Xue Christine,  Nowrangi Rajeev,  Smith John,  Acharya Patricia

Final Pr. ID: Paper #: 021

To assess whether a convolutional neural network (CNN) can be trained via transfer learning to accurately diagnose metaphyseal corner fractures on long bone radiographs by comparing its performance to that of a board-certified pediatric radiologist. Secondary endpoints will include the ability of the CNN to detect the type of long bone presented on the radiograph as well as identify the chronicity of the fracture. Read More

Authors:  Xue Christine , Nowrangi Rajeev , Smith John , Acharya Patricia

Keywords:  AI, Nonaccidental Trauma, metaphyseal corner fractures

Mcluckey Morgan,  Karmazyn Boaz,  Marine Megan

Final Pr. ID: Poster #: EDU-095

The aim of this exhibit is to familiarize the radiologist with low dose chest CT technique, articulate advantages of CT in detection of fractures that are occult or indeterminate on skeletal surveys, as well as discuss the limitations of chest CT.

Radiologists play a central role in the detection of non-accidental trauma. After bruising and soft tissue trauma, fractures are the most common inflicted injuries seen in children. Rib fractures are one of the most common fractures identified on skeletal surveys and are one of the most specific injuries in child abuse. Chest CT has been shown to increase detection of rib fractures in child abuse and postmortem studies. In addition, new techniques such as the use of tin filter allow performance of very low dose chest CT. This has led some to advocate for increased utilization of chest CT, as head CT and abdominal CT have already well-established indications for evaluation of child abuse.

Chest CT can be considered in children with negative skeletal survey and high clinical suspicion for child abuse, when the diagnosis of rib fractures is indeterminate, and when the finding of additional or more specific fractures may aid in management, such as in children with both anterolateral fractures and recent cardiopulmonary resuscitation.

Chest CT has some limitations. Uncommonly CT can miss acute nondisplaced fractures demonstrated on radiographs due to lower spatial resolution. In addition, developmental variations in the costochondral junction (CCJ) can mimic healed fractures in CT.

We will provide the protocol we use for low dose chest CT technique and discuss the value of axial oblique and 3-D reconstructions in detection of rib fractures. We will show examples of fractures seen on chest CT which are occult or indeterminate on radiography, such as buckle fractures or fractures in challenging locations such as the CCJ. We will further discuss the strengths of CT in the evaluation of other thoracic injuries of abuse including fractures of the sternum, scapula, and vertebral bodies. Cases of fractures found on post-mortem imaging will be included. Finally, we will also show cases demonstrating the utility of CT in identifying false positives and normal variants that may mimic fractures on radiographs. Read More

Authors:  Mcluckey Morgan , Karmazyn Boaz , Marine Megan

Keywords:  Nonaccidental Trauma, Thoracic

Noda Sakura,  Otjen Jeffrey,  Koti Ajay,  Schlatter Adrienne,  Blessing Matthew,  Feldman Ken,  Menashe Sarah

Final Pr. ID: Poster #: EDU-013

Faulty fetal packing is a concave depression of the skull caused in utero by external pressure (such as by the maternal pelvic bone, a fetal limb, or uterine fibroid). It has an incidence of approximately 1 in 10,000 births. Its main differential diagnosis is acquired ping-pong-type fracture of the malleable neonatal skull, which can raise suspicion for nonaccidental trauma if there is no history of accidental trauma. Most recent literature on faulty fetal packing describes only single case reports. This educational exhibit reviews a series of faulty fetal packing and neonatal calvarial fracture cases. After reviewing this educational exhibit, readers will be able to 1) describe the mechanism of faulty fetal packing, 2) describe the appearance of faulty fetal packing on multiple modalities including radiographs and CT, and 3) distinguish between prenatal faulty fetal packing, perinatal skull trauma, and neonatal accidental and non-accidental trauma. Read More

Authors:  Noda Sakura , Otjen Jeffrey , Koti Ajay , Schlatter Adrienne , Blessing Matthew , Feldman Ken , Menashe Sarah

Keywords:  nonaccidental trauma, Skull, child abuse

Marine Megan,  Smith Jessica,  Wanner Matthew,  Jennings Samuel,  Hibbard Roberta,  Karmazyn Boaz

Final Pr. ID: Poster #: SCI-030

The purpose is to evaluate how well hospitals comply with ACR-SPR guidelines for skeletal surveys in the cases of possible child abuse. Read More

Authors:  Marine Megan , Smith Jessica , Wanner Matthew , Jennings Samuel , Hibbard Roberta , Karmazyn Boaz

Keywords:  Skeletal Survey, Child Abuse, Nonaccidental Trauma

Marine Megan,  Smith Jessica,  Wanner Matthew,  Jennings Samuel,  Hibbard Roberta,  Karmazyn Boaz

Final Pr. ID: Poster #: SCI-029

The purpose is to evaluate how well non-pediatric radiologists interpret skeletal surveys in the cases of possible child abuse. Read More

Authors:  Marine Megan , Smith Jessica , Wanner Matthew , Jennings Samuel , Hibbard Roberta , Karmazyn Boaz

Keywords:  Skeletal Survey, Child Abuse, Nonaccidental Trauma