Human long bone development in vivo: analysis of the distal femoral epimetaphysis on MR images of fetuses

Bone age assessment from articular surface and epiphysis using deep neural networks

Bone age assessment is of great significance to genetic diagnosis and endocrine diseases. Traditional bone age diagnosis mainly relies on experienced radiologists to examine the regions of interest in hand radiography, but it is time-consuming and may even lead to a vast error between the diagnosis result and the reference. The existing computer-aided methods predict bone age based on general regions of interest but do not explore specific regions of interest in hand radiography. This paper aims to solve such problems by performing bone age prediction on the articular surface and epiphysis from hand radiography using deep convolutional neural networks. The articular surface and epiphysis datasets are established from the Radiological Society of North America (RSNA) pediatric bone age challenge, where the specific feature regions of the articular surface and epiphysis are manually segmented from hand radiography. Five convolutional neural networks, i.e., ResNet50, SENet, DenseNet-121, EfficientNet-b4, and CSPNet, are employed to improve the accuracy and efficiency of bone age diagnosis in clinical applications. Experiments show that the best-performing model can yield a mean absolute error (MAE) of 7.34 months on the proposed articular surface and epiphysis datasets, which is more accurate and fast than the radiologists. The project is available at https://github.com/YameiDeng/BAANet/, and the annotated dataset is also published at https://doi.org/10.5281/zenodo.7947923.

Femur development in fetal growth restriction as observed on prenatal magnetic resonance imaging

OBJECTIVE

To investigate human femur development in fetal growth restriction (FGR) by analyzing femur morphometrics and distal epimetaphyseal features on prenatal magnetic resonance imaging (MRI).

METHODS

This was a retrospective study of 111 fetuses (mean gestational age (GA), 27 + 2 weeks (range, 19-35 weeks)) with FGR associated with placental insufficiency without other major abnormalities and 111 GA-matched normal controls. On 1.5-Tesla echoplanar MRI, femur morphometrics, including diaphyseal length, epiphyseal length and epiphyseal width, were assessed. Using a previously reported grading system, epimetaphyseal features, including cartilaginous epiphyseal shape, metaphyseal shape and epiphyseal ossification, were analyzed qualitatively. To compare FGR cases and controls, the paired t-test was used to assess morphometrics, generalized estimating equations were used for epimetaphyseal features and time-to-event analysis was used to assess the visibility of epiphyseal ossification.

RESULTS

There were significant differences in femur morphometrics between FGR cases and controls (all parameters, P < 0.001), with bone shortening observed in FGR. No significant differences were found in the distribution of epimetaphyseal features between FGR cases and controls (epiphyseal shape, P = 0.341; metaphyseal shape, P = 0.782; epiphyseal ossification, P = 0.85). Epiphyseal ossification was visible at a median of 33.6 weeks in FGR cases and at 32.1 weeks in controls (P = 0.008).

CONCLUSIONS

On prenatal MRI, cases with FGR associated with placental insufficiency exhibit diaphyseal and epiphyseal shortening of the femur. However, FGR cases and normal controls share similarly graded distal epimetaphyseal features. Consequently, these features may not be appropriate MRI characteristics for the identification of FGR. © 2022 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Susceptibility-weighted imaging to evaluate normal and abnormal vertebrae in fetuses:a preliminary study

OBJECTIVE

To evaluate the performance of Susceptibility-weighted imaging (SWI) in visualizing normal and abnormal fetal vertebrae in vivo and in utero.

METHODS

Ninety-seven women with normal fetal vertebrae and 127 women suspected fetal vertebral anomalies on ultrasound were included in our study. SWI, TrueFISP and HASTE of the fetal spine were performed on 1.5-T MRI. The image quality and diagnostic performance between HASTE/TrueFISP and SWI were compared. Pearson correlations to correlate the L1 centrum ossification center (COC) measurements with gestational age (GA) were performed.

