Synthetic magnetic resonance-based relaxometry and brain volume: cutoff values for predicting neurocognitive outcomes in very preterm infants.

BACKGROUND: Early neurorehabilitation can enhance neurocognitive outcomes in very preterm infants (<32 weeks), and conventional magnetic resonance imaging (MRI) is commonly used to assess neonatal brain injury; however, the predictive value for neurodevelopmental delay is limited. Timely predictive quantitative biomarkers are needed to improve early identification and management of infants at risk of neurodevelopmental delay. OBJECTIVE: To evaluate the potential of quantitative synthetic MRI measurements at term-equivalent age as predictive biomarkers of neurodevelopmental impairment and establish practical cutoff values to guide clinical decision-making. MATERIALS AND METHODS: This retrospective study included 93 very preterm infants who underwent synthetic MRI at term-equivalent age between January 2017 and September 2020. Clinical outcomes were assessed using the Bayley-III scale of infant development (mean age 2.1 years). The predictive value for impaired development was analyzed using receiver operating characteristic curves for synthetic MRI-based volumetry and T1 and T2 relaxation measurements. RESULTS: The T1 relaxation time in the posterior limb of the internal capsule was a potent predictor of severe (sensitivity, 92%; specificity, 80%; area under the curve (AUC), 0.91) and mild or severe (AUC, 0.75) developmental impairment. T2 relaxation time in the posterior limb of the internal capsule was a significant predictor of severe impairment (AUC, 0.76), whereas the brain parenchymal volume was a significant predictor of severe (AUC, 0.72) and mild or severe impairment (AUC, 0.71) outperforming the reported qualitative MRI scores (AUC, 0.66). CONCLUSION: The proposed cutoff values for T1 relaxation time in the posterior limb of the internal capsule and for total brain volume measurements, derived from synthetic MRI, show promise as predictors of both mild and severe neurodevelopmental impairment in very preterm infants.

Automated analysis of trapeziometacarpal joint kinematics using four-dimensional computed tomography.

The aim of this study was to develop an automated approach model to define in vivo kinematics of the trapeziometacarpal (TMC) joint using four-dimensional computed tomography. A total of 15 healthy volunteers were included and their TMC joint kinematics were studied during a retropulsion-opposition-retropulsion movement. We used cardan angles estimated from transformation matrices using a ZYX-decomposition and analysed the motion of the thumb metacarpal relative to the trapezium, the thumb metacarpal relative to the index metacarpal, and the trapezium relative to the index metacarpal. The study also included an analysis of the joint hysteresis effect and a joint proximity model that estimated the joint contact area during a retropulsion-opposition-retropulsion movement. The automated approach significantly decreased the time needed to analyse each case and makes this model applicable for further research on TMC kinematics.