Related Experiment Video
Updated: May 1, 2026

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Early Detection of Traumatic Brain Hematoma Using Machine Learning-Powered Near-Infrared Spectroscopy: A Case Report
Neil Manjunath Salian1, Dhaval Shukla, Subhas Konar
1Department of Neurosurgery, National Institute of Mental Health and Neuro Sciences, Bengaluru, Karnataka, India.
Abstract:
Traumatic brain injury is a significant concern in contact sports, with concussions being the most common type. Kabaddi, a high-contact sport, carries a high risk of head injuries, yet limited research has addressed their acute neurological impact. Conventional imaging such as computed tomography is the gold standard for hematoma detection but is often inaccessible in field settings. This case report describes the first use of machine learning-powered near-infrared spectroscopy (mm) to detect a traumatic brain hematoma in a 13-year-old Kabaddi player. After a head injury, near-infrared spectroscopy indicated a suspected hematoma, which was confirmed by computed tomography showing a right frontoparietal subdural hematoma with a 6-mm midline shift and a left temporal bone fracture. mNIRS proved to be a rapid, noninvasive screening tool for on-field assessment. Further research is warranted to optimize its clinical utility and integration into sports medicine protocols.