RESULTS

The visibility of the fetal vertebral structures on the SWI images (3.58 ± 0.69) was significantly greater than those on the HASTE (1.98 ± 0.51, P < 0.001) and TrueFISP (2.63 ± 0.52, P < 0.001). The diagnostic accuracy of SWI (89.0%) was superior to HASTE/TrueFISP (48.0%) (P < 0.001) and the area under the curve (AUC) for SWI was 0.909 (P < 0.001). The height, transverse, sagittal diameter and area of L1 COC were linearly correlated with GA (all P < 0.001).

CONCLUSION

SWI proved to be a reliable method for depicting fetal vertebral structure and growth, which can significantly improve the diagnostic performance of vertebral anomalies in fetuses. This article is protected by copyright. All rights reserved.

Magnetic resonance imaging of the fetal musculoskeletal system

Diagnosing musculoskeletal pathology requires understanding of the normal embryological development. Intrinsic errors of skeletal development are individually rare but are of paramount clinical importance because anomalies can greatly impact patients' lives. An accurate assessment of the fetal musculoskeletal system must be performed to provide optimal genetic counseling as well as to drive therapeutic management. This manuscript reviews the embryology of skeletal development and the appearance of the maturing musculoskeletal system on fetal MRI. In addition, it presents a comprehensive review of musculoskeletal fetal pathology along with postnatal imaging.

It's time to recognize the perichondrium

The perichondrium is a complex structure centered at the chondro-osseous junction of growing bones. It plays an important role in both normal skeletal development and in pathological conditions. This review illustrates the normal anatomy, function and imaging appearance of the perichondrium from fetal development to older childhood. The radiologic appearance of the perichondrium in skeletal trauma, infection and tumors in which it plays a role also are reviewed.

Fetal MRI, lower acceptance by women in research vs. clinical setting

AIM

To determine acceptance of pregnant women to undergo fetal magnetic resonance imaging (MRI) examination in research and clinical setting.

METHODS

A prospective study included a research group [part of a study comparing brain ultrasound (US) to MRI in fetuses at risk for acquired brain damage] and a clinical group [fetuses with suspected (brain) anomalies after structural US examination] from 2011 to 2014. All women were advised to use sedatives. MRI declinations, use of sedation, MRI duration and imaging quality were compared between both groups.

RESULTS

Study participation was accepted in 57/104 (55%) research cases. Fetal MRI was performed in 34/104 (33%) research and 43/44 (98%) clinical cases. Reasons to decline study participation were MRI related in 41%, and participation was too burdensome in 46%. Acceptance was highest for indication infection and lowest in alloimmune thrombocytopenia and monochorionic twin pregnancy. Sedatives were used in 14/34 research and 43/43 clinical cases. Scan duration and quality were comparable (21 and 20 min in research and clinical cases, respectively, moderate/good quality in both groups).

CONCLUSIONS

Pregnant women consider MRI more burdensome than professionals realize. Two-third of women at risk for fetal brain damage decline MRI examination. Future studies should evaluate which information about fetal MRI is supportive.

Three-dimensional growth of tibial shaft ossification in the human fetus: a digital-image and statistical analysis

PURPOSES

Tibial shaft ossification in terms of its size and growth may be criticalin describing both the fetal stage and maturity, and in identifying innate disorders. The present study was executed to quantitatively assess ossification of the tibial shaft, taking its morphometric linear, planar and volumetric parameters into account.

MATERIALS AND METHODS

With the use of methods of CT, digital-image analysis and statistics, the evolutionof tibial shaft ossification in 47 spontaneously aborted human fetuses at the age of 17-30 weeks was studied.

RESULTS

Without any male-female and right-left morphometric differences, the best fit growth dynamics fortibial shaft ossification was modelled by the following functions: y = 5.312 + 0.034 × (age)² ± 0.001 (R² = 0.89) for its length, y = - 2.855 + 0.307 × age ± 0.009 (R² = 0.96) for its proximal transverse diameter, y = - 0.758 + 0.153 × age ± 0.005 (R² = 0.88) for its middle transverse diameter, y = - 1.844 + 0.272 × age ± 0.09 (R² = 0.90) for its distal transverse diameter, y = - 40.263 + 0.258 × (age)² ± 0.007 (R² = 0.94) for its projection surface area, and y = - 287.996 + 1.186 × (age)² ± 0.037 (R² = 0.92) for its volume. The femoral-to-tibial ossification length ratio was 1.15 ± 0.1.

