Convolutional neural networks (CNNs) are a subtype of artificial neural network that have shown strong performance in computer vision tasks including image classification. To date, there has been ...limited application of CNNs to chest radiographs, the most frequently performed medical imaging study. We hypothesize CNNs can learn to classify frontal chest radiographs according to common findings from a sufficiently large data set.
Our institution's research ethics board approved a single-center retrospective review of 35,038 adult posterior-anterior chest radiographs and final reports performed between 2005 and 2015 (56% men, average age of 56, patient type: 24% inpatient, 39% outpatient, 37% emergency department) with a waiver for informed consent. The GoogLeNet CNN was trained using 3 graphics processing units to automatically classify radiographs as normal (n = 11,702) or into 1 or more of cardiomegaly (n = 9240), consolidation (n = 6788), pleural effusion (n = 7786), pulmonary edema (n = 1286), or pneumothorax (n = 1299). The network's performance was evaluated using receiver operating curve analysis on a test set of 2443 radiographs with the criterion standard being board-certified radiologist interpretation.
Using 256 × 256-pixel images as input, the network achieved an overall sensitivity and specificity of 91% with an area under the curve of 0.964 for classifying a study as normal (n = 1203). For the abnormal categories, the sensitivity, specificity, and area under the curve, respectively, were 91%, 91%, and 0.962 for pleural effusion (n = 782), 82%, 82%, and 0.868 for pulmonary edema (n = 356), 74%, 75%, and 0.850 for consolidation (n = 214), 81%, 80%, and 0.875 for cardiomegaly (n = 482), and 78%, 78%, and 0.861 for pneumothorax (n = 167).
Current deep CNN architectures can be trained with modest-sized medical data sets to achieve clinically useful performance at detecting and excluding common pathology on chest radiographs.
Artificial intelligence (AI) models for decision support have been developed for clinical settings such as radiology, but little work evaluates the potential impact of such systems. In this study, ...physicians received chest X-rays and diagnostic advice, some of which was inaccurate, and were asked to evaluate advice quality and make diagnoses. All advice was generated by human experts, but some was labeled as coming from an AI system. As a group, radiologists rated advice as lower quality when it appeared to come from an AI system; physicians with less task-expertise did not. Diagnostic accuracy was significantly worse when participants received inaccurate advice, regardless of the purported source. This work raises important considerations for how advice, AI and non-AI, should be deployed in clinical environments.
Medical datasets are often highly imbalanced with over-representation of prevalent conditions and poor representation of rare medical conditions. Due to privacy concerns, it is challenging to ...aggregate large datasets between health care institutions. We propose synthesizing pathology in medical images as a means to overcome these challenges. We implement a deep convolutional generative adversarial network (DCGAN) to create synthesized chest X-rays based upon a modest sized labeled dataset. We used a combination of real and synthesized images to train deep convolutional neural networks (DCNNs) to detect pathology across five classes of chest X-rays. The comparative study of DCNNs trained with the combination of real and synthesized images showed that these networks can outperform similar networks trained solely with real images in pathology classification. This improved performance is largely attributable to the balancing of the dataset using DCGAN synthesized images, where classes that are lacking in example images are preferentially augmented.
Abstract
Machine learning (ML) holds great promise in transforming healthcare. While published studies have shown the utility of ML models in interpreting medical imaging examinations, these are ...often evaluated under laboratory settings. The importance of real world evaluation is best illustrated by case studies that have documented successes and failures in the translation of these models into clinical environments. A key prerequisite for the clinical adoption of these technologies is demonstrating generalizable ML model performance under real world circumstances. The purpose of this study was to demonstrate that ML model generalizability is achievable in medical imaging with the detection of intracranial hemorrhage (ICH) on non-contrast computed tomography (CT) scans serving as the use case. An ML model was trained using 21,784 scans from the RSNA Intracranial Hemorrhage CT dataset while generalizability was evaluated using an external validation dataset obtained from our busy trauma and neurosurgical center. This real world external validation dataset consisted of every unenhanced head CT scan (n = 5965) performed in our emergency department in 2019 without exclusion. The model demonstrated an AUC of 98.4%, sensitivity of 98.8%, and specificity of 98.0%, on the test dataset. On external validation, the model demonstrated an AUC of 95.4%, sensitivity of 91.3%, and specificity of 94.1%. Evaluating the ML model using a real world external validation dataset that is temporally and geographically distinct from the training dataset indicates that ML generalizability is achievable in medical imaging applications.
