CT‐based volumetric measures obtained through deep learning: Association with biomarkers of neurodegeneration

• DOI: 10.1002/alz.13445
• Publication Date: 2023-09-28T11:08:20Z
• Authors: Meera Srikrishna; Nicholas J. Ashton; Alexis Moscoso; Joana B. Pereira; Rolf A. Heckemann; Danielle van Westen; Giovanni Volpe; Joel Simrén; Anna Zettergren; Silke Kern; Lars‐Olof Wahlund; Bibek Gyanwali; Saima Hilal; Joyce Chong Ruifen; Henrik Zetterberg; Kaj Blennow; Eric Westman; Christopher Chen; Ingmar Skoog; Michael Schöll
• Keywords: cognition; plasma biomarkers; deep learning; Brain segmentation; CSF biomarkers; Research Articles; 3. Good health; CT; dementia

Abstract INTRODUCTION Cranial computed tomography (CT) is an affordable and widely available imaging modality that used to assess structural abnormalities, but not quantify neurodegeneration. Previously we developed a deep‐learning–based model produced accurate robust cranial CT tissue classification. MATERIALS AND METHODS We analyzed 917 744 magnetic resonance (MR) scans from the Gothenburg H70 Birth Cohort, 204 241 MR participants of Memory Clinic Singapore. tested associations between six CT‐based volumetric measures (CTVMs) existing clinical diagnoses, fluid biomarkers, cognition. RESULTS CTVMs differentiated cognitively healthy individuals dementia prodromal patients with high accuracy levels comparable MR‐based measures. were significantly associated cognition biochemical markers DISCUSSION These findings suggest potential future use as informative first‐line examination tool for neurodegenerative disease diagnostics after further validation. Highlights Computed (CT)–based can distinguish controls, well controls. associate relevant cognitive, biochemical, neuroimaging diseases. Model performance, in terms brain classification, was consistent across two cohorts diverse nature. Intermodality agreement our automated established (MR)–based image segmentations stronger than visual assessment.