Pediatric Radiology最新文献

Background: Diffusion magnetic resonance imaging has emerged as an opportunity to explore brain white matter fiber tracts (WMFTs) through 3D digital reconstruction. This method could be useful in investigating the relationship between positional plagiocephaly and developmental problems; however, this has not been fully explored.

Objective: Evaluate WMFTs of healthy infants in two age groups with a range of positional plagiocephaly from normal to severe.

Materials and methods: This exploratory study, conducted at a free-standing, quaternary pediatric hospital in the Northeastern United States, utilized an existing database of healthy infants' MRIs obtained between 1 month and 4 months of age. MRIs were included if deemed good quality and had complete T1- and diffusion-weighted sequences and excluded if there were measurement disagreements or MRI data processing problems. Positional plagiocephaly severity was calculated using the Cranial Vault Asymmetry Index (CVAI). A repeated-measures regression model was constructed to assess the association of positional plagiocephaly severity with WMFTs fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD).

Results: Median age of 18 infants was 64.5 (IQR 71) days at the time of MRI. FA had a negative association with CVAI overall (β±SE=-0.53±0.51% per unit CVAI, P=0.32) and in both age groups. MD and RD had a positive association with CVAI overall (β±SE=1.31±0.46% per unit CVAI, P=0.013; β±SE=1.54±0.54% per unit CVAI, P=0.012) and in both age groups and all pathways.

Conclusion: As the severity of positional plagiocephaly increases, differences in WMFT formation are observed, suggesting the need for longitudinal studies with cognitive and behavioral assessments.

Cranial fasciitis is a rare pediatric skull tumor with a challenging radiological diagnosis, particularly when it arises in unusual locations or presents with atypical features. We report a case of cranial fasciitis originating from the ethmoid bone in a 17-month-old girl, presenting as isolated facial asymmetry instead of the usual scalp mass. Cross-sectional and metabolic imaging demonstrated an aggressive-appearing ethmoid sinus mass mimicking rhabdomyosarcoma, with extensive bone destruction and intracranial extension. The patient underwent complete surgical resection of the lesion. This report emphasizes the need to consider cranial fasciitis in the differential diagnosis of any aggressive-appearing skull mass in childhood, and underscores the importance of recognizing the wide anatomical and clinical spectrum of the lesion.

The integration of artificial intelligence (AI) in pediatric radiology requires an interdisciplinary approach that prioritizes transparency, accountability and collaboration between developers, clinicians and regulatory bodies. The development of AI models that are specifically designed to analyze pediatric imaging data has the potential to improve diagnosis and treatment outcomes, but it also requires careful consideration of the ethical implications. This review highlights the importance of the unique challenges posed by AI in pediatric imaging data, including regulatory hurdles, bias mitigation and the need for human oversight. Facing this situation, pediatric radiologists need to be equipped with the skills and knowledge to critically evaluate AI outputs and address potential biases and limitations. This requires ongoing education and training in pediatric radiology as well as AI. The integration of AI in pediatric radiology requires a collaborative approach that involves not only developers and clinicians but also patients and families. Ultimately, the integration of AI in pediatric imaging needs to be a coordinated effort from all stakeholders to prioritize the long-term safety and health of the young patients.

Background: Childhood obesity is rising worldwide, leading to an increased prevalence of metabolic dysfunction-associated steatotic liver disease, the most common pediatric liver disease. Ultrasound-derived fat fraction (UDFF) is a recently developed technique for quantifying hepatic fat that has been validated in children, providing a rapid, reliable, and non-invasive alternative to conventional imaging.

Objective: To evaluate the association between weight status, expressed as body mass index (BMI) z-score, and UDFF in a pediatric population without underlying liver disease.

