Automatic multispectral MRI segmentation of human hippocampal subfields: an evaluation of multicentric test–retest reproducibility

Abstract

Accurate and reproducible automated segmentation of human hippocampal subfields is of interest to study their roles in cognitive functions and disease processes. Multispectral structural MRI methods have been proposed to improve automated hippocampal subfield segmentation accuracy, but the reproducibility in a multicentric setting is, to date, not well characterized. Here, we assessed test–retest reproducibility of FreeSurfer 6.0 hippocampal subfield segmentations using multispectral MRI analysis pipelines (22 healthy subjects scanned twice, a week apart, at four 3T MRI sites). The harmonized MRI protocol included two 3D-T1, a 3D-FLAIR, and a high-resolution 2D-T2. After within-session T1 averaging, subfield volumes were segmented using three pipelines with different multispectral data: two longitudinal (“long_T1s” and “long_T1s_FLAIR”) and one cross-sectional (“long_T1s_FLAIR_crossT2”). Volume reproducibility was quantified in magnitude (reproducibility error—RE) and space (DICE coefficient). RE was lower in all hippocampal subfields, except for hippocampal fissure, using the longitudinal pipelines compared to long_T1s_FLAIR_crossT2 (average RE reduction of 0.4–3.6%). Similarly, the longitudinal pipelines showed a higher spatial reproducibility (1.1–7.8% of DICE improvement) in all hippocampal structures compared to long_T1s_FLAIR_crossT2. Moreover, long_T1s_FLAIR provided a small but significant RE improvement in comparison to long_T1s (p = 0.015), whereas no significant DICE differences were found. In addition, structures with volumes larger than 200 mm3 had better RE (1–2%) and DICE (0.7–0.95) than smaller structures. In summary, our study suggests that the most reproducible hippocampal subfield FreeSurfer segmentations are derived from a longitudinal pipeline using 3D-T1s and 3D-FLAIR. Adapting a longitudinal pipeline to include high-resolution 2D-T2 may lead to further improvements.

Subcortical Volume Changes in Early Menopausal Women and Correlation With Neuropsychological Tests

“…For analysis, the T1-weighted images with indistinct gray-white demarcation and obvious motion artifacts were excluded first. The cerebrum cortical reconstruction and volumetric segmentation were conducted by the FreeSurfer image analysis suite (version 6.0), which has good test-retest reliability on scanners of different manufacturers ( Chiappiniello et al, 2021 ). Field strengths were freely obtained online, 1 and technical details of these procedures are described in prior publications.…”

“…The FreeSurfer automated segmentation procedure is a novel and reliable method to measure hippocampal subregional volumes ( Schmidt et al, 2018 ; Chiappiniello et al, 2021 ; Figueiredo et al, 2021 ). It has been authenticated in many studies, including those investigating Alzheimer’s disease ( Novak and Einstein, 2013 ), schizophrenia ( Tost et al, 2013 ), and fetal alcohol spectrum disorder ( Guio et al, 2016 ).…”

The Learning Curve and Inter-Observer Variability in Contouring the Hippocampus under the Hippocampal Sparing Guidelines of Radiation Therapy Oncology Group 0933

“…In a similar study, analyzing difference between manual delineation of hippocampi on T1-weight MRI scans, the Dice similarity coefficient is in range between 0.18, 0.24 and 0.31 for the three cohorts [30]. In other studies using automatic contouring of hippocampi, the Dice coefficient index was higher compared to manually delineation hippocampi and was in range 0.65 to 0.95 [20,31]. The manually delineated hippocampi based on a deep learning course showed satisfying results, which could have a positive impact on improving delineation structures under RT planning.…”

“…The hippocampal bodies were contoured as two separate structures without conjunction with the amygdala in all cases. Variations in the spatial localization of the hippocampi delineated by different doctors were described in three directions: right-left (X), cranio-caudal (Y), and forward-backward (Z), in relation to the reference contour of the hippocampus delineated by the most experienced radiation oncologist with expertise in neuroimaging, according to RTOG atlas recommendations [20]. It needs to be emphasized that previously delineated hippocampi were invisible to subsequently participants.…”

Comparison and Reliability of Hippocampal Subfield Segmentations Within FreeSurfer Utilizing T1- and T2-Weighted Multispectral MRI Data

“…This is a drawback for FreeSurfer that longitudinal hippocampal processing is only possible using a T1-weighted image and not available for multispectral contrast inputs. This should be addressed in future releases of this software package as it was recently shown that the longitudinal approach outperformed cross-sectional hippocampal processing ( Chiappiniello et al, 2020 ). In this investigation, authors also used a multispectral approach, however, focusing on the recon-all stream and not directly on the hippocampal subfield tool, as we did in our analysis.…”