Imaging of Neonatal Child Abuse with an Emphasis on Abusive Head Trauma

The diagnosis and management of abusive head injuries in infancy in Egypt

Child death owed to abuse and negligence is not uncommon, and its real incidence is unknown. The most common cause of fatal child abuse is head trauma. Abusive head injuries (AHI) most often involve brain injury of infants and young children. The outcomes of AHI vary from complete recovery to severe brain damage and death. This article highlights the diagnosis and management of AHI in infancy in Egypt, with a special focus on the social, medical, and legal aspects. The authors emphasize the importance of reporting cases suspected of AHI to the relevant authorities; this will guard against the recurrence of abuse to the child and will have a positive impact on the community.

Using deep learning and natural language processing models to detect child physical abuse

BACKGROUND

The recognition of child physical abuse can be challenging and often requires a multidisciplinary assessment. Deep learning models, based on clinical characteristics, laboratory studies, and imaging findings, were developed to facilitate unbiased identification of children who may have been abused.

METHODS

Level 1 pediatric trauma center registry data from 1/1/2010-1/31/2020 were queried for abused children and matched participants with non-abusive trauma. Observations were de-identified and divided into training and validation sets. Model 1 used patient demographics (age, gender, and insurance type) and clinical characteristics (vital signs, shock index pediatric age-adjusted, Glasgow Coma Score, lactate, base deficit, and international normalized ratio). Model 2 used the same features as Model 1, but with the text of the radiology reports of head computed tomography, brain MRIs, and skeletal surveys. Google's latest BERT Natural Language Processing (NLP) model, which was pre-trained on a large corpus, was used for fine-tuning Model 2. Accuracy, sensitivity, specificity, F1 scores, and positive predictive values were used to assess performance.

RESULTS

Of 1,312 patients, 737 (56.2%) were abused. Model 1 had an accuracy of 86.3%, sensitivity of 87.2%, specificity of 85.1%, F1 score of 0.86, and positive predictive value (PPV) of 88.7% for the validation set with an area under the receiver Operating Curve (ROC AUC) of 0.86. NLP based Model 2 had an accuracy of 93.4%, sensitivity 92.5%, specificity of 94.6%, F1 score of 0.93, and PPV of 95.9% for the validation set, with a ROC AUC of 0.94. Most features had weak individual correlations with abuse (r < 0.3).

CONCLUSIONS

Deep learning models accurately distinguished child physical abuse from non-abuse, and NLP further improved the accuracy of the models. Such models could be developed to run in real-time in the electronic medical record and alert clinicians when certain criteria are met, which would prompt them to pursue the diagnosis of abuse.

LEVEL OF EVIDENCE

III STUDY TYPE: Diagnostic.

Virtopsy and Living Individuals Evaluation Using Computed Tomography in Forensic Diagnostic Imaging

The applications of forensic radiology involve both Virtopsy both studies on living people - to demonstrate bone age, search for foreign bodies, such as voluntary injection of drug ovules or surgical sponges accidentally forgotten, to assess gunshot wounds, to evaluate injuries by road accidents, and cases of violence or abuse (both in adults and in children). Computed tomography is the most used imaging tool used in forensic pathology and its indications are mainly focused on cases of unnatural deaths or when a crime is suspected. It is preferred over the standard autopsy in selected cases, such as in putrefied, carbonized or badly damaged bodies; or as a preliminary evaluation in mass disasters.

Mimics of child abuse: Can choking explain abusive head trauma?

Choking is one of the alternative explanations of abusive head trauma in children that have been offered in courtroom testimony and in the media. Most of these explanations - including choking - are not scientifically supported. This article highlights four points. (1) The origins of choking as an explanation for intracranial and retinal hemorrhages are speculative. (2) Choking has been used in high profile court testimony as an explanation for the death of a child thought to have been abused. (3) A case report that proposes choking as an alternative explanation for the death of a child diagnosed with abusive head trauma includes omissions and misrepresentations of facts. (4) There was a decision by the editor of the journal that published the case report that it was not necessary to include all the facts of the case; moreover, the editor indicated that facts are not required when presenting an alternative explanation. The use of scientifically unsupported alternative explanations for abusive head trauma based on inaccurate and biased information constitutes further victimization of the abused child and represents a travesty of justice.