Objectives This study aimed at developing a convolutional neural network (CNN) able to automatically quantify and characterize the level of degeneration of rotator cuff (RC) muscles from shoulder CT images including muscle atrophy and fatty infiltration. Methods One hundred three shoulder CT scans from 95 patients with primary glenohumeral osteoarthritis undergoing anatomical total shoulder arthroplasty were retrospectively retrieved. Three independent radiologists manually segmented the premorbid boundaries of all four RC muscles on standardized sagittal-oblique CT sections. This premorbid muscle segmentation was further automatically predicted using a CNN. Automatically predicted premorbid segmentations were then used to quantify the ratio of muscle atrophy, fatty infiltration, secondary bone formation, and overall muscle degeneration. These muscle parameters were compared with measures obtained manually by human raters. Results Average Dice similarity coefficients for muscle segmentations obtained automatically with the CNN (88% ± 9%) and manually by human raters (89% ± 6%) were comparable. No significant differences were observed for the subscapularis, supraspinatus, and teres minor muscles (p > 0.120), whereas Dice coefficients of the automatic segmentation were significantly higher for the infraspinatus (p < 0.012). The automatic approach was able to provide good-very good estimates of muscle atrophy (R 2 = 0.87), fatty infiltration (R 2 = 0.91), and overall muscle degeneration (R 2 = 0.91). However, CNN-derived segmentations showed a higher variability in quantifying secondary bone formation (R 2 = 0.61) than human raters (R 2 = 0.87). Conclusions Deep learning provides a rapid and reliable automatic quantification of RC muscle atrophy, fatty infiltration, and overall muscle degeneration directly from preoperative shoulder CT scans of osteoarthritic patients, with an accuracy comparable with that of human raters. Key Points • Deep learning can not only segment RC muscles currently available in CT images but also learn their pre-existing locations and shapes from invariant anatomical structures visible on CT sections. • Our automatic method is able to provide a rapid and reliable quantification of RC muscle atrophy and fatty infiltration from conventional shoulder CT scans. • The accuracy of our automatic quantitative technique is comparable with that of human raters.

ObjectivesWe studied in a clinical setting the age dependent T1 relaxation time as a marker of normal late brain maturation and compared it to conventional techniques, namely the apparent diffusion coefficient (ADC).Materials and methodsForty-two healthy subjects ranging from ages 1 year to 20 years were included in our study. T1 brain maps in which the intensity of each pixel corresponded to T1 relaxation times were generated based on MR imaging data acquired using a MP2RAGE sequence. During the same session, diffusion tensor imaging data was collected. T1 relaxation times and ADC in white matter and grey matter were measured in seven clinically relevant regions of interest and were correlated to subjects’ age.ResultsIn the basal ganglia, there was a small, yet significant, decrease in T1 relaxation time (-0.45 ≤R≤-0.59, p<10−2) and ADC (-0.60≤R≤-0.65, p<10−4) as a function of age. In the frontal and parietal white matter, there was a significant decrease in T1 relaxation time (-0.62≤R≤-0.68, p<10−4) and ADC (-0.81≤R≤-0.85, p<10−4) as a function of age. T1 relaxation time changes in the corpus callosum and internal capsule were less relevant for this age range. There was no significant difference between the correlation of T1 relaxation time and ADC with respect to age (p-value = 0.39). The correlation between T1 relaxation and ADC is strong in the white matter but only moderate in basal ganglia over this age period.ConclusionsT1 relaxation time is a marker of brain maturation or myelination during late brain development. Between the age of 1 and 20 years, T1 relaxation time decreases as a function of age in the white matter and basal ganglia. The greatest changes occur in frontal and parietal white matter. These regions are known to mature in the final stage of development and are mainly composed of association circuits. Age-correlation is not significantly different between T1 relaxation time and ADC. Therefore, T1 relaxation time does not appear to be a superior marker of brain maturation than ADC but may be considered as complementary owing the intrinsic differences in bio-physical sensitivity. This work may serve as normative ranges in clinical imaging routines.

No abstract

No abstract

We present a case of popliteal venous aneurysm causing recurrent pulmonary embolism successfully treated by surgical resection.