Graph Learning for Cortical Parcellation from Tensor Decompositions of Resting-State fMRI.

Cortical parcellation has long been a cornerstone in the field of neuroscience, enabling the cerebral cortex to be partitioned into distinct, non-overlapping regions that facilitate the interpretation and comparison of complex neuroscientific data. In recent years, these parcellations have frequently been based on the use of resting-state fMRI (rsfMRI) data. In parallel, methods such as independent components analysis have long been used to identify large-scale functional networks with significant spatial overlap between networks. Despite the fact that both forms of decomposition make use of the same spontaneous brain activity measured with rsfMRI, a gap persists in establishing a clear relationship between disjoint cortical parcellations and brain-wide networks. To address this, we introduce a novel parcellation framework that integrates NASCAR, a three-dimensional tensor decomposition method that identifies a series of functional brain networks, with state-of-the-art graph representation learning to produce cortical parcellations that represent near-homogeneous functional regions that are consistent with these brain networks. Further, through the use of the tensor decomposition, we avoid the limitations of traditional approaches that assume statistical independence or orthogonality in defining the underlying networks. Our findings demonstrate that these parcellations are comparable or superior to established atlases in terms of homogeneity of the functional connectivity across parcels, task contrast alignment, and architectonic map alignment. Our methodological pipeline is highly automated, allowing for rapid adaptation to new datasets and the generation of custom parcellations in just minutes, a significant advancement over methods that require extensive manual input. We describe this integrated approach, which we refer to as Untamed, as a tool for use in the fields of cognitive and clinical neuroscientific research. Parcellations created from the Human Connectome Project dataset using Untamed, along with the code to generate atlases with custom parcel numbers, are publicly available at https://untamed-atlas.github.io.

Artificial Intelligence Outcome Prediction in Neonates with Encephalopathy (AI-OPiNE).

Purpose To develop a deep learning algorithm to predict 2-year neurodevelopmental outcomes in neonates with hypoxic-ischemic encephalopathy using MRI and basic clinical data. Materials and Methods In this study, MRI data of term neonates with encephalopathy in the High-dose Erythropoietin for Asphyxia and Encephalopathy (HEAL) trial (ClinicalTrials.gov: NCT02811263), who were enrolled from 17 institutions between January 25, 2017, and October 9, 2019, were retrospectively analyzed. The harmonized MRI protocol included T1-weighted, T2-weighted, and diffusion tensor imaging. Deep learning classifiers were trained to predict the primary outcome of the HEAL trial (death or any neurodevelopmental impairment at 2 years) using multisequence MRI and basic clinical variables, including sex and gestational age at birth. Model performance was evaluated on test sets comprising 10% of cases from 15 institutions (in-distribution test set, n = 41) and 10% of cases from two institutions (out-of-distribution test set, n = 41). Model performance in predicting additional secondary outcomes, including death alone, was also assessed. Results For the 414 neonates (mean gestational age, 39 weeks ± 1.4 [SD]; 232 male, 182 female), in the study cohort, 198 (48%) died or had any neurodevelopmental impairment at 2 years. The deep learning model achieved an area under the receiver operating characteristic curve (AUC) of 0.74 (95% CI: 0.60, 0.86) and 63% accuracy in the in-distribution test set and an AUC of 0.77 (95% CI: 0.63, 0.90) and 78% accuracy in the out-of-distribution test set. Performance was similar or better for predicting secondary outcomes. Conclusion Deep learning analysis of neonatal brain MRI yielded high performance for predicting 2-year neurodevelopmental outcomes. Keywords: Convolutional Neural Network (CNN), Prognosis, Pediatrics, Brain, Brain Stem Clinical trial registration no. NCT02811263 Supplemental material is available for this article. © RSNA, 2024 See also commentary by Rafful and Reis Teixeira in this issue.

Integrated Short-TE and Hadamard-edited Multi-Sequence (ISTHMUS) for advanced MRS.