Materials and methods: Abdominal ultrasounds including a UDFF measurement in patients without liver disease were retrospectively evaluated. We calculated the BMI z-score for all patients and classified them as normal weight (zBMI) or overweight (zBMI≥1). UDFF values were compared across zBMI categories, sex and ethnicity (Hispanic/Latino or non-Hispanic/non-Latino), using chi-square tests (P<0.05). Pearson's correlation was used to assess the relationship between continuous UDFF values and zBMI. Logistic regression analyses were performed to assess associations between elevated UDFF (>6%) and adiposity markers (zBMI and ultrasound-measured abdominal wall thickness (AWT)), as well as ethnicity.

Results: Of 223 subjects, 93 (41 males/52 females, mean age of 8.2 years) met the inclusion criteria. In the normal-weight population (n=46), 41 (89.1%) had normal UDFF, and 5 (10.9%) had elevated UDFF. In the overweight group (n =47), 29 (61.7%) had normal UDFF and 18 (38.3%) had elevated UDFF. UDFF values showed a positive correlation with zBMI, and higher zBMI increased the odds of elevated UDFF. In the multivariable model including zBMI, AWT, and ethnicity, ultrasound-measured abdominal wall thickness was the strongest predictor of elevated UDFF.

Conclusion: BMI z-score was positively associated with hepatic fat content and with higher odds of elevated UDFF. When multiple factors were considered together, ultrasound-measured abdominal wall thickness showed the strongest independent association with elevated UDFF supporting the central role of adiposity in pediatric hepatic fat accumulation. UDFF may serve as a valuable complement to routine clinical markers, using zBMI, for early identification of children with hepatic steatosis. Larger prospective studies are needed to define its role in clinical practice.

Palpable lymph nodes are among the most common indications for ultrasound evaluation in the pediatric population. Ultrasound provides valuable insight into nodal composition by assessing greyscale morphology, color Doppler vascularity, and vascular resistance, with emerging techniques such as elastography and contrast-enhanced ultrasound offering additional diagnostic potential. Although certain sonographic features may suggest a benign or malignant etiology, imaging overlap often exists posing diagnostic challenges. This review provides an overview of the sonographic appearance of the most common pediatric lymphadenopathies, including reactive hyperplasia, bacterial and viral lymphadenitis, necrotizing and granulomatous lymphadenitis, malignant and atypical entities. Characteristic and non-specific imaging features are discussed, along with practical approaches to interpretation and current strategies for diagnosis and management.

Background: Myelination is a key biomarker of healthy brain maturation, and its disruption can signal neurodevelopmental disorders.

Objective: The study aimed to enhance the accuracy and interpretability of brain age prediction in early infancy by incorporating the biological process of myelination as an attention mechanism into deep learning models.

Materials and methods: A fully automated deep learning framework, called myelination-attention-empowered model (MAENet), was developed through retrospective analysis of structural magnetic resonance imaging (sMRI) data from 603 participants who met the inclusion criteria, aged 0-2 years, collected in a local hospital between July 2017 and June 2024. The MAENet consisted of four modules: a multiscale information fusion channel (MSIF-channel) on the T2WI brain image, a myelination-empowered feature extraction channel (MEFE-channel) on an automated and standardized segmentation of the white matter image, a communication mechanism that enabled inter-channel information flow and enhanced the MSIF-channel's sensitivity to myelination-related features, and a myelination-attention mechanism that dynamically emphasized myelination-sensitive regions.

Results: The proposed MAENet model exhibited superior performance over multiple deep learning models, including ResNet-50, VGG, Inception, SFCN, Skewed, FiA-Net, and TSAN. The mean absolute error (MAE) between the predicted brain age and chronological age was significantly reduced by 18%-41% in the subgroup of 0-1-year-old infants, 25%-37% in the subgroup of 1-2-year-old infants, and 18%-40% in the whole group of 0-2-year-old infants in the experimental comparison (P < 0.05). The brain regions attended to by the MAENet model were visualized and consistent with the well-known developmental trajectories of white matter myelination in early infancy.

Conclusion: The MAENet model demonstrated a significant improvement in brain age prediction accuracy in 0-2-year-olds by effectively leveraging the developmental process of myelination.