CONCLUSIONS

The size of tibial shaft ossification displays neither sex nor laterality differences. Tibial shaft ossification follows quadratic functions with respect to its length, projection surface area and volume, and linear functions with respect to its proximal, middle and distal transverse diameters. The obtained morphometric data of tibial shaft ossification are considered normative age-specific references of relevance in both the estimation of fetal ages and the ultrasound diagnostics of congenital defects.

Quantitative anatomy of the primary ossification center of the femoral shaft in human fetuses

Purpose

Early clinical distinction of congenital defects in the femur is extremely important, as it determines the prognosis of the development of the lower limb. This study was performed to quantitatively examine the primary center of ossification in the femoral shaft with respect to its linear, planar, and volumetric parameters.

Materials and methods

Using methods of CT, digital-image analysis, and statistics, the size of the primary ossification center of the femoral shaft in 47 spontaneously aborted human fetuses aged 17–30 weeks was studied.

Results

With no sex and laterality differences, the best fit growth dynamics for femoral shaft ossification center was modelled by the following functions: y = 5.717 + 0.040 × (age)² ± 2.905 (R² = 0.86) for its length, y = −3.579 + 0.368 × age ± 0.529 (R² = 0.88) for its proximal transverse diameter, y = −1.105 + 0.187 × age ± 0.309 (R² = 0.84) for its middle transverse diameter, y = −2.321 + 0.323 × age ± 0.558 (R² = 0.83) for its distal transverse diameter, y = −50.306 + 0.308 × (age)² ± 18.289 (R² = 0.90) for its projection surface area, and y = −91.458 + 0.390 × (age)³ ± 92.146 (R² = 0.88) for its volume.

Conclusions

The size of the femoral shaft ossification center displays neither sex nor laterality differences. The ossification center in the femoral shaft follows quadratic functions with respect to its length and projection surface area, linear functions with respect to its proximal, middle, and distal transverse diameters, and a cubic function with respect to its volume. The obtained morphometric data of the femoral shaft ossification center are considered normative for respective prenatal weeks and may be of relevance in both the estimation of fetal ages and the ultrasound diagnostics of congenital defects.

Fetal MRI at 3T-ready for routine use?

Fetal MR now plays an important role in the clinical work-up of pregnant females. It is performed mainly at 1.5 T. However, the desire to obtain a more precise fetal depiction or the fact that some institutions have access only to a 3.0 T scanner has resulted in a growing interest in performing fetal MR at 3.0 T. The aim of this article was to provide a reference for the use of 3.0 T MRI as a prenatal diagnostic method.

Prenatal diagnosis of congenital upper limb differences: a current concept review

Congenital upper limb differences are frequently associated with complex syndromes. Ultrasonography is considered as the first-line diagnostic modality, and fetal MRI can be useful to further evaluate ill-defined areas. Genetic and non-invasive prenatal testing help to identify the underlying genetic disorder. The diagnostic assessment is a multidisciplinary task that should involve early prenatal consultations with specialists involved in case management and treatment planning. Obstetricians, geneticists, radiologists, psychologists and dedicated surgeons are needed to provide good parental education, prenatal and postnatal care, and successful outcomes. The purpose of this review is to provide an overview of the clinicopathologic background, current diagnostic and imaging procedures in affected fetuses.

