Genetic variants of phospholipase C-γ2 alter the phenotype and function of microglia and confer differential risk for Alzheimer's disease.

Genetic association studies have demonstrated the critical involvement of the microglial immune response in Alzheimer's disease (AD) pathogenesis. Phospholipase C-gamma-2 (PLCG2) is selectively expressed by microglia and functions in many immune receptor signaling pathways. In AD, PLCG2 is induced uniquely in plaque-associated microglia. A genetic variant of PLCG2, PLCG2P522R, is a mild hypermorph that attenuates AD risk. Here, we identified a loss-of-function PLCG2 variant, PLCG2M28L, that confers an increased AD risk. PLCG2P522R attenuated disease in an amyloidogenic murine AD model, whereas PLCG2M28L exacerbated the plaque burden associated with altered phagocytosis and Aß clearance. The variants bidirectionally modulated disease pathology by inducing distinct transcriptional programs that identified microglial subpopulations associated with protective or detrimental phenotypes. These findings identify PLCG2M28L as a potential AD risk variant and demonstrate that PLCG2 variants can differentially orchestrate microglial responses in AD pathogenesis that can be therapeutically targeted.

APOE4/4 is linked to damaging lipid droplets in Alzheimer's disease microglia.

Several genetic risk factors for Alzheimer's disease implicate genes involved in lipid metabolism and many of these lipid genes are highly expressed in glial cells1. However, the relationship between lipid metabolism in glia and Alzheimer's disease pathology remains poorly understood. Through single-nucleus RNA sequencing of brain tissue in Alzheimer's disease, we have identified a microglial state defined by the expression of the lipid droplet-associated enzyme ACSL1 with ACSL1-positive microglia being most abundant in patients with Alzheimer's disease having the APOE4/4 genotype. In human induced pluripotent stem cell-derived microglia, fibrillar Aß induces ACSL1 expression, triglyceride synthesis and lipid droplet accumulation in an APOE-dependent manner. Additionally, conditioned media from lipid droplet-containing microglia lead to Tau phosphorylation and neurotoxicity in an APOE-dependent manner. Our findings suggest a link between genetic risk factors for Alzheimer's disease with microglial lipid droplet accumulation and neurotoxic microglia-derived factors, potentially providing therapeutic strategies for Alzheimer's disease.

Deep learning-driven adaptive optics for single-molecule localization microscopy.

The inhomogeneous refractive indices of biological tissues blur and distort single-molecule emission patterns generating image artifacts and decreasing the achievable resolution of single-molecule localization microscopy (SMLM). Conventional sensorless adaptive optics methods rely on iterative mirror changes and image-quality metrics. However, these metrics result in inconsistent metric responses and thus fundamentally limit their efficacy for aberration correction in tissues. To bypass iterative trial-then-evaluate processes, we developed deep learning-driven adaptive optics for SMLM to allow direct inference of wavefront distortion and near real-time compensation. Our trained deep neural network monitors the individual emission patterns from single-molecule experiments, infers their shared wavefront distortion, feeds the estimates through a dynamic filter and drives a deformable mirror to compensate sample-induced aberrations. We demonstrated that our method simultaneously estimates and compensates 28 wavefront deformation shapes and improves the resolution and fidelity of three-dimensional SMLM through >130-µm-thick brain tissue specimens.

Case 326: Intra-Articular Osteoid Osteoma.

