AI-Assisted Spine X-Ray Analysis for Early Scoliosis Diagnosis Using YOLOv8-Based Object Detection and Automated Cobb Angle Estimation

Manju Darshan M J; Supreetha Gowda H D

AI-Assisted Spine X-Ray Analysis for Early Scoliosis Diagnosis Using YOLOv8-Based Object Detection and Automated Cobb Angle Estimation

Author(s)	Mr. Manju Darshan M J, Dr. Supreetha Gowda H D
Country	India
Abstract	Scoliosis is a musculoskeletal condition involving an abnormal lateral curvature of the spine that, if left undiagnosed, can progress to chronic pain, postural imbalance, and reduced respiratory function. Conventional diagnosis relies on manual radiographic interpretation and Cobb angle measurement performed by radiologists or orthopedic specialists, a process that is reliable but time-intensive and subject to inter-observer variability. This paper presents an automated, deep-learning-driven screening pipeline that detects vertebral structures and spinal abnormalities directly from spine radiographs using a YOLOv8 object-detection model trained on a three-class annotated dataset (Vertebra, Scoliosis Spine, Normal Spine). Detected vertebral bounding-box centroids are used to fit upper and lower spinal reference lines, from which an approximate Cobb angle is computed; the resulting angle is mapped to a four-level severity scale (normal, mild, moderate, severe). The complete pipeline is exposed through a Flask-based web application supporting user authentication, image upload, annotated result visualization, and persistent storage of prediction history in a relational database. Evaluation on held-out validation and test partitions shows strong detection performance, with an overall test mAP@50 of 94.7% and class-wise precision and recall exceeding 0.86 across all three categories. These results indicate that combining real-time object detection with automated angle estimation can provide a practical, reproducible first-pass screening tool, while the system is explicitly positioned as a decision-support aid rather than a replacement for clinical diagnosis.
Keywords	Scoliosis detection; YOLOv8; Cobb angle estimation; spine X-ray; deep learning; medical image analysis; object detection.
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 17, Issue 2, July-December 2026
Published On	2026-07-11

View / Download PDF File

About IJAIDR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us	Message on WhatsApp	+91-9687-182-185	editor@ijaidr.com

Journal of Advances in Developmental Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

AI-Assisted Spine X-Ray Analysis for Early Scoliosis Diagnosis Using YOLOv8-Based Object Detection and Automated Cobb Angle Estimation

Share this