Automated detection of enteric tubes misplaced in the respiratory tract on chest radiographs using deep learning with two centre validation.

AIM: To develop and test a model based on a convolutional neural network that can identify enteric tube position accurately on chest radiography. MATERIALS AND METHODS: The chest radiographs of adult patients were classified by radiologists based on enteric tube position as either critically misplaced (within the respiratory tract) or not critically misplaced (misplaced within the oesophagus or safely positioned below the diaphragm). A deep-learning model based on the 121-layer DenseNet architecture was developed using a training and validation set of 4,693 chest radiographs. The model was evaluated on an external test data set from a separate institution that consisted of 1,514 consecutive radiographs with a real-world incidence of critically misplaced enteric tubes. RESULTS: The receiver operator characteristic area under the curve was 0.90 and 0.92 for the internal validation and external test data sets, respectively. For the external data set with a prevalence of 4.4% of critically misplaced enteric tubes, the model achieved high accuracy (92%), sensitivity (80%), and specificity (92%) for identifying a critically misplaced enteric tube. The negative predictive value (99%) was higher than the positive predictive value (32%). CONCLUSION: The present study describes the development and external testing of a model that accurately identifies an enteric tube misplaced within the respiratory tract. This model could help reduce the risk of the catastrophic consequences of feeding through a misplaced enteric tube.

Augmented reality simulator for CT-guided interventions.

INTRODUCTION: CT-guided interventions are taught using a mentored approach on real patients. It is well established that simulation is a valuable training tool in medicine. This project assessed the feasibility and acceptance of replicating a CT-guided intervention using a bespoke software application with an augmented reality head-mounted display (ARHMD). METHODS: A virtual patient was generated using a CT dataset obtained from The Cancer Imaging Archive. A surface mesh of a virtual patient was projected into the field-of-view of the operator. ChArUco markers, placed on both the needle and agar jelly phantom, were tracked using RGB cameras built into the ARHMD. A virtual CT slice simulating the needle position was generated on voice command. The application was trialled by senior interventional radiologists and trainee radiologists with a structured questionnaire evaluating face validity and technical aspects. RESULTS: Sixteen users trialled the application and feedback was received from all. Eleven felt the accuracy and realism was adequate for training and twelve felt more confident about their CT biopsy skills after this training session. DISCUSSION: The study showed the feasibility of simulating a CT-guided procedure with augmented reality and that this could be used as a training tool. KEY POINTS: • Simulating a CT-guided procedure using augmented reality is possible. • The simulator developed could be an effective training tool for clinical practical skills. • Complexity of cases can be tailored to address the training level demands.

Remote reporting in the COVID-19 era: from pilot study to practice.

AIM: To assess the benefits and challenges of remote reporting using an intra-departmental teleradiology system. MATERIALS AND METHODS: A pilot of an in-hospital Trust radiologist reporting on in-hospital Trust patients via a remote login was undertaken. Reporting output, training impact, and quality improvement were measured. RESULTS: Reporting output increased by 140%. Trainee satisfaction was high in a qualitative survey, particularly for out-of-hours support and teaching. Clinicians found the service to be similar to the same service provided by a locally based radiologist. CONCLUSION: In the COVID-19 era, remote working has developed rapidly. This study shows that radiology departments can provide remote reporting that is equal in standard to reporting from within the hospital, and in addition, that there are advantages to output and training.

Assessment of the young adult hip joint using plain radiographs.

Radiographic examination remains the mainstay of the initial assessment of the young adult hip; however, common parameters are required to assist in the formation of accurate diagnoses and appropriate management plans. This paper aims to summarise the most important aspects of the assessment of plain radiographs performed on the young adult hip joint.

Physes around the shoulder girdle: normal development and injury patterns.

Traumatic injuries involving the scapula and clavicle in skeletally immature patients have unique characteristics that distinguish them from similar injuries in the mature skeleton. Fractures involving unossified cartilage and unfused epiphyses are difficult to appreciate on plain radiographs and computed tomography (CT) imaging. Knowledge of the developmental anatomy and normal radiological appearances during different stages of development of these bones is an essential prerequisite for the radiologist tasked with interpreting the imaging of such injuries in order to avoid potential diagnostic pitfalls. With increased availability and improved resolution of magnetic resonance imaging (MRI), we are now better able to distinguish between true joint dislocations and epiphyseal injuries. Making this distinction is important because it can have implications with regards to how the patient is managed and the prognosis.