HISTORY: A 15-year-old male patient presented with a 3-week history of inner left thigh pain provoked by activity and experienced occasionally at rest. The patient denied nighttime pain, fever, or chills. Laboratory investigation revealed the following normal values: hemoglobin level of 15.6 g/dL (normal range, 13-16 g/dL), platelet count of 240 × 103/µL (normal range, 140-440 × 103/µL), and total leukocyte count of 7100 cells/µL (normal range, 4500-11 000 cells/µL). The percentage of neutrophils was considered low at 44% (normal range, 54%-62%), and the percentage of eosinophils was slightly high at 3.7% (normal range, 0%-3%). An anteroposterior radiograph of the left hip is shown. Physical therapy was initiated, with no improvement after 2 weeks of therapy. The patient was referred to an orthopedist for further evaluation. At physical examination, the patient endorsed marked left hip pain with hip flexion to 90°, limited internal and external rotation (5° and 15°, respectively), and antalgic gait favoring the left leg. Hip MRI and further serologic analysis were requested for further evaluation. Although the serologic testing was performed at an outside laboratory, the physician reported positive immunoglobulin-G Lyme titers, normal C-reactive protein level, and normal erythrocyte sedimentation rate. Pelvic CT was requested. The patient was prescribed a course of doxycycline (100 mg twice daily for 28 days), with reported resolution of symptoms 2 weeks after initiation of treatment. Three weeks later, the patient presented to our department with recurrent left hip pain, which was similar in severity compared with the initial presentation. A second MRI examination of the left hip was performed 4 months after the initial presentation.

Case 326.

HISTORY: A 15-year-old boy presented with a 3-week history of inner left thigh pain provoked by activity and experienced occasionally at rest. He denied nighttime pain, fever, or chills. Laboratory investigation revealed the following normal values: hemoglobin level of 15.6 g/dL (normal range, 13-16 g/dL), platelet count of 240 × 103/µL (normal range, 140-440 × 103/µL), and total leukocyte count of 7100 cells/µL (normal range, 4500-11 000 cells/µL). The percentage of neutrophils was considered low at 44% (normal range, 54%-62%), and the percentage of eosinophils was slightly high at 3.7% (normal range, 0%-3%). An anteroposterior radiograph of the left hip is shown (Fig 1). Physical therapy was initiated, with no improvement after 2 weeks of therapy. The patient was referred to an orthopedist for further evaluation. On physical examination, the patient endorsed marked left hip pain with hip flexion to 90°, limited internal and external rotation (5° and 15°, respectively), and antalgic gait favoring the left leg. Hip MRI (Fig 2) and further serologic analysis were requested for further evaluation. Although the serologic testing was performed at an outside laboratory, the physician reported positive immunoglobulin-G Lyme titers, normal C-reactive protein level, and normal erythrocyte sedimentation rate. Pelvic CT was requested (Fig 3). The patient was prescribed a course of doxycycline (100 mg twice daily for 28 days), with reported resolution of symptoms 2 weeks after initiation of treatment. Three weeks later, he presented to our department with recurrent left hip pain, which was similar in severity compared with initial presentation. A second MRI of the left hip was performed 4 months after initial presentation (Fig 4).

Lower Extremity Growth according to AI Automated Femorotibial Length Measurement on Slot-Scanning Radiographs in Pediatric Patients.

Background Commonly used pediatric lower extremity growth standards are based on small, dated data sets. Artificial intelligence (AI) enables creation of updated growth standards. Purpose To train an AI model using standing slot-scanning radiographs in a racially diverse data set of pediatric patients to measure lower extremity length and to compare expected growth curves derived using AI measurements to those of the conventional Anderson-Green method. Materials and Methods This retrospective study included pediatric patients aged 0-21 years who underwent at least two slot-scanning radiographs in routine clinical care between August 2015 and February 2022. A Mask Region-based Convolutional Neural Network was trained to segment the femur and tibia on radiographs and measure total leg, femoral, and tibial length; accuracy was assessed with mean absolute error. AI measurements were used to create quantile polynomial regression femoral and tibial growth curves, which were compared with the growth curves of the Anderson-Green method for coverage based on the central 90% of the estimated growth distribution. Results In total, 1874 examinations in 523 patients (mean age, 12.7 years ± 2.8 [SD]; 349 female patients) were included; 40% of patients self-identified as White and not Hispanic or Latino, and the remaining 60% self-identified as belonging to a different racial or ethnic group. The AI measurement training, validation, and internal test sets included 114, 25, and 64 examinations, respectively. The mean absolute errors of AI measurements of the femur, tibia, and lower extremity in the test data set were 0.25, 0.27, and 0.33 cm, respectively. All 1874 examinations were used to generate growth curves. AI growth curves more accurately represented lower extremity growth in an external test set (n = 154 examinations) than the Anderson-Green method (90% coverage probability: 86.7% [95% CI: 82.9, 90.5] for AI model vs 73.4% [95% CI: 68.4, 78.3] for Anderson-Green method; χ2 test, P < .001). Conclusion Lower extremity growth curves derived from AI measurements on standing slot-scanning radiographs from a diverse pediatric data set enabled more accurate prediction of pediatric growth. © RSNA, 2024 Supplemental material is available for this article.