Growth plate stress distribution implications during bone development: a simple framework computational approach

Mechanical stimuli play a significant role in the process of long bone development as evidenced by clinical observations and in vivo studies. Up to now approaches to understand stimuli characteristics have been limited to the first stages of epiphyseal development. Furthermore, growth plate mechanical behavior has not been widely studied. In order to better understand mechanical influences on bone growth, we used Carter and Wong biomechanical approximation to analyze growth plate mechanical behavior, and explore stress patterns for different morphological stages of the growth plate. To the best of our knowledge this work is the first attempt to study stress distribution on growth plate during different possible stages of bone development, from gestation to adolescence. Stress distribution analysis on the epiphysis and growth plate was performed using axisymmetric (3D) finite element analysis in a simplified generic epiphyseal geometry using a linear elastic model as the first approximation. We took into account different growth plate locations, morphologies and widths, as well as different epiphyseal developmental stages. We found stress distribution during bone development established osteogenic index patterns that seem to influence locally epiphyseal structures growth and coincide with growth plate histological arrangement.

Foetal "black bone" MRI: utility in assessment of the foetal spine

OBJECTIVE

Foetal CT has recently been added to the foetal imaging armamentarium, but this carries with it the risks of ionizing radiation, both to the mother and the foetus. Foetal "black bone" MRI is a new technique that allows assessment of the foetal skeleton without the risk of exposure to ionizing radiation and is a potential new sequence in foetal MRI examination.

METHODS

Retrospective review of all foetal MRI studies over the past 4- to 5-year period identified 36 cases where susceptibility weighted imaging was used. Cases were selected from this group to demonstrate the potential utility of this sequence.

RESULTS

This sequence is most frequently useful not only in the assessment of spinal abnormalities, most commonly the bony abnormalities in myelomeningocele, but also in cases of scoliosis, segmentation anomalies and sacrococcygeal teratoma.

CONCLUSION

Although the utility of this sequence is still being evaluated, it provides excellent contrast between the mineralized skeleton and surrounding soft tissues compared with standard half Fourier acquisition single-shot turbo-spin echo sequences. Further assessment is required to determine whether black bone MRI can more accurately evaluate the level of bony defect in spina bifida aperta, an important prognostic factor. Potential further uses include the assessment of skeletal dysplasias, evaluation of the skull base and craniofacial skeleton in certain congenital anomalies and the post-mortem evaluation of the foetal skeleton potentially obviating the need for necropsy.

ADVANCES IN KNOWLEDGE

Foetal black bone MRI can be performed using susceptibility weighted imaging and allows better demonstration of the mineralized skeleton compared with standard sequences.

Normal mandibular growth and diagnosis of micrognathia at prenatal MRI

OBJECTIVE

The aim of this article is to present reference data for fetal mandibular growth on magnetic resonance imaging, enabling the diagnosis of micrognathia.

METHODS

Retrospectively, on 355 magnetic resonance scans of apparently facially normal fetuses [gestational age (GA), 20-36 weeks], mandibular anterior-posterior diameter (APD = mandibular size), inferior facial angle (IFA = mandibular position), and jaw index (APD normalized to biparietal diameter) were correlated with GA by Pearson correlation. APD-age relationship was modeled. A receding chin was subjectively determined. Ten fetuses with mandibular anomalies were compared with normal fetuses.

RESULTS

For GA, APD showed high correlation (r = 0.850; P < 0.001), IFA (r = 0.086; P = 0.119) no correlation, and jaw index (r = -0.139; P = 0.018) weak correlation. APD-age relationship was expressed by the following: APD = 0.281 + 0.989 * GA (r(2)  = 0.723). A receding chin was identified in 7/10 abnormal fetuses. APD, IFA, and jaw index of abnormal and normal fetuses were significantly different (P < 0.001). In 10/10 abnormal fetuses, IFA was <50.0°; in 7/10, jaw index was less than the fifth percentile (micrognathia); in 3/10, jaw index was at low normal range (retrognathia).

CONCLUSIONS

Subjective identification of micrognathia may be limited. Reference data provide quantitative evaluation of mandibular size and position. An IFA <50° reflects micrognathia or retrognathia; a jaw index less than the fifth percentile suggests micrognathia.