Rostad Bradley,  Richer Edward,  Riedesel Erica,  Alazraki Adina

Final Pr. ID: Paper #: 026

Foreign body ingestion is common in children. Timely diagnosis of the nature and location of the foreign body is important. A button battery which lodges in the esophagus can quickly cause severe esophageal and mediastinal injury. Machine learning that can detect anatomical regions of interest is an important step in computerized foreign body localization and may result in prioritization of radiographs with mediastinal foreign bodies. The purpose of this study is to develop a machine learning model to identify anatomical regions of interest on pediatric foreign body series radiographs. Read More

Authors:  Rostad Bradley , Richer Edward , Riedesel Erica , Alazraki Adina

Keywords:  Machine learning, Artificial intelligence, Foreign body

Rostad Bradley,  Richer Edward,  Riedesel Erica,  Alazraki Adina

Final Pr. ID: Paper #: 025

Timely diagnosis of foreign body ingestion in children is important, particularly in the case of an ingested button battery. A button battery which lodges in the esophagus can quickly cause severe esophageal and mediastinal injury. It is also important to distinguish an ingested button battery from a coin; a button battery may be misdiagnosed a coin because of its similar radiographic appearance. The purpose of this study is to develop a machine learning model to identify button batteries and coins on pediatric foreign body series radiographs. Read More

Authors:  Rostad Bradley , Richer Edward , Riedesel Erica , Alazraki Adina

Keywords:  Machine learning, Artificial intelligence, Button battery

Stanley Parker,  Stanley Charles

Final Pr. ID: Poster #: EDU-010 (T)

In 2017, roughly 2 trillion (2,000,000,000,000) medical images were produced, reviewed, reported, archived, and used in the detection and management of disease. Based on historical trends, this number has doubled every 5 years and is accelerating. This explosive growth in imaging data has created major opportunities for the use of Artificial Intelligence (AI). The question is less whether radiologists, and technologists, will be replaced by AI (they will not) and more about whether we could survive without AI. Although intelligent algorithms have been used for some time in segments of the imaging field, new methods of machine learning, based particularly on “deep learning”, are much more powerful. Many of the deep learning publications today point to the promise of significant advances in efficiency, precision, reproducibility, and prognostic abilities.
If AI will not replace radiologists/technologists but rather augment them with tools to meet the rising demands for diagnostic imaging, then it is imperative that we have a basic understanding of the concepts and language that defines this area of knowledge. In the not so distant past the average technologist understood the basics of film processing but wouldn’t even recognize the words DICOM or EMR; we are now at that point of change with AI. Deep learning, machine learning, neural networks, ground truth, the list goes on. The goal of this presentation is to provide a basic framework of the concepts, terminology, and references to how AI has, and likely, will be employed in medical imaging, thus making us better practitioners and partners with this technology.
Read More

Authors:  Stanley Parker , Stanley Charles

Keywords:  Artificial Intelligence, Medical Imaging, Technologist

Meda Karthik,  Milla Sarah,  Rostad Bradley

Final Pr. ID: Paper #: 024

Machine learning that can identify and localize objects in an image using a labeled bounding box is called object detection. The purpose of this study is to demonstrate object detection in identifying rickets on pediatric wrist radiographs. Read More

Authors:  Meda Karthik , Milla Sarah , Rostad Bradley

Keywords:  Machine learning, Artificial intelligence, rickets

Tsang Brian,  Gupta Aaryan,  Takahashi Marcelo,  Ola Tolulope,  Baffi Henrique,  Doria Andrea

Final Pr. ID: Poster #: SCI-031

1) To systematically review the uses of AI for magnetic resonance (MR) imaging assessment of primary pediatric cancer and identify common literature topics and knowledge gaps. 2) To assess the adherence of the existing literature to the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) guidelines. Read More

Authors:  Tsang Brian , Gupta Aaryan , Takahashi Marcelo , Ola Tolulope , Baffi Henrique , Doria Andrea

Keywords:  Artificial intelligence, Pediatric cancer, Magnetic resonance imaging

Sarrami Amir Hossein,  Wang Hongzhi,  Baratto Lucia,  Syeda-mahmood Tanveer,  Daldrup-link Heike