Intra-articular Osteoid Osteomas: Imaging Manifestations and Mimics.

Osteoid osteoma (OO) is the third most prevalent benign bone neoplasm in children. Although it predominantly affects the diaphysis of long bones, OO can assume an intra-articular location in the epiphysis or the intracapsular portions of bones. The most common location of intra-articular OO is the hip joint. The presentation of intra-articular OOs often poses a diagnostic enigma, both from clinical and radiologic perspectives. Initial symptoms are often vague and nonspecific, characterized by joint pain, stiffness, and limited range of motion, which frequently contributes to a delayed diagnosis. Radiographic findings range from normal to a subtle sclerotic focus, which may or may not have a lucent nidus. In contrast to their extra-articular counterparts, intra-articular lesions have distinct features at MRI, including synovitis, joint effusion, and bone marrow edema-like signal intensity. While CT remains the standard for identifying the nidus, even CT may be inadequate in visualizing it in some cases, necessitating the use of bone scintigraphy or fluorine 18-labeled sodium fluoride PET/CT for definitive diagnosis. Radiologists frequently play a pivotal role in suggesting this diagnosis. However, familiarity with the unique imaging attributes of intra-articular OO is key to this endeavor. Awareness of these distinctive imaging findings of intra-articular OO is crucial for avoiding diagnostic delay, ensuring timely intervention, and preventing unnecessary procedures or surgeries resulting from a misdiagnosis. The authors highlight and illustrate the different manifestations of intra-articular OO as compared with the more common extra-articular lesions with respect to clinical presentation and imaging findings. ©RSNA, 2024 Supplemental material is available for this article.

Perioperative Factors and Radiographic Severity Scores for Predicting the Duration of Mechanical Ventilation After Arterial Switch Surgery.

OBJECTIVES: Cardiac surgery on cardiopulmonary bypass (CPB) during the neonatal period can cause perioperative organ injuries. The primary aim of this study was to determine the incidence and risk factors associated with postoperative mechanical ventilation duration and acute lung injury after the arterial switch operation (ASO). The secondary aim was to examine the utility of the Brixia score for characterizing postoperative acute lung injury (ALI). DESIGN: A retrospective study. SETTING: A single-center university hospital. PARTICIPANTS: A total of 93 neonates with transposition of great arteries with intact ventricular septum (dTGA IVS) underwent ASO. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: From January 2015 to December 2022, 93 neonates with dTGA IVS were included in the study. The cohort had a median age of 4.0 (3.0-5.0) days and a mean weight of 3.3 ± 0.5 kg. About 63% of patients had ≥48 hours of postoperative mechanical ventilation after ASO. Risk factors included prematurity, post-CPB transfusion of salvaged red cells, platelets and cryoprecipitate, and postoperative fluid balance by univariate analysis. The larger transfused platelet volume was associated with the risk of ALI by multivariate analysis. The median baseline Brixia scores were 11.0 (9.0-12.0) and increased significantly in the postoperative day 1 in patients who developed moderate ALI 24 hours after admission to the intensive care unit (15.0 [13.0-16.0] v 12.0 [10.0-14.0], p = 0.046). CONCLUSIONS: Arterial switch operation results in a high incidence of ≥48-hour postoperative mechanical ventilation. Blood component transfusion is a potentially modifiable risk factor. The Brixia scores also may be used to characterize postoperative acute lung injury.