BACKGROUND: To examine data quality and reproducibility using ISTHMUS, which has been implemented as the standardized MR spectroscopy sequence for the multi-site Healthy Brain and Child Development (HBCD) study. METHODS: ISTHMUS is the consecutive acquisition of short-TE PRESS (32 transients) and long-TE HERCULES (224 transients) data with dual-TE water reference scans. Voxels were positioned in the centrum semiovale, dorsal anterior cingulate cortex, posterior cingulate cortex and bilateral thalamus regions. After acquisition, ISTHMUS data were separated into the PRESS and HERCULES portions for analysis and modeled separately using Osprey. In vivo experiments were performed in 10 healthy volunteers (6 female; 29.5±6.6 years). Each volunteer underwent two scans on the same day. Differences in metabolite measurements were examined. T2 correction based on the dual-TE water integrals were compared with: 1) T2 correction based on the default white matter and gray matter T2 reference values in Osprey and 2) shorter WM and GM T2 values from recent literature. RESULTS: No significant difference in linewidth was observed between PRESS and HERCULES. Bilateral thalamus spectra had produced significantly higher (p<0.001) linewidth compared to the other three regions. Linewidth measurements were similar between scans, with scan-to-scan differences under 1 Hz for most subjects. Paired t-tests indicated a significant difference only in PRESS NAAG between the two thalamus scans (p=0.002). T2 correction based on shorter T2 values showed better agreement to the dual-TE water integral ratio. CONCLUSIONS: ISTHMUS facilitated data acquisition and post-processing and reduced operator workload to eliminate potential human error.

Integrated Short-TE and Hadamard-edited Multi-Sequence (ISTHMUS) for Advanced MRS.

Background: To examine data quality and reproducibility using ISTHMUS, which has been implemented as the standardized MR spectroscopy sequence for the multi-site Healthy Brain and Child Development (HBCD) study. Methods: ISTHMUS is the consecutive acquisition of short-TE PRESS (32 transients) and long-TE HERCULES (224 transients) data with dual-TE water reference scans. Voxels were positioned in the centrum semiovale, dorsal anterior cingulate cortex, posterior cingulate cortex and bilateral thalamus regions. After acquisition, ISTHMUS data were separated into the PRESS and HERCULES portions for analysis and modeled separately using Osprey. In vivo experiments were performed in 10 healthy volunteers (6 female; 29.5±6.6 years). Each volunteer underwent two scans on the same day. Differences in metabolite measurements were examined. T2 correction based on the dual-TE water integrals were compared with: 1) T2 correction based the default white matter and gray matter T2 reference values in Osprey; 2) shorter WM and GM T2 values from recent literature; and 3) reduced CSF fractions. Results: No significant difference in linewidth was observed between PRESS and HERCULES. Bilateral thalamus spectra had produced significantly higher (p<0.001) linewidth compared to the other three regions. Linewidth measurements were similar between scans, with scan-to-scan differences under 1 Hz for most subjects. Paired t-tests indicated a significant difference only in PRESS NAAG between the two thalamus scans (p=0.002). T2 correction based on shorter T2 values showed better agreement to the dual-TE water integral ratio. Conclusions: ISTHMUS facilitated and standardized acquisition and post-processing and reduced operator workload to eliminate potential human error.

Genetic and Congenital Anomalies in Infants With Hypoxic-Ischemic Encephalopathy.

BACKGROUND: Infants with hypoxic ischemic encephalopathy (HIE) may have underlying conditions predisposing them to hypoxic-ischemic injury during labor and delivery. It is unclear how genetic and congenital anomalies impact outcomes of HIE. METHODS: Infants with HIE enrolled in a phase III trial underwent genetic testing when clinically indicated. Infants with known genetic or congenital anomalies were excluded. The primary outcome, i.e., death or neurodevelopmental impairment (NDI), was determined at age two years by a standardized neurological examination, Bayley Scales of Infant Development, Third Edition (BSID-III), and the Gross Motor Function Classification Scales. Secondary outcomes included cerebral palsy and BSID-III motor, cognitive, and language scores at age two years. RESULTS: Of 500 infants with HIE, 24 (5%, 95% confidence interval 3% to 7%) were diagnosed with a genetic (n = 15) or congenital (n = 14) anomaly. Infants with and without genetic or congenital anomalies had similar rates of severe encephalopathy and findings on brain magnetic resonance imaging. However, infants with genetic or congenital anomalies were more likely to have death or NDI (75% vs 50%, P = 0.02). Among survivors, those with a genetic or congenital anomaly were more likely to be diagnosed with cerebral palsy (32% vs 13%, P = 0.02), and had lower BSID-III scores in all three domains than HIE survivors without such anomalies. CONCLUSIONS: Among infants with HIE, 5% were diagnosed with a genetic or congenital anomaly. Despite similar clinical markers of HIE severity, infants with HIE and a genetic or congenital anomaly had worse neurodevelopmental outcomes than infants with HIE alone.