Authors:
J. Angelin Jeba, S. Rubin Bose, V. Sathiesh Kumar, O. Jeba Singh, B. Judy Flavia, R. Regin, S. Suman Rajest, M. Mohamed Sameer Ali

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 Electronics Engineering, MIT Campus, Anna University, Chennai, Tamil Nadu, India. Centre for Academic Research, Alliance University, Bengaluru, Karnataka, India. Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. School of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Department of Research and Development, Dhaanish Ahmed College of Engineering, Chennai, Tamil Nadu, India.

Abstract:

A progressive loss of brain function is the hallmark of Alzheimer's disease (AD), which is frequently linked to a serious deterioration in cognitive abilities. Early detection is therefore essential for successful treatments. Introducing two specialised models for medical imaging and CSF values analysis, this work addresses the critical demand for precise diagnostic instruments in radiology and neurology. The first model, Biomarker, is carefully designed for biomarker analysis and achieves 100% accuracy in AD identification on a dataset of 810 patients. The deployment of a Random Forest Classifier enables this noteworthy result, demonstrating Biomarker's effectiveness in differentiating AD cases from non-AD cases. The second model, YOLOv8, is exceptional at identifying brain anomalies in MRI scans. Using advanced post-processing algorithms, batch normalisation, and convolutional layers, YOLOv8 achieves an impressive mean Average Precision (mAP) of 99.8%. YOLOv8 accurately distinguished among the various phases of cognitive impairment, including extremely mild, mild, moderate, and non-demented cases. Both approaches show great promise for early diagnosis and clinical decision-making when combined with the Biomarker (Random Forest) model for detecting Alzheimer's disease, thereby improving patient outcomes.

Keywords: Random Forest Classifier; Performance Metrics; Convolutional Neural Network; CT-Scan and MRI; Brain Function; Alzheimer's Disease; Radiology and Neurology; Mean Average Precision.

Received on: 01/07/2025, Revised on: 12/09/2025, Accepted on: 11/10/2025, Published on: 09/03/2026

DOI: 10.69888/FTSBE.2026.000673

FMDB Transactions on Sustainable Biomedical Engineering, 2026 Vol. 1 No. 1, Pages: 56–72

• Views : 48
• Downloads : 15

Download PDF