Integrating multimodality magnetic resonance imaging to the Allen Mouse Brain Common Coordinate Framework.

High-resolution magnetic resonance imaging (MRI) affords unique image contrasts to nondestructively probe the tissue microstructure; validation of MRI findings with conventional histology is essential to better understand the MRI contrasts. However, the dramatic difference in the spatial resolution and image contrast of these two techniques impedes accurate comparison between MRI metrics and traditional histology. To better validate various MRI metrics, we acquired whole mouse brain multigradient recalled-echo and multishell diffusion MRI datasets at 25-µm isotropic resolution. The recently developed Allen Mouse Brain Common Coordinate Framework (CCFv3) provides opportunities to integrate multimodal and multiscale datasets of the whole mouse brain in a common three-dimensional (3D) space. The T2*, quantitative susceptibility mapping, diffusion tensor imaging, and neurite orientation dispersion and density imaging parameters were compared with both serial two-photon tomography images and 3D Nissl staining images in the CCFv3 at the same spatial resolution. The correlation between MRI and Nissl staining strongly depends on different metrics and different regions of the brain. Integrating different imaging modalities to the same space may substantially improve our understanding of the complexity of the brain at different scales.

The added value of skeletal surveys in the initial evaluation of children diagnosed with Langerhans cell histiocytosis in the era of staging 18 F-FDG PET/CT: A retrospective study.

OBJECTIVE: Currently, there is no consensus protocol on the initial staging evaluation for Langerhans cell histiocytosis (LCH). Our institutional protocol consists of a skeletal survey and a whole-body positron emission tomography with 2-deoxy-2-[fluorine-18] fluoro-D-glucose integrated with computed tomography (FDG PET/CT) study. The utility of the PET/CT lies in its sensitivity in detecting osseous and extra-osseous lesions, and in determining the baseline metabolic activity of LCH lesions to assess treatment response. However, the added utility of the skeletal survey in staging LCH is unclear. Therefore, this study retrospectively assessed the added diagnostic value of skeletal surveys in patients with baseline PET/CTs for initial staging of LCH. METHODS: We retrospectively searched the medical records of all patients less than or equal to 18 years old at a large children's hospital (May 2013 to September 2021). The inclusion criteria were (a) biopsy-proven diagnosis of LCH and (b) initial staging PET/CT and skeletal survey performed less than or equal to 1 month apart. A blinded pediatric radiologist reviewed the skeletal surveys and another reviewed the PET/CTs in identifying LCH osseous lesions. RESULTS: Our study cohort consisted of 49 children with 86 LCH osseous lesions. In non-extremity locations, PET/CT identified 70/70 (100%) osseous lesions, while skeletal surveys detected 43/70 (61.4%) osseous lesions. In the extremities, PET/CT identified 13/16 (81.3%) osseous lesions, while skeletal surveys detected 15/16 (93.8%) osseous lesions. CONCLUSION: Skeletal surveys increased the detection rate of osseous lesions in the extremities, but added no diagnostic value to the detection of osseous lesions in non-extremity locations. Therefore, we propose to abbreviate the skeletal survey to include only extremity radiographs.

Reconsidering pregnancy screening policies for minors: patient-specific estimate of fetus and effective dose for potentially pregnant minors undergoing optimized dose CT of the pelvis.