Final Pr. ID: Poster #: EDU-078

Medical Imaging has a crucial role in the diagnosis and management of pediatric cancer patients by providing information about tumor location and quantitative measures of tumor size and metabolic activity at baseline as well as during and after therapy. The standard imaging plan for staging and re-staging of pediatric malignancies includes a high-resolution MRI or CT scan of the local tumor and whole body staging for the detection of metastases on CT, MRI and/or PET scans. Children with lymphomas, sarcomas, germ cell tumors and a few other tumor types are referred to whole body 18F-FDG PET scanning, either coupled with CT or MRI. Artificial intelligence (AI) algorithms can facilitate staging and re-staging of cancers in children by providing 1) rapid detection and delineation of tumoral lesions, 2) automated measurements of tumoral size and metabolic activity, 3) relating tumor measurements to internal standard such as liver and blood pool, 4) assigning a score according to tumor-specific staging systems.

Detecting tumors on a whole-body scan is a challenging task, especially in children whose organs undergo changes in size and composition with increasing age. Moreover, the tumors in children can arise from almost anywhere in the body, from head to toe. With successful applications on object detection, AI methods are promising for automatic tumor detection from whole-body scans as well. In this tutorial, we will introduce a few popular AI methods for such purpose. These methods include U-Net, Vision Transformers (ViT), and the hybrid of the above methods such as O-Net Transformer or TransUNet. Read More

Authors:  Sarrami Amir Hossein , Wang Hongzhi , Baratto Lucia , Syeda-mahmood Tanveer , Daldrup-link Heike

Keywords:  Hybrid Imaging, Artificial Intelligence, Cancer

Li Jason,  Betke Margaret,  Gill Christopher,  Thompson Russell,  Wang Kaihong,  Etter Lauren,  Camelo Ingrid,  Chang Hailey,  Setty Bindu,  Castro Ilse,  Pieciak Rachel

Final Pr. ID: Poster #: SCI-015

Point of Care Lung ultrasound has proven in multiple studies to be superior to CXR to diagnose pneumonia in children especially in limited resource settings. This non-radiating, portable and adaptable technique, brings an opportunity to detect pneumonia with higher accuracy than CXR. Ultrasound imaging interpretation is challenging. To deal with this complexity, we created a "brightness profiles" data reduction technique to identify specific anatomical structures identified by lung ultrasound using artificial intelligence. We use this technique to demonstrate how data reduction can help identify common anatomical landmarks and abnormal findings, and aid in the interpretation of ultrasound diagnosed pediatric pneumonia. Read More

Authors:  Li Jason , Betke Margaret , Gill Christopher , Thompson Russell , Wang Kaihong , Etter Lauren , Camelo Ingrid , Chang Hailey , Setty Bindu , Castro Ilse , Pieciak Rachel

Keywords:  Artificial intelligence, pneumonia, ultrasound

Cho Yeon Jin,  Park Hyoung Suk,  Jeon Kiwan,  Choi Young Hun,  Choi Gayoung,  Lee Seul Bi,  Lee Seunghyun,  Cheon Jung-eun,  Kim Woo Sun,  Ryu Young Jin,  Hwang Jae-yeon

Final Pr. ID: Paper #: 060

The purpose of this study was to develop a convolutional neural network (CNN)-based deep learning algorithm for the automated detection of developmental dysplasia of the hip (DDH) on conventional radiography and to assess its feasibility and diagnostic performance. Read More

Authors:  Cho Yeon Jin , Park Hyoung Suk , Jeon Kiwan , Choi Young Hun , Choi Gayoung , Lee Seul Bi , Lee Seunghyun , Cheon Jung-eun , Kim Woo Sun , Ryu Young Jin , Hwang Jae-yeon

Keywords:  Artificial intelligence, Developmental dysplasia of the hip, Conventional radiography

Alkhulaifat Dana,  Rafful Patricia,  Lopez Rippe Julian,  Khalkhali Vahid,  Welsh Michael,  Wieczkowski Sydney,  Reid Janet,  Sotardi Susan

