Vascular retinal biomarkers improves the detection of the likely cerebral amyloid status from hyperspectral retinal images.

INTRODUCTION: This study investigates the relationship between retinal image features and ß-amyloid (Aß) burden in the brain with the aim of developing a noninvasive method to predict the deposition of Aß in the brain of patients with Alzheimer's disease. METHODS: Retinal images from 20 cognitively impaired and 26 cognitively unimpaired cases were acquired (3 images per subject) using a hyperspectral retinal camera. The cerebral amyloid status was determined from binary reads by a panel of 3 expert raters on 18F-florbetaben positron-emission tomography (PET) studies. Image features from the hyperspectral retinal images were calculated, including vessels tortuosity and diameter and spatial-spectral texture measures in different retinal anatomical regions. RESULTS: Retinal venules of amyloid-positive subjects (Aß+) showed a higher mean tortuosity compared with the amyloid-negative (Aß-) subjects. Arteriolar diameter of Aß+ subjects was found to be higher than the Aß- subjects in a zone adjacent to the optical nerve head. Furthermore, a significant difference between texture measures built over retinal arterioles and their adjacent regions were observed in Aß+ subjects when compared with the Aß-. A classifier was trained to automatically discriminate subjects combining the extracted features. The classifier could discern Aß+ subjects from Aß- subjects with an accuracy of 85%. DISCUSSION: Significant differences in texture measures were observed in the spectral range 450 to 550 nm which is known as the spectral region known to be affected by scattering from amyloid aggregates in the retina. This study suggests that the inclusion of metrics related to the retinal vasculature and tissue-related textures extracted from vessels and surrounding regions could improve the discrimination performance of the cerebral amyloid status.

To distinguish flexible and rigid lumbar curve from MRI texture analysis in adolescent idiopathic scoliosis: A feasibility study.

BACKGROUND: Imaging in side bending, supine, traction, fulcrum, and push prone are examples of methods used to evaluate the curve reduction of scoliotic spine. However, being able to determine spine curve flexibility from MRI would eliminate the need of additional X-ray radiation related to radiograph acquisition in side-bending. PURPOSE/HYPOTHESIS: To find specific texture features of lumbar postural muscles on MRI that can distinguish flexible from rigid lumbar scoliotic curves. We hypothesized that the changes occurring in postural muscles with scoliosis can be seen with MRI. STUDY TYPE: Retrospective study case control. POPULATION: With Institutional Review Board approval and informed consent, 15 adolescents with idiopathic scoliosis and scheduled for surgery were involved. FIELD STRENGTH/SEQUENCE: T1 -weighted MR images were performed on a 1.5T system using a spin echo sequence in the axial direction. ASSESSMENT: The spinal erector, quadratus lumborum and psoas major muscles were analyzed using textural features. STATISTICAL TESTS: Principal component analysis (PCA) and agglomerative hierarchical clustering (AHC) were used to classify the lumbar postural muscles and calculate performance metrics. The lumbar flexibility index, measured from suspension tests, was used as ground truth measurement. RESULTS: The five discriminant features (out of 34 tested features) obtained from PCA were able to keep over 90% of the variability of the dataset. The right and left spinal erector and the left psoas major had the highest performance metrics to classify the spinal curve flexibility, with an accuracy over 0.80, a sensitivity over 0.82, a specificity over 0.68, and a Matthews correlation coefficient over 0.57. DATA CONCLUSION: This study analyzed MRI using texture information of muscle to distinguish flexible from rigid scoliotic curves. Some postural muscle such as the spinal erector and the psoas major are more likely to reflect the curve flexibility of a scoliotic participant. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2017.

Texture analysis for automatic segmentation of intervertebral disks of scoliotic spines from MR images.

This paper presents a unified framework for automatic segmentation of intervertebral disks of scoliotic spines from different types of magnetic resonance (MR) image sequences. The method exploits a combination of statistical and spectral texture features to discriminate closed regions representing intervertebral disks from background in MR images of the spine. Specific texture features are evaluated for three types of MR sequences acquired in the sagittal plane: 2-D spin echo, 3-D multiecho data image combination, and 3-D fast imaging with steady state precession. A total of 22 texture features (18 statistical and 4 spectral) are extracted from every closed region obtained from an automatic segmentation procedure based on the watershed approach. The feature selection step based on principal component analysis and clustering process permit to decide among all the extracted features which ones resulted in the highest rate of good classification. The proposed method is validated using a supervised k-nearest-neighbor classifier on 505 MR images coming from three different scoliotic patients and three different MR acquisition protocols. Results suggest that the selected texture features and classification can contribute to solve the problem of oversegmentation inherent to existing automatic segmentation methods by successfully discriminating intervertebral disks from the background on MRI of scoliotic spines.