BACKGROUND: Only verbal pregnancy screening is recommended for post-menarcheal females undergoing pelvic radiographs. In contrast, usually, a urine/serum pregnancy test for pelvic computed tomographic (CT) exams is required out of concern for higher radiation exposure. OBJECTIVE: To estimate patient-specific fetus absorbed dose to a potentially pregnant minor from an optimized dose CT of the pelvis for femoral version and surgical planning and provide evidence that such examinations of the pelvis can be performed with only verbal pregnancy screening. METHODS AND METHODS: A retrospective study was performed on 102 female patients between 12-18 years of age (15.4 ± 2.1 years) who underwent optimized dose CT of the pelvis for orthopedic evaluation of femoral version and surgical planning. Optimized CT exams were performed with weight-adjusted kVp and tube current modulation. Patient-specific dose from the optimized dose CT was calculated using the National Cancer Institute Dosimetry System for CT (NCICT) database by matching each patient to a phantom from the NCI non-reference phantom library based on patient sex, weight, and height. The calculated absorbed uterus dose was used as a surrogate for the fetus dose. Furthermore, patient-specific organ doses were used to estimate the effective dose. The strengths of the linear relationships between the dose metrics and patient characteristics were assessed using Pearson correlation coefficients through linear regression. RESULTS: The mean patient-specific effective dose for an optimized dose CT of the pelvis was 0.54 ± 0.20 mSv (range: 0.15-1.22 mSv). The mean estimated absorbed uterine dose was 1.57 ± 0.67 mGy (range: 0.42-4.81 mGy). Both effective dose and estimated uterine dose correlated poorly with patient physical characteristics (R = -0.26; 95% CI: [-0.43, -0.007] for age, R = 0.03; 95% CI: [-0.17, 0.22] for weight) but correlated strongly (R = 0.79, 95% CI: [0.7, 0.85]) with CTDIvol. CONCLUSION: The estimated fetus dose in case of pregnancy was significantly lower than 20 mGy for urine/serum pregnancy screening, suggesting that the pregnancy screening protocols in minors undergoing optimized dose CT require reassessment and may be safely performed by verbal attestation only.

Pelvis radiographs in children with cerebral palsy: effects of patient positioning on calculating migration percentages.

BACKGROUND: Hip displacement in children with cerebral palsy (CP) is monitored by measuring migration percentage on anteroposterior pelvis radiographs. However, proper positioning for radiography in children with spasticity is difficult. The reliability and accuracy of migration percentage as a function of patient positioning is unknown. OBJECTIVE: To determine the effects of patient positioning on migration percentage measurements in children with CP. MATERIALS AND METHODS: We identified children with CP (≤18-year-old) with pelvis CT and anteroposterior pelvis radiograph obtained <6 months apart (10/2018-11/2021). Digitally reconstructed radiographs were generated from each pelvis CT, to simulate nine different patient positions: neutral; 10° and 20° lordosis and kyphosis; and 10° and 20° right rotation and left rotation. Two radiologists measured migration percentages from the simulated and real pelvis radiographs. We used Spearman's rho to assess inter-rater reliability, and Wilcoxon signed rank test to determine statistical significance. RESULTS: We studied sixty-three children (male=41; median age=8 years; range=4-18 years). The two radiologists' migration percentage measurements were highly correlated with each other across all simulated and real radiographs (Spearman's rho=0.86-0.99, P<0.01). For both readers and hips, migration percentages measured from real radiographs were significantly different from those measured from neutral simulated radiographs (P<0.01), with median absolute difference=5-6 percentage points (PP) and interquartile range (IQR)=9-12 PP. When comparing migration percentage measurements from neutral simulated radiographs to those in kyphosis/lordosis and right/left rotations, median absolute differences were 2-4 PP (IQR=3-8 PP) and 4-15 PP (IQR=6-17 PP), respectively. CONCLUSION: Inter-rater reliability of measured migration percentages is high, but accuracy decreases with patient positional changes.

Dating birth-related clavicular fractures: pediatric radiologists versus artificial intelligence.

