Screening for Normal Pressure Hydrocephalus on Head CT Using Automated Callosal Angle Assessment

Area of Science:

• Neurology
• Radiology
• Medical Imaging

Background:

• Normal pressure hydrocephalus (NPH) is an underdiagnosed condition in older adults, causing gait impairment and fall risk.
• The callosal angle is a known imaging biomarker for NPH, but its application on routine head CT needs validation.
• Early diagnosis of NPH is crucial for timely intervention and management.

Purpose of the Study:

• To validate an automated measurement of the callosal angle on routine head CT scans for NPH detection.
• To compare automated measurements with manual assessments performed by radiologists.
• To evaluate the diagnostic performance of the automated callosal angle measurement as a screening tool for NPH.

Main Methods:

• Retrospective analysis of 198 patients with probable NPH and 198 matched controls.
• Manual callosal angle measurements by residents and neuroradiologists.
• Comparison of automated and manual measurements using intraclass correlation (ICC) and diagnostic accuracy metrics.

Main Results:

• Automated callosal angle measurements showed strong agreement with manual measurements (ICC = 0.90).
• An automated threshold of <90° yielded 84.1% accuracy, 90.4% sensitivity, and 77.8% specificity.
• An optimized threshold of 95° achieved 85.9% accuracy, sensitivity, and specificity, with an AUC of 0.915.

Conclusions:

• Automated callosal angle assessment on routine head CT is a reliable and scalable method for NPH detection.
• This automated tool can facilitate earlier diagnosis and treatment of NPH, a reversible cause of dementia.
• The validated automated measurement supports its use as a screening tool for NPH.