Search

The rich collection of annotated datasets piloted the robustness of deep learning techniques to effectuate the implementation of diverse medical imaging tasks. Over 15% of deaths include children under age five are caused by pneumonia globally. In this study, we describe our deep learning based approach for the identification and localization of pneumonia in Chest X-rays (CXRs) images. Researchers usually employ CXRs for the diagnostic imaging study. Several factors such as positioning of the patient and depth of inspiration can change the appearance of the chest X-ray, complicating interpretation further. Our identification model ( https://github.com/amitkumarj441/identify_pneumonia ) is based on Mask-RCNN, a deep neural network which incorporates global and local features for pixel-wise segmentation. Our approach achieves robustness through critical modifications of the training process and a novel post-processing step which merges bounding boxes from multiple models. The proposed identification model achieves better performances evaluated on chest radiograph dataset which depict potential pneumonia causes.

Resumen La ecografía es un estudio de imágenes diagnósticas con una amplia historia de uso en todas las especialidades de la Medicina; su advenimiento ha ayudado al enfoque diagnóstico e, incluso, al abordaje terapéutico de los pacientes. Desde su origen en el siglo XIX con mediciones de la velocidad del sonido en el agua, hasta el desarrollo de las máquinas de ultrasonografía sustentadas en los avances de la tecnología, la física y la ingeniería, se ha utilizado de manera notable en la Medicina. No hay duda de que, en el campo de la Cirugía General, ha sido útil para el manejo de los pacientes con enfermedad abdominal. La tecnología ecográfica permite identificar las características de un órgano normal y, cuando este patrón se afecta, orienta sobre la causa o determina la enfermedad que puede estar produciendo la alteración. En la presente revisión, se hace un recuento histórico del nacimiento de la ecografía, su aplicación en el campo de la medicina y su utilidad para el cirujano general en diversas circunstancias de la práctica quirúrgica. Abstract Ultrasound is a diagnostic imaging study with a broad history of use in all medical specialties. Its advent has helped the medical approach as well as the therapeutic approach of patients. From its origin in the nineteenth century with the measurements of the speed of sound in the water, until the development of ultrasonic machines, sustained in the advances of technology in physics and engineering, has been used in a remarkable ways in medicine. There is no doubt that in the field of general surgery it has been useful for the management of patients with abdominal pathologies, mainly to identify the characteristics of a normal organ and when this pattern of normality is affected the ultrasound technology allows to identify which disease is in place. This review provides a historical review of how ultrasound was born, and how it began to be applied in the field of medicine and its utility for the general surgeon in various scenarios of surgical practice.

INTRODUCTION AND OBJECTIVES:A previous randomized trial found that obtaining ultrasound (US) as the first diagnostic imaging study for adults with suspected ureteral stones reduces radiation without increasing return ED visits or missed diagnoses. These results are consistent with EAU guidelines that US should be the first imaging modality for nephrolithiasis. We sought to determine the potential reduction in radiation exposure for an US-first approach using population-based diagnostic imaging patterns in the ED.METHODS:We performed a cross-sectional study of adults who presented to 70 emergency departments in South Carolina between 1996 and 2017 using all payer hospital claims data. All ED encounters within a 6 month “episode” following the initial ED visit were identified, along with ultrasound and CT ordered at the visits. Patterns of US and CT during the first two consecutive ER visits were determined at the patient- and hospital-level. Imaging patterns in which only a CT was performed were classified...

Gastrointestinal bleeding (GIB) occurs in up to 40% of patients with continuous-flow (CF) left ventricular assist devices (LVADs). We sought to identify targets to improve hospital resource utilization and decrease readmissions after GIB. We performed a single-center, retrospective analysis of LVAD-associated GIB resulting in hospital admission between July 2011 and April 2014. Follow-up data were collected through March 2015. We analyzed 57 admissions for GIB in 23 patients. One or more diagnostic imaging study was performed in 47% of admissions, with a definite or probable source of GIB identified in 23%. A total of 76 endoscopies were performed (≥ 1 endoscopy in 79% of admissions, ≥ 2 in 42%). Definite or probable bleeding sources were identified in 25% and 12% of endoscopies, respectively. Patients who underwent multiple endoscopies were no more likely to have a bleeding source identified (OR 1.48; 95% CI 0.50–4.32; p = 0.59) and had longer hospital stays (11.1 vs. 7.8 days, p

Diagnostic imaging has mirrored the steady growth of healthcare utilization in the USA. This has created greater opportunity for diagnostic errors, which can be costly in terms of morbidity and mortality as well as dollars and cents. The purposes of this study were to describe all return visits to a tertiary care urban pediatric emergency department (PED) resulting from diagnostic imaging discrepancies and to calculate the costs of these return visits. From July 2014 to February 2015, all children who underwent a diagnostic imaging study during an ED visit were assembled. Analysis was performed on all children who were called back and returned to the ED following a discrepant read. Direct and indirect costs to the patient, family, hospital, and society for these return visits were calculated. During the study period, 8310 diagnostic imaging studies were performed, with 207 (2.5%) discrepant reads. Among the discrepant reads, 37 (0.4% of total, 17.9% of discrepant) patients had a return visit to the ED for further management. Including ED charges, time and travel costs to the family, and costs of radiation exposure, return visits for radiologic discrepancies over this 8-month period cost a total of $84,686.47, averaging $2288.82 per patient. Though the overall diagnostic imaging discrepancy rate among our study population was low, the clinically significant discrepancies requiring return ED visits were potentially high risk, and costly for the patient, family, and healthcare system.

After-hours radiologic interpretation by nonradiology attendings or resident radiologists introduces the risk of discrepancies. Clinical outcomes following radiologic discrepancies among pediatric emergency department (ED) patients are poorly described. In particular, children with special healthcare needs (CSHCN), have more opportunities for discrepancies and potential consequences than non- CSHCN. Our objective was to determine the rates and types of radiologic discrepancies, and to compare CSHCN to non-CSHCN.From July 2014 to February 2015, all children who underwent a diagnostic imaging study at a free-standing children's ED were included. Data collected included radiologic studies - type and location - and clinical details - chief complaint and CSHCN type. Differences between preliminary reads and final pediatric radiology attending reads were defined as discrepancies, and categorized by clinical significance. Descriptive statistics, z-tests, and chi-square were used.Over 8months, 8310 visits (7462 unique patients) had radiologic studies (2620 CSHCN, 5690 non-CSHCN). A total of 198 (2.4%) radiologic discrepancies [56 (28.3%) CSHCN, 142 (71.7%) non-CSHCN] were found. Chief complaints for CSCHN were more often within the cardiac, pulmonary and neurologic systems (p0.001 for each), whereas non-CSHCN presented with more trauma (p0.001). The rates of discrepancies (CSHCN 2.1%, non- CSHCN 2.5%, p=0.3) and severity of clinical consequences (p=0.6) were not significantly different between CSHCN and non-CSHCN.Though the frequency and type of radiologic studies performed between CSHCN and non-CSHCN were different, we found no significant difference in the rate of radiologic discrepancies or the rate of clinically significant radiologic discrepancies.