BACKGROUND: Fracture dating from skeletal surveys is crucial in the diagnosis and investigation of infant abuse. However, this task is challenging because of the subjective nature of the radiologic interpretation and the lack of ground truth. Researchers have used birth-related clavicle fractures as a surrogate to study the radiographic pattern of healing; however, they did not elucidate the accuracy performance of the radiologists in dating fractures. OBJECTIVE: To determine the accuracy of radiologists in dating birth-related clavicle fractures and compare their performance to that achieved by computer algorithm. MATERIALS AND METHODS: We used a previously assembled birth-related clavicle fracture database consisting of 416 anteroposterior clavicle radiographs as the study cohort. The average and standard deviation of the fracture age within this database were 24 days and 18 days, respectively. Three blinded radiologists independently estimated the ages of the clavicle fractures depicted in the radiographs within the database. We compared these estimation results to those made by a recently published deep-learning (DL) model conducted with the identical infant cohort. We calculated standard error metrics to compare the accuracy performances of the radiologists and the computer model. RESULTS: The intra- and inter-reader agreements of the fracture age estimates by the radiologists were moderate to good. The radiologists estimated the fracture ages with a mean absolute error (MAE) of 6.1-7.1 days, and standard deviation of the absolute error of 6.3-8.3 days. The accuracy performances of the three radiologists were not significantly different from one another. In comparison, the DL model estimated the age of clavicle fractures with an MAE of 4.2 days, significantly lower than all of the radiologists (P < 0.001). CONCLUSION: Three experienced pediatric radiologists dated clavicular fractures with moderate-good intra- and inter-reader agreements. The correlations between the radiologists' estimates and the ground truth were moderate to good. The fracture ages assigned by the DL model showed superior correlation with the ground truth compared to radiologists' dating estimates.

Virtual 3D femur model to assess femoral version: comparison to the 2D axial slice approach.

BACKGROUND: Quantifying femoral version is crucial in diagnosing femoral version abnormalities and for accurate pre-surgical planning. There are numerous methods for measuring femoral version, however, reliability studies for most of these methods excluded children with hip deformities. OBJECTIVE: To propose a method of measuring femoral version based on a virtual 3D femur model, and systematically compare its reliability to the widely used Murphy's 2D axial slice technique. MATERIALS AND METHODS: We searched our imaging database to identify hip/femur CTs performed on children (<18 years old) with a clinical indication of femoral version measurement (September 2021-August 2022). Exclusion criteria were prior hip surgery, and inadequate image quality or field-of-view. Two blinded radiologists independently measured femoral version using the virtual 3D femur model and Murphy's 2D axial slice method. To assess intrareader variability, we randomly selected 20% of the study sample for re-measurements by the two radiologists >2 weeks later. We analyzed the reliability and correlation of these techniques via intraclass correlation coefficient (ICC), Bland-Altman analysis, and deformity subgroup analysis. RESULTS: Our study sample consisted of 142 femurs from 71 patients (10.6±4.4 years, male=31). Intra- and inter-reader correlations for both techniques were excellent (ICC≥0.91). However, Bland-Altman analysis revealed that the standard deviation (SD) of the absolute difference between the two radiologists for the Murphy method (mean 13.7°) was larger than that of the 3D femur model technique (mean 4.8°), indicating higher reader variability. In femurs with hip flexion deformity, the SD of the absolute difference for the Murphy technique was 17°, compared to 6.5° for the 3D femur model technique. In femurs with apparent coxa valga deformity, the SD of the absolute difference for the Murphy technique was 10.4°, compared to 5.2° for the 3D femur model technique. CONCLUSION: The 3D femur model technique is more reliable than the Murphy's 2D axial slice technique in measuring femoral version, especially in children with hip flexion and apparent coxa valga deformities.

INPP5D deficiency attenuates amyloid pathology in a mouse model of Alzheimer's disease.

INTRODUCTION: Inositol polyphosphate-5-phosphatase (INPP5D) is a microglia-enriched lipid phosphatase in the central nervous system. A non-coding variant (rs35349669) in INPP5D increases the risk for Alzheimer's disease (AD), and elevated INPP5D expression is associated with increased plaque deposition. INPP5D negatively regulates signaling via several microglial cell surface receptors, including triggering receptor expressed on myeloid cells 2 (TREM2); however, the impact of INPP5D inhibition on AD pathology remains unclear. METHODS: We used the 5xFAD mouse model of amyloidosis to assess how Inpp5d haplodeficiency regulates amyloid pathogenesis. RESULTS: Inpp5d haplodeficiency perturbs the microglial intracellular signaling pathways regulating the immune response, including phagocytosis and clearing of amyloid beta (Aß). It is important to note that Inpp5d haploinsufficiency leads to the preservation of cognitive function. Spatial transcriptomic analysis revealed that pathways altered by Inpp5d haploinsufficiency are related to synaptic regulation and immune cell activation. CONCLUSION: These data demonstrate that Inpp5d haplodeficiency enhances microglial functions by increasing plaque clearance and preserves cognitive abilities in 5xFAD mice. Inhibition of INPP5D is a potential therapeutic strategy for AD.