Final Pr. ID: Poster #: EDU-047

Artificial intelligence (AI) applications for radiology have undergone exponential growth in recent years, owing to the development of large datasets for use in machine learning algorithms and technological advancements in the field of imaging informatics. However, the advancement of AI algorithms in pediatric radiology has lagged behind adult applications. Currently, only seven commercially-available AI algorithms have received FDA approval for use in the pediatric population [1]. One of the major factors limiting the use of AI in pediatric radiology is the lack of the requisite large pediatric imaging datasets.
In AI research and implementation, pediatric radiologists serve as stewards of imaging data. As such, pediatric radiologists should be trained in AI data management, including best practices for the selection, curation, de-identification, and storage of radiology data. Since a necessary first step in the development of AI algorithms requires the curation of large datasets, pediatric radiologists should have a basic understanding of how to archive imaging data for AI research and validation. However, few resources are currently available to provide targeted education for pediatric radiologists with respect to AI data curation..
The aim of this educational exhibit is to provide an educational resource specifically for pediatric radiologists which teaches best practices for data management in AI research, including the selection of patient cohorts, data anonymization techniques, image annotation and segmentation methods, and data storage tools. This exhibit integrates our professional experience, with a thorough literature review of prior AI research, into an educational resource to teach data science methodologies for the management of AI research and clinical implementation to the pediatric radiology community.

References
1. AI Central. https://aicentral.acrdsi.org/. Accessed 18 Oct 2022 Read More

Authors:  Alkhulaifat Dana , Rafful Patricia , Lopez Rippe Julian , Khalkhali Vahid , Welsh Michael , Wieczkowski Sydney , Reid Janet , Sotardi Susan

Keywords:  Artificial Intelligence, Education, Data Management

Bedoya M. Alejandra,  Iwasaka-neder Jade,  Bixby Sarah,  Tsai Andy

Final Pr. ID: Poster #: SCI-005

Knowing the gestational status of a neonate (premature or full-term) impacts a pediatric radiologist’s ability to generate a reasonable differential diagnosis of neonatal diseases. Though crucial, this information is often unavailable at the time of the babygram interpretation. Conventional methods based on measuring the clavicular-pubic length (CPL) and the transverse diameter of the chest (TDC) from a babygram have been proposed as reliable estimators of a neonate’s gestational age. In this study, we aimed to compare these two conventional methodologies to that of a deep-learning (DL) model in determining a neonate’s gestational status. Read More

Authors:  Bedoya M. Alejandra , Iwasaka-neder Jade , Bixby Sarah , Tsai Andy

Keywords:  Artificial intelligence, Clavicular-pubic length, Chest width

Knight Jessica,  Jaremko Jacob,  Zhou Yuyue,  Keen Christopher,  Rakkundedeth Abhilash,  Ghasseminia Siyavash,  Wichuk Stephanie,  Brilz Alan,  Alves Pereira Fatima,  Kirschner David

Final Pr. ID: Poster #: SCI-021

Wrist trauma is common in children and generally requires radiography for diagnosis. Many children who receive radiographs do not have fractures and are thus subjected to unnecessary radiation exposure along with increased wait times in the emergency department (ED). Ultrasound (US) is safe, cost-effective, portable and sensitive in visualizing cortical disruption, potentially making it a valuable tool for bedside assessment of fractures. This study aims to determine the feasibility of using US to detect distal radius fractures (DRF) in children and to contrast the accuracy of hand-held device compatible 2D transducers to 3D transducers that can only be used with traditional US machines. In order to address difficulties in US image interpretation by inexperienced users, we investigated the utility of an artificial intelligence (AI) network. Read More

Authors:  Knight Jessica , Jaremko Jacob , Zhou Yuyue , Keen Christopher , Rakkundedeth Abhilash , Ghasseminia Siyavash , Wichuk Stephanie , Brilz Alan , Alves Pereira Fatima , Kirschner David

Keywords:  Artificial Intelligence, Ultrasound, Musculoskeletal

Cheng Jocelyn,  Leesmidt Kantheera,  Durand Rachelle,  Mongan John,  Cort Kayla,  Courtier Jesse

Final Pr. ID: Poster #: EDU-001

In recent years, the applications of artificial intelligence in radiology have increased exponentially. Randomized controlled trials demonstrating the potential of AI in radiology have been centered around deep learning techniques with artificial neural networks, such as convolutional neural networks (CNNs). Current applications of CNNs include the detection, segmentation and quantification of pathologic conditions. These tasks use deductive AI, which uses discriminative models to identify discrepancies between images. Generative AI is an emerging technique that has a number of exciting novel applications in the field of Pediatric Radiology. This exhibit will provide a practical overview to familiarize pediatric radiologists with this topic and review current and potential future use cases.

Generative adversarial networks (GAN) require two opposing networks, a generator that synthesizes new images, and a discriminator that differentiates between real and synthesized images. Through an iterative process, GAN creates highly realistic images that have exciting applications in image creation, translation between imaging modalities, acceleration of image synthesis and identification of pathologic abnormalities.

