Abusive head trauma (AHT) is an extreme form of physical child abuse, a subset of which is shaken baby syndrome (SBS). While traumatic injury in children is most readily observed as marks of contusion on the body, AHT/SBS may result in internal injuries that can put the life of the child in danger. One pivotal sign associated with AHT/SBS that cannot be spotted with the naked eye is retinal injury (RI), an early sign of which is retinal hemorrhage (RH) in cases with rupture of the retinal vasculature. If not addressed, RI can lead to irreversible outcomes, such as visual loss. It is widely assumed that the major cause of RI is acceleration-deceleration forces that are repeatedly imposed on the patient during abusive shaking. Still, due to the controversial nature of this type of injury, few investigations have ever sought to delve into its biomechanical and/or biochemical features using realistic models. As such, our knowledge regarding AHT-/SBS-induced RI is significantly lacking. In this mini-review, we aim to provide an up-to-date account of the traumatology of AHT-/SBS-induced RI, as well as its biomechanical and biochemical features, while focusing on some of the experimental models that have been developed in recent years for studying retinal hemorrhage in the context of AHT/SBS.

Retinal hemorrhages in pediatric patients can be a diagnostic challenge for ophthalmologists. These hemorrhages can occur due to various underlying etiologies, including abusive head trauma, accidental trauma, and medical conditions. Accurate identification of the etiology is crucial for appropriate management and legal considerations. In recent years, deep learning techniques have shown promise in assisting healthcare professionals in making more accurate and timely diagnosis of a variety of disorders. We explore the potential of deep learning approaches for differentiating etiologies of pediatric retinal hemorrhages. Our study, which spanned multiple centers, analyzed 898 images, resulting in a final dataset of 597 retinal hemorrhage fundus photos categorized into medical (49.9%) and trauma (50.1%) etiologies. Deep learning models, specifically those based on ResNet and transformer architectures, were applied; FastViT-SA12, a hybrid transformer model, achieved the highest accuracy (90.55%) and area under the receiver operating characteristic curve (AUC) of 90.55%, while ResNet18 secured the highest sensitivity value (96.77%) on an independent test dataset. The study highlighted areas for optimization in artificial intelligence (AI) models specifically for pediatric retinal hemorrhages. While AI proves valuable in diagnosing these hemorrhages, the expertise of medical professionals remains irreplaceable. Collaborative efforts between AI specialists and pediatric ophthalmologists are crucial to fully harness AIs potential in diagnosing etiologies of pediatric retinal hemorrhages.