Density map estimation with convolutional neural networks to count radiopaque markers on colonic transit studies.

BACKGROUND: A radiopaque marker study measures colonic transit time for work-up of primary constipation. It requires the patient to ingest multiple tiny radiopaque markers, which the radiologist must count manually on follow-up abdominal radiographs. Counting these markers is tedious but cognitively simple. OBJECTIVE: To develop a convolutional neural network (CNN) capable of counting the number of radiopaque markers on abdominal radiographs. MATERIALS AND METHODS: The image archive at a large tertiary children's hospital was searched to identify abdominal radiographs performed in children for the indication of a radiopaque marker study. To establish the ground truth, a radiologist manually labeled the coordinates of the radiopaque markers in each radiograph and thereby generated a density map for that radiograph. A CNN was trained to estimate this density map from its corresponding abdominal radiograph. Spatially integrating the output density map provided an estimate of the number of markers in the radiograph. To assess model accuracy, mean absolute error and root mean square error were calculated. RESULTS: The study cohort consisted of 436 radiographs (mean number of markers per radiograph: 34). This cohort was randomly divided into training, validation and testing sets consisting of 306, 65 and 65 radiographs, respectively. Based on the testing set, the CNN accurately estimated the number of markers in each radiograph with mean absolute error=2.6 markers and root mean square error=3.9 markers. CONCLUSION: The proposed CNN generated promising results in counting the number of radiopaque markers on abdominal radiographs and offers the potential of automating the interpretation of colonic transit studies.

Machine learning to identify distal tibial classic metaphyseal lesions of infant abuse: a pilot study.

BACKGROUND: The classic metaphyseal lesion (CML) is an injury that is highly specific for infant abuse, and the distal tibia is one of the most common sites of occurrence. A machine learning tool that identifies distal tibial CMLs on infant skeletal surveys could assist radiologists in the diagnosis of infant abuse. OBJECTIVE: To develop and evaluate a machine learning-based classification algorithm to identify distal tibial CMLs on skeletal surveys performed for suspected infant abuse. MATERIALS AND METHODS: We reviewed medical records of infants (≤1 year old) who had skeletal surveys for suspected abuse at a large tertiary children's hospital over the past 13 years to identify those at low and high risk for abuse. Normal distal tibial radiographs from the low-risk group formed the normal study cohort; radiographs with distal tibial CMLs from the high-risk group formed the abnormal study cohort. We used these two cohorts to train a machine learning algorithm to classify distal tibial radiographs as normal or abnormal. We systematically evaluated this algorithm using a fivefold cross-validation procedure and statistically analyzed the results. RESULTS: The normal study cohort consisted of 177 radiographs from 89 infants, and the abnormal study cohort consisted of 73 radiographs from 35 infants. Our machine learning algorithm showed an overall performance accuracy of 93% and Kappa of 0.84. The overall sensitivity and specificity of the model were 88% and 96%, respectively. CONCLUSION: Our developed machine learning model shows encouraging results as an automated tool to identify CMLs of the distal tibia on skeletal surveys performed for suspected infant abuse.

Diagnostic challenges with central venous catheter studies.