Primary vertebral tumors in children are rare but clinically significant due to their potential for neurological compromise, spinal instability, and growth-related deformities. These lesions encompass a wide histologic spectrum, ranging from benign entities, such as osteoid osteoma and osteoblastoma, to aggressive malignancies like Ewing sarcoma and osteosarcoma. Despite their rarity, timely and accurate diagnosis is essential, as delays can result in irreversible neurological deficits. Imaging is fundamental across the clinical continuum, from detection and characterization to treatment planning and post-therapeutic monitoring. Conventional radiography remains a common first-line tool, although limited in sensitivity for early or subtle findings. Computed tomography (CT) is particularly useful for evaluating cortical integrity and detecting tumor mineralization, findings that are especially relevant in lesions such as osteoid osteoma, where CT also plays a key role in planning and guiding percutaneous treatment. Magnetic resonance imaging (MRI), the modality of choice, offers superior soft tissue contrast and is essential for assessing bone marrow involvement and adjacent neural structures. Advanced MRI techniques, such as diffusion-weighted imaging (DWI) and dynamic contrast-enhanced studies, further enhance lesion characterization and may help predict response to treatment. Minimally invasive techniques, including image-guided biopsy, radiofrequency ablation, and cryoablation, have expanded the therapeutic techniques, particularly for benign lesions. This review provides a comprehensive overview of the classification and imaging features of pediatric spinal bone tumors, with emphasis on the strengths and limitations of each modality and the evolving role of interventional radiology in diagnosis and treatment.

Background: Accurate Cobb angle measurement is essential in adolescent idiopathic scoliosis (AIS), but evidence on artificial intelligence (AI) performance in pediatric patients, especially with severe curves, is limited.

Objective: This study evaluated the accuracy of a commercially available deep learning software in measuring Cobb angles in surgical cases of AIS and compared its performance with that of radiology residents.

Methods: A total of 151 preoperative anteroposterior whole spine X-rays were analyzed. The ground truth was established by consensus between a pediatric radiologist and a pediatric orthopedic surgeon. The mean absolute error (MAE) of the AI and six radiology residents was calculated. Wilcoxon signed-rank test and intraclass correlation coefficients (ICC) were computed.

Results: The dataset included 151 angles categorized as moderate (13; 20-39°), severe (74; 40-59°), and extreme (64; ≥60°). Overall, the AI's MAE was 4.57° [95%CI 3.85°, 5.29°], significantly higher than that of the residents' (P=0.0017). Scoliosis severity significantly affected MAE in both groups. For extreme scoliosis, the MAE of the AI (6.53° [95%CI 5.06°, 7.86°]) was significantly higher than that of the residents (3.61° [95%CI, 3.17°, 3.97°]) (P<0.001). No significant difference in MAE was observed between patients with moderate and severe scoliosis (P=0.86). The agreement between the ICC between the AI and the ground truth was 0.89 [95%CI 0.69, 0.95].

Conclusion: The Cobb angle measurements obtained with this software agree with those of experts for moderate and severe scoliosis with a clinically acceptable average error, without significant difference compared to radiology residents. AI accuracy significantly decreases in patients with extreme scoliosis, highlighting the need for radiologist oversight.

Bifid ribs are rare congenital anomalies, usually asymptomatic and discovered incidentally on imaging. However, they can cause significant pain, posing diagnostic and therapeutic challenges, particularly in pediatric populations. We report a case of a 3-year-old female with chronic anterior chest wall pain attributed to a bifid right fourth rib. Imaging confirmed the anomaly without associated soft tissue masses. After transient symptom relief with lidocaine patches and intercostal nerve blocks, she underwent successful percutaneous image-guided cryoneurolysis of the right third to fifth intercostal nerves. The patient experienced marked pain relief post-procedure without complications, with sustained benefit at 4 months. Bifid ribs may be an under-recognized cause of intercostal neuralgia in children. Image-guided percutaneous cryoneurolysis may represent a safe and effective treatment option for symptomatic relief.