Authors:
J. Angelin Jeba, S. Rubin Bose, V. Karrthik Kishore, Bushra Rehman
Addresses:
Department of Electronics and Communication Engineering, S.A. Engineering College, Chennai, Tamil Nadu, India. Department of Electronics and Communication Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Institute of Pathology and Diagnostic Medicine, Khyber Medical University, Hayatabad, Peshawar, Pakistan.
Abstract:
Brain cancer, including gliomas, meningiomas, and pituitary tumors, is complex and difficult to diagnose and cure. Brain tumors account for 2% of cancer cases in India, affecting children and adults. Despite advances in imaging tools like MRI, CT, and PET, precise early detection improves results, especially in regions with limited specialist healthcare. This study uses a modified DETR (identification Transformer) model for brain tumor identification using real-time object detection and localization. DETR's advanced feature extraction and bounding box prediction enable precise brain tumor type identification. The proposed approach predicts tumor grades using a deep learning model trained on a large dataset of annotated medical images with a recall of 0.991, precision of 0.996, and F1 score of 0.995. This research combines the improved DETR model's real-time detection with standard imaging modalities to increase diagnosis accuracy, assessment time, and radiologists' decision-making. In India, brain cancer management is hindered by a lack of advanced diagnostic tools and high treatment costs. The updated DETR model's strengths are combined with neuro-oncology clinical procedures in this AI-driven method to improve brain cancer detection and patient outcomes.
Keywords: Brain Cancer; Gliomas and Meningiomas; Excess Hormones; Imaging Techniques; AI and ML; Healthcare Landscape; Malignant Brain Tumour; Brain Cancer Detection; Detection Transformer.
Received: 09/04/2024, Revised: 22/07/2024, Accepted: 03/09/2024, Published: 07/12/2024
AVE Trends in Intelligent Health Letters, 2024 Vol. 1 No. 4 , Pages: 193-205