Fidelity of 3D Printed Brains from MRI Scan in Children with Pathology (Prior Hypoxic Ischemic Injury)

Abstract

Cortical injury on the surface of the brain in children with hypoxic ischemic injury (HII) can be difficult to demonstrate to non-radiologists and lay people using brain images alone. Three-dimensional (3D) printing is helpful to communicate the volume loss and pathology due to HII in children's brains. 3D printed models represent the brain to scale and can be held up against models of normal brains for appreciation of volume loss. If 3D printed brains are to be used for formal communication, e.g., with medical colleagues or in court, they should have high fidelity of reproduction of the actual size of patients' brains. Here, we evaluate the size fidelity of 3D printed models from MRI scans of the brain, in children with prior HII. Twelve 3D prints of the brain were created from MRI scans of children with HII and selected to represent a variety of cortical pathologies. Specific predetermined measures of the 3D prints were made and compared to measures in matched planes on MRI. Fronto-occipital length (FOL) and bi-temporal/bi-parietal diameters (BTD/BPD) demonstrated high interclass correlations (ICC). Correlations were moderate to weak for hemispheric height, temporal height, and pons-cerebellar thickness. The average standard error of measurement (SEM) was 0.48 cm. Our results demonstrate high correlations in overall measurements of each 3D printed model derived from brain MRI scans versus the original MRI, evidenced by high ICC values for FOL and BTD/BPD. Measures with low correlation values can be explained by variability in matching the plane of measurement to the MRI slice orientation.

Keywords: 3D printing; Cerebral palsy; Fidelity; Hypoxic ischemic injury; Magnetic resonance imaging; Pediatric brain.

Fig. 1

Comparison of axial T1W MRI image (a) with vertex view (b) and side oblique view (c) of the corresponding 3D printed model. The areas depicted by the red arrows correspond to subtle localized atrophy involving the peri-rolandic region of both cerebral hemispheres, sustained from a remote prior profound hypoxic ischemic injury in an 11-year 5-month-old girl. The black arrows depict the inter-hemispheric/lentiform separation due to the atrophy. The MRI image (a) depicts one aspect of the atrophy while the 3D printed model image (c) demonstrates the extension of involvement to the inferior and lateral aspects of the left pre-central gyrus to good effect

Fig. 2

T1W MRI images in sagittal (a), coronal (b), and axial (c) planes with images of the corresponding 3D printed model in different views (d–f). This is a 10-year 6-month-old girl who suffered a remote prior hypoxic ischemic injury at term, with a combined partial prolonged and profound injury. The corpus callosum is diffusely thinned as a result of global atrophy (white arrows in (a)). The green arrows depict areas of focal cystic encephalomalacia and ulegyria within the anterior inter-arterial watershed of both cerebral hemispheres and the corresponding overlying cortical atrophy. The blue chevron depicts an area that required a manual patch (after automatic segmentation) to cover a defect due to the segmentation not recognizing the residual thin cortex overlying the cystic encephalomalacia. The red arrows depict atrophy with deep sulci of the posterior inter-arterial watershed bilateral. The asterisks (white and black) depict inter-hemispheric fissure widening due to the diffuse atrophy and associated lentiform separation of the hemispheres at the involved regions

Fig. 3

Linear measures on T1W MRI images. Fronto-occipital measures in the axial plane (a). Bi-temporal or bi-parietal diameter in the axial plane (b). Hemispheric height in the coronal plane (c)

Fig. 4

Linear measures on T1W MR images. Fronto-occipital measure on the midline sagittal image (a). Pons cerebellar thickness on midline sagittal image (b). Temporal height in the coronal plane (c)

Fig. 5

Demonstrative images of linear measures of the 3D print models using the digital Vernier calipers. Right hemispheric fronto-occipital length (a). Bi-parietal or bi-temporal diameter (b). Left hemispheric height (c)

Fig. 6

Graphic display of correlations between the MRI linear measures (MRI) and measurements obtained from the 3D print models for right fronto-occipital length (a), left fronto-occipital length (b), bi-temporal or bi-parietal diameter (c), and pons cerebellar thickness (d). (Key: r, correlation coefficient; p, statistical significance; Spearman r, Spearman’s correlation coefficient; ICC, interclass correlation coefficient; SEM, standard error of measurement). Units of SEM in centimeters (cm)

Fig. 7

Graphic display of correlations between the MRI linear measures (MRI) and measurements obtained from the 3D print models (print) for right hemispheric height (a), left hemispheric height (b), right temporal height (c), and left temporal height (d). (Key: r, correlation coefficient; p, statistical significance; Spearman r, Spearman’s correlation coefficient; ICC (agreement), interclass correlation coefficient agreement; ICC (consistency), interclass correlation coefficient consistency; SEM, standard error of measurement). Units of SEM in centimeters (cm)

Fig. 8

Two models of different children’s brains side by side for comparison. Both patients suffered partial prolonged hypoxic ischemic injury at term gestation, now with cerebral palsy, who had delayed MR imaging — a 2-year 2-month-old boy (a) and a 11-year 6-month-old boy (b). Note, the more severe atrophy in the 11-year-old (b) evident from increased depth and width of the sulci in the watershed region (red arrows in each model) as well as the low overall brain volume (i.e., an 11-year-old with a similar or smaller brain volume than a 2-year-old) when visualizing the models side by side