A Novel Approach for Manual Segmentation of the Amygdala and Hippocampus in Neonate MRI

Niloofar Hashempour; Jetro J Tuulari; Harri Merisaari; Kristian Lidauer; Iiris Luukkonen; Jani Saunavaara; Riitta Parkkola; Tuire Lähdesmäki; Satu J Lehtola; Maria Keskinen; John D Lewis; Noora M Scheinin; Linnea Karlsson; Hasse Karlsson

doi:10.3389/fnins.2019.01025

A Novel Approach for Manual Segmentation of the Amygdala and Hippocampus in Neonate MRI

Front Neurosci. 2019 Sep 24:13:1025. doi: 10.3389/fnins.2019.01025. eCollection 2019.

Authors

Niloofar Hashempour^{1

2}, Jetro J Tuulari^{1

2

3}, Harri Merisaari⁴, Kristian Lidauer¹, Iiris Luukkonen¹, Jani Saunavaara⁵, Riitta Parkkola⁶, Tuire Lähdesmäki⁷, Satu J Lehtola¹, Maria Keskinen¹, John D Lewis⁸, Noora M Scheinin^{1

2

9}, Linnea Karlsson^{1

10}, Hasse Karlsson^{1

2}

Affiliations

¹ FinnBrain Birth Cohort Study, Turku Brain and Mind Center, Institute of Clinical Medicine, University of Turku, Turku, Finland.
² Department of Psychiatry, Turku University Hospital, University of Turku, Turku, Finland.
³ Turku Collegium for Science and Medicine, University of Turku, Turku, Finland.
⁴ Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States.
⁵ Department of Medical Physics, Turku University Hospital, Turku, Finland.
⁶ Department of Radiology, Turku University Hospital, University of Turku, Turku, Finland.
⁷ Department of Pediatric Neurology, Turku University Hospital, University of Turku, Turku, Finland.
⁸ Montreal Neurological Institute, McGill University, Montreal, QC, Canada.
⁹ Turku PET Centre, University of Turku, Turku, Finland.
¹⁰ Department of Child Psychiatry, Turku University Hospital, University of Turku, Turku, Finland.

Abstract

The gross anatomy of the infant brain at term is fairly similar to that of the adult brain, but structures are immature, and the brain undergoes rapid growth during the first 2 years of life. Neonate magnetic resonance (MR) images have different contrasts compared to adult images, and automated segmentation of brain magnetic resonance imaging (MRI) can thus be considered challenging as less software options are available. Despite this, most anatomical regions are identifiable and thus amenable to manual segmentation. In the current study, we developed a protocol for segmenting the amygdala and hippocampus in T2-weighted neonatal MR images. The participants were 31 healthy infants between 2 and 5 weeks of age. Intra-rater reliability was measured in 12 randomly selected MR images, where 6 MR images were segmented at 1-month intervals between the delineations, and another 6 MR images at 6-month intervals. The protocol was also tested by two independent raters in 20 randomly selected T2-weighted images, and finally with T1 images. Intraclass correlation coefficient (ICC) and Dice similarity coefficient (DSC) for intra-rater, inter-rater, and T1 vs. T2 comparisons were computed. Moreover, manual segmentations were compared to automated segmentations performed by iBEAT toolbox in 10 T2-weighted MR images. The intra-rater reliability was high ICC ≥ 0.91, DSC ≥ 0.89, the inter-rater reliabilities were satisfactory ICC ≥ 0.90, DSC ≥ 0.75 for hippocampus and DSC ≥ 0.52 for amygdalae. Segmentations for T1 vs. T2-weighted images showed high consistency ICC ≥ 0.90, DSC ≥ 0.74. The manual and iBEAT segmentations showed no agreement, DSC ≥ 0.39. In conclusion, there is a clear need to improve and develop the procedures for automated segmentation of infant brain MR images.

Keywords: amygdala; automated segmentation; brain; hippocampus; infants; magnetic resonance imaging; manual segmentation.