Reason for exam Imaging Reporting and Data System (RI-RADS): A grading system to standardize radiology requisitions

Classification of Musculoskeletal Radiograph Requisition Appropriateness Using Machine Learning

Objective: Poor quality imaging requisitions lower report quality and impede good patient care. Manual control of such requisitions is time consuming and can be a source of friction with referring physicians. The purpose of this study was to determine if poor quality requisitions could be identified automatically using machine learning and natural language processing techniques in order to allow for more efficient workflow. Methods: Exam indications from 50 000 musculoskeletal radiograph requisitions were manually classified, reviewed and deemed 'appropriate' or 'inappropriate' by two staff radiologists based on ACR appropriateness criteria. The requisitions were divided into training and test groups (80/20 split). The training set was pre-processed, converted to a bag-of-words model and used to train a Multinomial Naïve Bayes classifier which was then applied to the test set. Results: Out of 50 000 requisitions, 12 253 (24.5%) were deemed to contain an inappropriate indication. A Naive Bayes model correctly classified requisitions with an accuracy of 98%. In the test set, 107 of 7561 (1.4%) appropriate requisitions were incorrectly flagged and 92 of 2439 (3.8%) inappropriate requisitions were not flagged. Conclusions: Accurate automated identification of inappropriate indications on musculoskeletal requisitions is feasible using machine learning and natural language processing.

IMPACT (Information Medically Pertinent in Acute Computed Tomography) requests: Delphi study to develop criteria standards for adequate clinical information in computed tomography requests in the Australian emergency department

INTRODUCTION

Inadequate clinical information in medical imaging requests negatively affects the clinical relevance of imaging performed and the quality of resultant radiology reports. Currently, there are no published Australian guidelines on what constitutes adequate clinical information in computed tomography (CT) requests. This study aimed to determine specific items of clinical information radiologists require in CT requests for acute chest, abdomen and blunt trauma examinations, to support optimal reporting.

METHODS

A panel of 24 CT-reporting consultant radiologists participated in this e-Delphi consensus study. Panellists undertook multiple online survey rounds of open-ended, dichotomous and Likert scale questions, receiving feedback following each. Round 1 responses formulated lists for each CT examination. Round 2 set a threshold of 80% agreement after dichotomous scoring. Round 3 accepted items which averaged 4 or more on a 5-point Likert scale. Round 4 required panellists to rank items within the aggregated, accepted lists, based on panellists' perceived level of usefulness.

RESULTS

The large numbers of round 1 items (chest: 101, abdomen: 76, blunt trauma: 80) were rationalised and grouped into categories to facilitate efficiency during subsequent rounds. Twenty-three chest, 24 abdomen and 17 blunt trauma items met the 80% agreement threshold in round 2. Items below threshold were included in round 3; numbering 44, 19 and 23 for chest, abdomen and blunt trauma, respectively. Through the e-Delphi process, we formulated clinical information criteria standards for three CT types.

CONCLUSIONS

The developed standards will guide Australian referrers in providing adequate clinical information in CT requests, to support optimal reporting, diagnosis and treatment.

Talking Points: Enhancing Communication Between Radiologists and Patients

Radiologists communicate along multiple pathways, using written, verbal, and non-verbal means. Radiology trainees must gain skills in all forms of communication, with attention to developing effective professional communication in all forms. This manuscript reviews evidence-based strategies for enhancing effective communication between radiologists and patients through direct communication, written means and enhanced reporting. We highlight patient-centered communication efforts, available evidence, and opportunities to engage learners and enhance training and simulation efforts that improve communication with patients at all levels of clinical care.

Clinical decision support system recommendations: how often do radiologists and clinicians accept them?

OBJECTIVE

To assess the acceptance and reliability of clinical decision support system (CDSS) imaging referral scores (ESR iGuide).

METHODS

A pilot study was conducted in a tertiary hospital. Four different experts were invited to rate 40 simulated clinical cases on a 5-level scale, for the level of agreement with the ESR iGuide's recommended procedures. In cases of disagreement, physicians were asked to indicate the reason. Descriptive measures were calculated for the level of agreement. We also explored the degree of agreement between four different specialists, and examined the cases in which clinicians disagreed with ESR iGuide best practice recommendations.

RESULTS

The mean rating of the four experts for the 40 clinical simulated cases was 4.17 ± 0.65, median 4.25 (on a scale of 1-5). All four raters totally agreed with the system recommendation in 75% of cases. No significant relationship was found between the degree of agreement and the number of indications and the patient's age or gender. In an optimistic scenario, using a binary agree/disagree variable, the Overall Percentage Agreement for the rating of the 40 simulated cases between the four experts was 77.28%. There were a total of 20 disagreements out of 160 cases with the ESR iGuide, of which 7 were among the two radiologists.

CONCLUSIONS

CDSS can be an effective tool for guiding the selection of appropriate imaging examinations, thus cutting costs due to unnecessary imaging scans. Since this is a pilot study, further research on a larger scale, preferably at national level, is required.

KEY POINTS

• The average of the mean rating of the four experts was 4.17 ± 0.65, median 4.25, on a scale of 1-5 where 5 represents total agreement with the CDSS tool. • In an optimistic scenario, using a binary agree/disagree variable, the Overall Percentage Agreement between the four experts was 77.28%. • Radiologists had fewer disagreements with the recommendations of the CDSS tool than other physicians, indicating a better fit of the support system to radiology experts' perspective.

