Establishing a Baseline for Human Cortical Folding Morphological Variables: A Multisite Study

SAN: Mitigating spatial covariance heterogeneity in cortical thickness data collected from multiple scanners or sites

In neuroimaging studies, combining data collected from multiple study sites or scanners is becoming common to increase the reproducibility of scientific discoveries. At the same time, unwanted variations arise by using different scanners (inter-scanner biases), which need to be corrected before downstream analyses to facilitate replicable research and prevent spurious findings. While statistical harmonization methods such as ComBat have become popular in mitigating inter-scanner biases in neuroimaging, recent methodological advances have shown that harmonizing heterogeneous covariances results in higher data quality. In vertex-level cortical thickness data, heterogeneity in spatial autocorrelation is a critical factor that affects covariance heterogeneity. Our work proposes a new statistical harmonization method called spatial autocorrelation normalization (SAN) that preserves homogeneous covariance vertex-level cortical thickness data across different scanners. We use an explicit Gaussian process to characterize scanner-invariant and scanner-specific variations to reconstruct spatially homogeneous data across scanners. SAN is computationally feasible, and it easily allows the integration of existing harmonization methods. We demonstrate the utility of the proposed method using cortical thickness data from the Social Processes Initiative in the Neurobiology of the Schizophrenia(s) (SPINS) study. SAN is publicly available as an R package.

SAN: mitigating spatial covariance heterogeneity in cortical thickness data collected from multiple scanners or sites

In neuroimaging studies, combining data collected from multiple study sites or scanners is becoming common to increase the reproducibility of scientific discoveries. At the same time, unwanted variations arise by using different scanners (inter-scanner biases), which need to be corrected before downstream analyses to facilitate replicable research and prevent spurious findings. While statistical harmonization methods such as ComBat have become popular in mitigating inter-scanner biases in neuroimaging, recent methodological advances have shown that harmonizing heterogeneous covariances results in higher data quality. In vertex-level cortical thickness data, heterogeneity in spatial autocorrelation is a critical factor that affects covariance heterogeneity. Our work proposes a new statistical harmonization method called SAN (Spatial Autocorrelation Normalization) that preserves homogeneous covariance vertex-level cortical thickness data across different scanners. We use an explicit Gaussian process to characterize scanner-invariant and scanner-specific variations to reconstruct spatially homogeneous data across scanners. SAN is computationally feasible, and it easily allows the integration of existing harmonization methods. We demonstrate the utility of the proposed method using cortical thickness data from the Social Processes Initiative in the Neurobiology of the Schizophrenia(s) (SPINS) study. SAN is publicly available as an R package.

Prognostic associations of cortical gyrification in minimally medicated schizophrenia in an early intervention setting

The aberration in cortical gyrification seen in schizophrenia likely originates in the earliest phase of life, as gyrification begins in utero and reaches its peak in infancy. However, emerging observations have indicated a later reduction in gyrification, especially in early adulthood, may also occur in schizophrenia. At present, it is unclear whether the baseline and later gyrification reduction has any prognostic importance in schizophrenia. We address this question in a longitudinal design in patients minimally medicated at inception. About 108 minimally medicated (duration of medication = <14 days of antipsychotics) patients and 106 healthy controls underwent structural magnetic resonance imaging, with 34 patients being selectively re-scanned when clinically stable following antipsychotic treatment. The cortical surface from each structural image was reconstructed, and the local gyrification index and cortical thickness were computed for each vertex on the surface. We found minimally medicated schizophrenia patients during the first episode had a relatively higher gyrification in bilateral supramarginal, left superior temporal, and right posterior cingulate and paracentral regions. However, poor prognostic features were more likely in patients with lower baseline gyrification. Longitudinal reductions in left superior parietal and right precentral gyrification were associated with lower improvements in both positive and negative symptoms over time. The spatial pattern of longitudinal changes in gyrification was distinct from the changes in cortical thickness. These results indicated that schizophrenia is characterized by a relative hypergyrification in parieto-temporal and medial cortical areas at a group level at first presentation, but poor outcomes relate to lower-gyrification elsewhere both at the onset and during the early course. The early post-onset reduction of gyrification is rather limited in space and magnitude, but occurs unrelated to the progressive thinning, representing a distinct, prognostically important structural trajectory.