Bone age determination using only the index finger: a novel approach using a convolutional neural network compared with human radiologists.

BACKGROUND: Recently developed convolutional neural network (CNN) models determine bone age more accurately than radiologists. OBJECTIVE: The purpose of this study was to determine whether a CNN and radiologists can accurately predict bone age from radiographs using only the index finger rather than the whole hand. MATERIALS AND METHODS: We used a public anonymized dataset provided by the Radiological Society of North America (RSNA) pediatric bone age challenge. The dataset contains 12,611 hand radiographs for training and 200 radiographs for testing. The index finger was cropped from these images to create a second dataset. Separate CNN models were trained using the whole-hand radiographs and the cropped second-digit dataset using the consensus ground truth provided by the RSNA bone age challenge. Bone age determination using both models was compared with ground truth as provided by the RSNA dataset. Separately, three pediatric radiologists determined bone age from the whole-hand and index-finger radiographs, and the consensus was compared to the ground truth and CNN-model-determined bone ages. RESULTS: The mean absolute difference between the ground truth and CNN bone age for whole-hand and index-finger was similar (4.7 months vs. 5.1 months, P=0.14), and both values were significantly smaller than that for radiologist bone age determination from the single-finger radiographs (8.0 months, P<0.0001). CONCLUSION: CNN-model-determined bone ages from index-finger radiographs are similar to whole-hand bone age interpreted by radiologists in the dataset, as well as a model trained on the whole-hand radiograph. In addition, the index-finger model performed better than the ground truth compared to subspecialty trained pediatric radiologists also using only the index finger to determine bone age. The radiologist interpreting bone age can use the second digit as a reliable starting point in their search pattern.

Correction to: Pediatric ileocolic intussusception: new observations and unexpected implications.

In this article, the fourth author's name is misspelled. The correct spelling, as shown above, should be "Nadia F. Mahmood."

Renal Ultrasound for Infants Younger Than 2 Months With a Febrile Urinary Tract Infection.

OBJECTIVE: The purpose of this study is to determine the performance of renal ultrasound for detecting vesicoureteral reflux (VUR) and obstructive uropathies in infants younger than 2 months with a febrile urinary tract infection (UTI). MATERIALS AND METHODS: We performed a retrospective cohort study of infants younger than 2 months with fever and culture-proven UTI presenting from July 1, 2008, through December 31, 2011, with renal ultrasound and voiding cystourethrogram (VCUG) performed within 30 days of UTI diagnosis. Two pediatric radiologists independently reviewed the renal ultrasound and VCUG findings. Results of the renal ultrasound were deemed abnormal if collecting system dilation, renal size asymmetry, collecting system duplication, urothelial thickening, ureteral dilation, or bladder anomalies were present. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of renal ultrasound were calculated using VCUG as reference standard. RESULTS: Of the 197 patients included (mean [SD] age, 33 [ 15 ] days; male-to-female ratio, 2:1), 25% (n = 49) had VUR grades I-V, with 16% (n = 31) having VUR grades III-V and 8% (n = 15) having VUR grades IV-V. For grades I-V VUR, sensitivity was 32.7% (95% CI, 20.0-47.5%), specificity was 69.6% (95% CI, 61.5-76.9%), PPV was 26.2% (95% CI, 15.8-39.1%), and NPV was 75.7% (95% CI, 67.6-82.7%). For grades III-V VUR, sensitivity was 51.6% (95% CI, 33.1-69.9%), specificity was 72.9% (95% CI, 65.5-79.5%), PPV was 26.2% (95% CI, 15.8-39.1%), and NPV was 89.0% (95% CI, 82.5-93.7%). For grades IV-V VUR, sensitivity was 86.7% (95% CI, 59.5-98.3%), specificity was 73.6% (95% CI, 66.6-79.9%), PPV was 21.3% (95% CI, 11.9-33.7%), and NPV was 98.5% (95% CI, 94.8-99.8%). No obstructive uropathies were diagnosed by VCUG in patients with normal renal ultrasound findings. CONCLUSION: In infants younger than 2 months, a normal renal ultrasound makes the presence of grades IV and V VUR highly unlikely but does not rule out lower grades of VUR.

Thyroid surgery outcomes at a children's hospital: The value of a multidisciplinary team approach.

BACKGROUND: Our purpose is to describe the structure, function and outcomes of our multidisciplinary pediatric thyroid program and to evaluate our experience in comparison to other high-volume centers. METHODS: We reviewed all thyroid operations performed 10/2012 through 09/2019, and examined number of cases per year, patient demographics, procedures, final diagnoses and results. Primary outcomes were hypoparathyroidism and recurrent laryngeal nerve (RLN) injury at 12 months. Data were analyzed using descriptive statistics and univariate analyses. RESULTS: We performed 294 thyroid operations on 279 patients. Seventy-nine percent were female. Median age was 15 years (IQR: 12-17). Operations included total thyroidectomy (65%), lobectomy (30%) and completion thyroidectomy (5%). Most common diagnoses were Graves' disease (35%), malignancy (29%), and benign nodule (20%). We developed an evidence-based clinical pathway and conducted weekly multidisciplinary meetings. A clinical data specialist reviewed process and outcome measures routinely. Overall, 6 patients (2.0%) had hypoparathyroidism and 2 (0.7%) had unilateral RLN injury at 12 months. Two of the patients with clinical suspicion of permanent hypoparathyroidism were ultimately weaned off calcium. Both patients with RLN injury had extensive locally advanced malignant disease involving the nerve. CONCLUSIONS: Our multidisciplinary team achieved excellent long-term outcomes for pediatric thyroid surgery comparable to other high-volume pediatric and adult centers.

Acute abdominal pain presenting as a rare appendiceal duplication: a case report.

INTRODUCTION: Appendiceal duplication is a rare anomaly that can manifest as right lower quadrant pain. There are several variations described for this condition. We recommend aggressive operative management should this anatomical variation present in the presence of acute appendicitis. CASE PRESENTATION: We report the case of a 15-year-old African American girl who presented to our hospital with right lower quadrant pain and was subsequently found to have appendiceal duplication. CONCLUSION: There are two categorical systems that have described and stratified appendiceal duplication. Both classification systems have been outlined and referenced in this case report. A computed tomography scan has been included to provide a visual aid to help identify true vermiform appendiceal duplication. The presence of this anatomical abnormality is not a reason for surgical intervention; however, should this be found in the setting of acute appendicitis, aggressive resection of both appendices is mandatory.

Macrodystrophia lipomatosa: A troubled second big toe.

We report the case of a 4-year-old boy who presented to our facility with an enlarged right second toe. After radiography ruled out acute pathology, an MRI showed increased adipose tissue at the distal aspect of the appendage. Although the parent felt that the appendage was unaesthetic, there was no functional issue, and further discussion with the pediatricians and pediatric surgeons eliminated the need for immediate intervention. Macrodystrophia lipomatosa is a rare entity that predominantly affects children. Distributed evenly in males and females, the disease refers to hypertrophy of fibroadipose tissue of the distal upper and lower extremities.

Prenatal and Neonatal MRI of Sacrococcygeal Teratoma With Surgical Correlation.

We report a case in which a sacrococcygeal teratoma was identified and characterized on prenatal and postnatal MRI, and correlated with subsequent surgical resection.