques and attention mechanisms to address these challenges more effectively. The proposed algorithm significantly improves osteosarcoma cell microscopic image retrieval efficiency and accuracy using deep hashing and attention mechanisms. Image preprocessing includes adaptive histogram equalization and dataset augmentation to enhance quality and diversity. Feature extraction employs the WRN-AM model to map high-dimensional features to a low-dimensional hash code space, improving retrieval efficiency. Finally, similarity matching via Hamming distance allows rapid and precise identification of similar images. The study shows notable advancements: the WRN-AM model achieves 93.2% classification accuracy and 97.09% mAP using 64-bit hash codes. These findings underscore the technique’s effective performance in extracting and categorizing diverse microscopic cell data efficiently and reliably. This innovative approach provides a robust solution for retrieving and classifying microscopic data of osteosarcoma cells and other cell types, speeding up clinical diagnosis and medical research. It facilitates quicker access and analysis of patient image data, enhancing diagnostic precision and treatment planning for healthcare professionals. Concurrently, it supports researchers in leveraging medical image data more efficiently, fostering progress and innovation in the medical field.