The coronavirus disease 2019 (COVID-19) pandemic is a global health care emergency. Although reverse-transcription polymerase chain reaction testing is the reference standard method to identify ...patients with COVID-19 infection, chest radiography and CT play a vital role in the detection and management of these patients. Prediction models for COVID-19 imaging are rapidly being developed to support medical decision making. However, inadequate availability of a diverse annotated data set has limited the performance and generalizability of existing models. To address this unmet need, the RSNA and Society of Thoracic Radiology collaborated to develop the RSNA International COVID-19 Open Radiology Database (RICORD). This database is the first multi-institutional, multinational, expert-annotated COVID-19 imaging data set. It is made freely available to the machine learning community as a research and educational resource for COVID-19 chest imaging. Pixel-level volumetric segmentation with clinical annotations was performed by thoracic radiology subspecialists for all COVID-19-positive thoracic CT scans. The labeling schema was coordinated with other international consensus panels and COVID-19 data annotation efforts, the European Society of Medical Imaging Informatics, the American College of Radiology, and the American Association of Physicists in Medicine. Study-level COVID-19 classification labels for chest radiographs were annotated by three radiologists, with majority vote adjudication by board-certified radiologists. RICORD consists of 240 thoracic CT scans and 1000 chest radiographs contributed from four international sites. It is anticipated that RICORD will ideally lead to prediction models that can demonstrate sustained performance across populations and health care systems.
Artificial intelligence (AI)-generated clinical advice is becoming more prevalent in healthcare. However, the impact of AI-generated advice on physicians' decision-making is underexplored. In this ...study, physicians received X-rays with correct diagnostic advice and were asked to make a diagnosis, rate the advice's quality, and judge their own confidence. We manipulated whether the advice came with or without a visual annotation on the X-rays, and whether it was labeled as coming from an AI or a human radiologist. Overall, receiving annotated advice from an AI resulted in the highest diagnostic accuracy. Physicians rated the quality of AI advice higher than human advice. We did not find a strong effect of either manipulation on participants' confidence. The magnitude of the effects varied between task experts and non-task experts, with the latter benefiting considerably from correct explainable AI advice. These findings raise important considerations for the deployment of diagnostic advice in healthcare.
Due to a combination of increasing indications for MR imaging, increased MRI accessibility, and extensive global armed conflict over the last few decades, an increasing number of patients now and in ...the future will present with retained metallic ballistic debris of unknown composition. To date, there are no guidelines on how to safely image these patients which may result in patients who would benefit from MRI not receiving it. In this article, we review the current literature pertaining to the MRI safety of retained ballistic materials and present the process we use to safely image these patients.
Background. Aseptic abscesses (AA) are sterile lesions that represent an extraintestinal manifestation (EIM) of inflammatory bowel disease (IBD). Though Canada has the highest prevalence of IBD in ...the world, reports of IBD-associated AA are absent in Canada. This may represent a different IBD phenotype or underrecognition and underreporting. Purpose. To explore AA as a possible EIM of IBD and evaluate clinical and investigative findings among patients with IBD-associated AA. Methods. Retrospective chart and literature reviews were performed to find cases of IBD-associated AA at our institution and in the literature. Results. We identified 2 cases of IBD-associated AA in our institution. Both patients had ulcerative colitis and presented with fever, abdominal pain, and weight loss. Radiological workup and aspiration showed sterile splenic abscesses. The AA were unresponsive to antibiotics. One patient improved on corticosteroids and one underwent splenectomy. We retrieved 37 cases of IBD-associated AA from the literature. All patients showed no evidence of infection, failed to resolve with antibiotics, and, if attempted, improved on corticosteroids. Conclusions. Our cases are the first reported in Canada. They support literature which suggests AA as an EIM of IBD and may help increase recognition and reporting of this phenomenon.
Abstract Introduction Thrombocytosis is common following elective splenectomy and major trauma. However, little is known about the in-hospital course of platelet count (PC) and incidence of ...thrombocytosis after splenic trauma. Extreme thrombocytosis (PC > 1000 × 109 ) is associated with increased risk of venous thromboembolism (VTE) in primary thrombocytosis leading to the use of acetylsalicylic acid (ASA) for risk reduction, but the need for this agent in splenic trauma is undefined. Methods Retrospective cohort study of all patients with splenic trauma between April 1, 2010 and March 31, 2014. The in-hospital course of PC was assessed based on splenic injury management type. The association of management type with thrombocytosis was evaluated using a multivariable logistic regression model adjusting for potential confounders. The association of thrombocytosis, extreme thrombocytosis, and ASA use for the outcome of VTE was explored. Results 156 patients were eligible, PC initially increased in all patients with the highest peak after total splenectomy. The incidence of thrombocytosis was 41.0% (64/156). Thrombocytosis was more likely following splenectomy compared with spleen preserving strategies independent of length of stay, injury grade, ISS, age and transfusion (OR 7.58, 95% CI: 2.26–25.45). Splenectomy was associated with extreme thrombocytosis (OR 10.39, 95% CI: 3.59–30.07). Conclusions Thrombocytosis in splenic trauma is more likely after splenectomy than with spleen preserving strategies. Splenectomy is associated with extreme thrombocytosis. There was insufficient data in our study to determine the use of ASA as primary prevention of VTE after splenic trauma.