A central venous catheter is typically made of silicone rubber or polyurethane and inserted into a large central vein to provide prolonged and direct access to central venous circulation. These catheters provide a safe and effective method to administer intravenous medications, nutritional supplements, fluids and blood products. However, a myriad of complications is associated with central venous catheters, including, but not limited to, mechanical malfunction or fracture, kinking, erroneous placement, line infection, fibrin sheath formation and venous thrombosis. Following clinical and radiographic evaluation, contrast-enhanced line studies constitute the next best diagnostic tool to assess the functionality of central venous catheters. However, there is a lack of standardization in the literature outlining how these studies should be performed. In addition, the interpretation of these studies can be problematic for general pediatric radiologists, many of whom are often not familiar with placement or manipulation of these catheters. In this pictorial review, we highlight the challenges associated with performing and interpreting fluoroscopically guided contrast injection studies, using case studies drawn from a large tertiary children's hospital database for illustration. Revealing these challenges and understanding their causative mechanisms can improve the performance of these line studies.

Deep learning of birth-related infant clavicle fractures: a potential virtual consultant for fracture dating.

BACKGROUND: In infant abuse investigations, dating of skeletal injuries from radiographs is desirable to reach a clear timeline of traumatic events. Prior studies have used infant birth-related clavicle fractures as a surrogate to develop a framework for dating of abuse-related fractures. OBJECTIVE: To develop and train a deep learning algorithm that can accurately date infant birth-related clavicle fractures. MATERIALS AND METHODS: We modified a deep learning model initially designed for face-age estimation to date infant clavicle fractures. We conducted a computerized search of imaging reports and other medical records at a tertiary children's hospital to identify radiographs of birth-related clavicle fracture in infants ≤ 3 months old (July 2003 to March 2021). We used the resultant database for model training, validation and testing. We evaluated the performance of the deep learning model via a four-fold cross-validation procedure, and calculated accuracy metrics: mean absolute error (MAE), root mean square error (RMSE), intraclass correlation coefficient (ICC) and cumulative score. RESULTS: The curated database consisted of 416 clavicle radiographs from 213 infants. Average chronological age (equivalent to fracture age) at time of imaging was 24 days. This model estimated the ages of the clavicle fractures with MAE of 4.2 days, RMSE of 6.3 days and ICC of 0.919. On average, 83.7% of the fracture age estimates were accurate to within 7 days of the ground truth. CONCLUSION: Our deep learning study provides encouraging results for radiographic dating of infant clavicle fractures. With further development and validation, this model might serve as a virtual consultant to radiologists estimating fracture ages in cases of suspected infant abuse.

The impact of social distancing during the COVID-19 pandemic on rates of pediatric idiopathic intussusception.

BACKGROUND: The causes of idiopathic ileocolic intussusception are unknown, with infection as the most likely culprit. Recently, social distancing measures were implemented during the coronavirus disease 2019 (COVID-19) pandemic to decrease transmissible infectious diseases, creating an opportune setting to study the role of infection on the pathogenesis of intussusception on a population level. OBJECTIVE: To investigate the impact of social distancing on intussusception. MATERIALS AND METHODS: We retrospectively reviewed air contrast enemas and pylorus ultrasounds performed between March 2010 and March 2021 to identify cases of ileocolic intussusception and hypertrophic pyloric stenosis (HPS), using the latter as a negative control. The study time frame was divided into two periods: pre-pandemic (March 2010-February 2020) and pandemic (April 2020-March 2021). The number of cases that occurred in these two time periods were compared using the Poisson regression model. RESULTS: Of the 407 cases of idiopathic ileocolic intussusception identified, 396 occurred in the pre-pandemic time period (mean = 39.6 per 12-month period) and 11 occurred in the 12-month pandemic time period. The mean monthly number of intussusceptions showed a decline of 72% (95% confidence interval [CI] 49-85%) between the pre-pandemic and pandemic time periods (3.3 vs. 0.9 monthly cases; P < 0.001). In contrast, the mean monthly number of HPS did not differ significantly (P = 0.19). CONCLUSION: Social distancing-imposed to curb the spread of infection during the COVID-19 pandemic-resulted in a significant decline in ileocolic intussusception, lending strong support to the theory that infection is the dominant cause of intussusception.