ChatGPT is a popular generative AI model that generates coherent responses to queries, and uses a Generative Pre-trained Transformer (GPT) network to encode and decode language data. ChatGPT is pre-trained on large volume text data to learn how language is used in different contexts. An encoder processes the input sequence and a decoder generates the output sequence. Both encoder and decoder have a multi-head self-attention mechanism allows differential weighing of the sequence to infer meaning and context. Potential applications in pediatric radiology are broad, including patient communication, providing appropriate imaging recommendations, report generation and review of radiologic safety guidelines and regulations. However, generative AI models require careful implementation, training and validation to reduce potential pitfalls in unintentionally synthesizing content that does not exist, or conversely, removing clinically relevant information from an image.

The purpose of this exhibit is to: (1) provide an introduction to Generative AI (GAN) and ChatGPT (2) review use cases of GAN and ChatGPT in pediatric radiology (3) provide practical methods for evaluating the performance of artificial intelligence algorithms in pediatric radiology applications. Read More

Authors:  Cheng Jocelyn , Leesmidt Kantheera , Durand Rachelle , Mongan John , Cort Kayla , Courtier Jesse

Keywords:  Generative Artificial Intelligence, Artificial Intelligence, ChatGPT

Rafful Patricia,  Alkhulaifat Dana,  Lopez Rippe Julian,  Khalkhali Vahid,  Welsh Michael,  Venkatakrishna Shyam Sunder,  Wieczkowski Sydney,  Reid Janet,  Sotardi Susan

Final Pr. ID: Poster #: EDU-044

Artificial intelligence (AI) has the potential to improve many aspects of pediatric radiology. AI solutions have emerged to perform a wide range of tasks, including facilitating the workload of radiologists, improving image quality, performing motion correction, and reducing contrast dose. While a myriad of courses for learning AI are available, most resources lack the necessary radiology perspective required for clinical application. Few interactive resources are available to teach a systematic approach for pediatric radiologists when analyzing AI literature. Since AI methodologies in pediatric radiology research fundamentally differ from classical statistical analyses, an educational poster dedicated to interpreting pediatric radiology AI literature would assist in bridging this gap.

The purpose of this education exhibit is to provide a resource for pediatric radiologists that teaches a systematic approach for the interpretation of AI research publications. Knowledge delivery will be made by eLearning slide-based presentation. The major topics regarding AI literature will focus on: problem identification; data curation; data annotation, artificial intelligence models; validation and performance. Next, we review prior literature on important criteria for AI publications, including Checklist for Artificial Intelligence in Medical Imaging (CLAIM), the AI version of the Standards for Reporting of Diagnostic Accuracy Studies (STARD-AI), Transparent Report of Multivariable Prediction Model of Individual Prognosis and Diagnosis for AI (TRIPOD-AI) and Prediction Model Risk of Bias Assessment Tool for AI (PROBAST-AI). Lastly, we present these criteria in the context of pediatric AI research applications.
AI research methodology in pediatric radiology differs from classical research methodologies. Therefore, this education exhibit is intended to teach the systematic interpretation of the primary AI literature to pediatric radiologists in order to translate scientific knowledge into meaningful clinical information. Read More

Authors:  Rafful Patricia , Alkhulaifat Dana , Lopez Rippe Julian , Khalkhali Vahid , Welsh Michael , Venkatakrishna Shyam Sunder , Wieczkowski Sydney , Reid Janet , Sotardi Susan

Keywords:  Artificial Intelligence, Radiology, Research

Randhawa Hari,  Park Brandon,  Khanna Praneet,  Jung Daniel,  Das Ayushman,  Ginn Kevin,  Mitchell Grace

