Child abuse is the principal cause of traumatic injury and death in children 0-36 months old. Radiologic findings are a critically important piece of evidence necessary for assessment by a Child Protection Team to assess possible abuse. Current algorithmic strategies for the detection of abuse are sensitive, but compromise specificity. The purpose of this study was to determine the performance of deep learning to diagnose child abuse using Natural Language Processing of the unstructured free-text from Electronic Health Records (EHR), including essential information from Radiology Reports. Read More

Meeting name: SPR 2020 Annual Meeting & Postgraduate Course , 2020

Authors: Annapragada Akshaya, Donaruma-kwoh Marcella, Annapragada Ananth, Starosolski Zbigniew

Keywords: Child Abuse, decision support, Convolutional Neural Network

The incidence of head and neck malignancies among children has increased substantially in the last five decades. The immunosuppressive tumor micro-environment (TME) represents a major roadblock in the treatment of head and neck tumors. We have demonstrated previously that an elevated level of myeloid derived suppressor cells (MDSC) in the tumor correlates directly with a poor prognosis, and that MDSC levels can be reduced using inhibitors of inducible nitric oxide synthase (iNOS). In this pre-clinical study, we utilized non-contrast magnetic resonance imaging (MRI) and nanoparticle contrast-enhanced computed tomography (CECT) to monitor tumor responses to TME-directed combinatorial therapies that included iNOS inhibition, immunomodulation, checkpoint inhibition, and radiation. Read More

Meeting name: SPR 2020 Annual Meeting & Postgraduate Course , 2020

Authors: Devkota Laxman, Bhavane Rohan, Badachhape Andrew, Veeramachaneni Ratna, Stupin Igor, Bhandari Prajwal, Ghaghada Ketan, Sikora Andrew, Annapragada Ananth

Keywords: Immunotherapy, Tumor, Imaging

Neuroblastoma (NB) is the most common non-cranial solid tumor in childhood. The MYCN oncogene plays a crucial role in tumorigenesis and angiogenesis in NB and is a strong indicator of poor prognosis. The tyrosine hydroxylase (TH)-MYCN transgenic mouse model of NB is extensively utilized; however, little is known about disease progression in this model. In this work, we use multi-modal imaging to study tumor progression and vascular architecture in TH-MYCN transgenic, allograft, and syngeneic mouse models of NB. Read More

Meeting name: SPR 2020 Annual Meeting & Postgraduate Course , 2020

Authors: Badachhape Andrew, Tao Ling, Joshi Sanshuv, Devkota Laxman, Stupin Igor, Sarkar Poonam, Annapragada Ananth, Barbieri Eveline, Ghaghada Ketan

Keywords: Neuroblastoma, Magnetic Resonance Imaging, Tumor