Final Pr. ID: Poster #: EDU-003

MRI is crucial to the diagnosis, treatment planning, and monitoring of pediatric posterior fossa tumors; recent advances in machine and deep learning offer the prospect of increased diagnostic accuracy and minimally invasive management strategies. Machine learning and deep learning represent two artificial intelligence (AI) applications of radiomics that show promising results in the identification and classification of pediatric posterior fossa tumors, such as medulloblastoma, diffuse intrinsic pontine glioma, ependymoma, and pilocytic astrocytoma, based on MRI data. Many AI models demonstrate comparable, or in some cases, better, performance than human radiologists. Beyond diagnosis, machine and deep learning algorithms have been developed for tumor histology determination, prognostication, and molecular subtyping, to name just a few examples. In its early stages, however, the technology has several barriers to overcome before mainstream use. Studies of AI algorithms are often conducted with small sample sizes due to the relative rarity of pediatric posterior fossa tumors; many are also conducted at a single institution. Algorithms are susceptible to class imbalance as well, making it more difficult to identify tumors with lower prevalence. Multi-institution studies are important in addressing those limitations, though they have their own cumbersome requirements, such as data harmonization. Deep learning training strategies, such as federated learning, data augmentation, and transfer learning, help to address the problems introduced by working with small data sets. The purpose of this exhibit is to review the methods, results, and limitations identified in studies testing the efficacy of machine and deep learning models for the identification and classification of pediatric posterior fossa tumors and to discuss their potential for use in the clinical setting. Read More

Authors:  Randhawa Hari , Park Brandon , Khanna Praneet , Jung Daniel , Das Ayushman , Ginn Kevin , Mitchell Grace

Keywords:  Artificial Intelligence, Deep Learning, Machine Learning

Thompson Russell,  Pieciak Rachel,  Gill Christopher,  Li Jason,  Wang Kaihong,  Etter Lauren,  Camelo Ingrid,  Castro-aragon Ilse,  Setty Bindu,  Chang Hailey,  Betke Margaret

Final Pr. ID: Poster #: SCI-008

CXR is the most common imaging method to diagnose pneumonia in children in limited-resource settings. There is a need to simplify and expedite its interpretation. By using a machine learning model to first classify and interpret the pneumonia images and then incorporate those characteristic imaging findings patterns into a simulated mobile app, health care workers can use their mobile devices to interpret those findings based on preloaded images built into their mobile devices corresponding to pneumonia. Read More

Authors:  Thompson Russell , Pieciak Rachel , Gill Christopher , Li Jason , Wang Kaihong , Etter Lauren , Camelo Ingrid , Castro-aragon Ilse , Setty Bindu , Chang Hailey , Betke Margaret

Keywords:  Artificial intelligence, pneumonia, CXR

Alkhulaifat Dana,  Rafful Patricia,  Lopez Rippe Julian,  Khalkhali Vahid,  Welsh Michael,  Wieczkowski Sydney,  Reid Janet,  Sotardi Susan

Final Pr. ID: Poster #: EDU-043

In recent years, artificial intelligence (AI) applications for radiology have undergone exponential growth. AI has the potential to assist pediatric radiologists in several respects, including identification of various pathologies and prioritization of studies with urgent imaging findings. Although image analysis is an important application of AI, there are many additional use cases where AI can assist radiologists, including radiology operations and quality analysis. Therefore, it is critical that radiologists familiarize themselves with the wide range of AI methodologies and applications.
Although efforts have been made to create formal education for radiologists, currently available resources typically rely upon didactic lectures, without real-life context for the technology. However, case-based education is a principal component of radiology education. Furthermore, prior research has demonstrated that trainees and radiologists are interested in case-based approaches for learning AI [1].

Thus, the purpose of this educational exhibit is to provide an informative resource specifically for pediatric radiologists which utilizes a case-based learning approach. This resource uses 5 cases intended to demonstrate the variety of uses of AI in pediatric radiology, while illustrating the underlying process of AI model selection and development. Specifically, these cases will be selected to encompass AI topics including: image classification, lesion segmentation, natural language processing for imaging reports, recommender systems to enhance radiologist practice and predictive modeling for radiology operations research. This approach allows us to explore decision-points in AI model building, while examining potential pitfalls that may arise. This educational exhibit was developed based on a thorough literature review of prior AI research combined with our professional experience, and aims to provide educational resource intended to teach AI model selection and development to pediatric radiologists.

References
1. Velez-Florez MC, Ghosh A, Patton D, et al (2022) Artificial intelligence curriculum needs assessment for a pediatric radiology fellowship program: what, how, and why? Acad Radiol. https://doi.org/10.1016/j.acra.2022.04.026 Read More

Authors:  Alkhulaifat Dana , Rafful Patricia , Lopez Rippe Julian , Khalkhali Vahid , Welsh Michael , Wieczkowski Sydney , Reid Janet , Sotardi Susan

Keywords:  Artificial Intelligence, Case-Based Learning, Education