Reason for Exam Imaging Reporting and Data System: Consensus Reached on Quality Assessment of Radiology Requisitions

OBJECTIVE

The aim of this study was to reach consensus on quality assessment of clinical information in imaging requisitions using Reason for exam Imaging Reporting and Data System (RI-RADS).

METHODS

A Delphi study was conducted in September 2018 with a panel of 87 radiologists with diverse levels of experience from various settings (community hospitals, private hospitals, university hospitals, and clinics), of which 74.7% completed the survey. The agreement was assessed in the following subjects: (a) presumed effect of standardization, (b) the standardized system for information, (c) the scoring system for evaluation of requisitions, and (d) the implementation of RI-RADS. The consensus threshold was set at 51% responding (strongly) agree. The rate of lawsuits preventable with clinical information was also assessed.

RESULTS

Consensus was reached on all objectives of the study with a high level of agreement. Radiologists agreed on the need for standardization of imaging requisitions and attributed it to increased speed and accuracy of interpretations. Three categories of information were determined as key indicators of quality: impression, clinical findings, and clinical question. The scoring system is intended to grade requisitions based on the presence of these categories. Radiologists also agreed that RI-RADS will encourage physicians to improve requisitions. Among radiologists who responded to the survey, 12.6% had experienced at least one lawsuit potentially preventable with sufficient information in requisitions.

CONCLUSIONS

Reason for exam Imaging Reporting and Data System can be used as a standard for quality assessment of requisitions. Its use may improve the quality of patient care and reduce lawsuits against radiologists.

A reporting and analysis framework for structured evaluation of COVID-19 clinical and imaging data

The COVID-19 pandemic has worldwide individual and socioeconomic consequences. Chest computed tomography has been found to support diagnostics and disease monitoring. A standardized approach to generate, collect, analyze, and share clinical and imaging information in the highest quality possible is urgently needed. We developed systematic, computer-assisted and context-guided electronic data capture on the FDA-approved mint Lesion^TM software platform to enable cloud-based data collection and real-time analysis. The acquisition and annotation include radiological findings and radiomics performed directly on primary imaging data together with information from the patient history and clinical data. As proof of concept, anonymized data of 283 patients with either suspected or confirmed SARS-CoV-2 infection from eight European medical centers were aggregated in data analysis dashboards. Aggregated data were compared to key findings of landmark research literature. This concept has been chosen for use in the national COVID-19 response of the radiological departments of all university hospitals in Germany.

Requests for radiologic imaging: Prevalence and determinants of inadequate quality according to RI-RADS

PURPOSE

To determine the prevalence and determinants of radiologic imaging requests that are of inadequate quality according to the Reason for exam Imaging Reporting and Data System (RI-RADS).

METHODS

This study included a random sample of 673 radiologic examinations performed at a tertiary care center. The quality of each imaging request was graded according to RI-RADS. Ordinal regression analysis was performed to determine the association of RI-RADS grade with patient age, gender, and hospital status, indication for imaging, requesting specialty, imaging modality, body region, time of examination, and relationship with previous imaging within the past one year.

RESULTS

RI-RADS grades A (adequate request), B (barely adequate request), C (considerably limited request), and D (deficient request) were assigned to 159 (23.6 %), 166 (24.7 %), 214 (31.8 %), and 134 (19.9 %) of cases, respectively. Indication for imaging, requesting specialty, and body region were independently significantly associated with RI-RADS grades. Specifically, routine preoperative imaging (odds ratio [OR]: 3.422, P = 0.030) and transplantation imaging requests (OR: 8.710, P = 0.000) had a higher risk of poorer RI-RADS grades, whereas infection/inflammation as indication for imaging (OR: 0.411, P = 0.002), pediatrics as requesting specialty (OR: 0.400, P = 0.007), and head (OR: 0.384, P = 0.017), spine (OR: 0.346, P = 0.016), and upper extremity (OR: 0.208, P = 0.000) as body regions had a lower risk of poorer RI-RADS grades.

CONCLUSION

The quality of radiologic imaging requests is inadequate in >75 % of cases, and is affected by several factors. The data from this study can be used as a baseline and benchmark for further investigation and improvement.

Imaging Findings of SARS-CoV-2 Infection in Pediatrics: A Systematic Review of Coronavirus Disease 2019 (COVID-19) in 850 Patients

RATIONALE AND OBJECTIVES

Children with COVID-19 seem to have a relatively milder disease and better prognosis; however, severe disease or death could still occur in this age group. Although the knowledge on the clinical and epidemiology of COVID-19 in pediatric patients is being accumulated rapidly, relevant comprehensive review on its radiological manifestations is still lacking. The present article reviews the radiological characteristics of COVID-19 in pediatrics, based on the previous studies.

MATERIALS AND METHODS

We conducted a systematic literature search for published articles by using Medline, Scopus, Google Scholar and Embase online databases. All studies describing CT findings of COVID-19 in pediatrics (<18years) were included.

RESULTS

A total of 39 studies with 850 pediatric patients were reviewed. 225 (26.5%) of patients had normal CT findings. Ground-glass opacities and consolidations were the most common CT abnormalities (384/625, 61.5%). Other findings were halo sign, interstitial opacities, bronchial wall thickening, and crazy-paving sign. Approximately 55% of patients had unilateral pulmonary findings. Most studies found peripheral and lower-lobe distribution to be a prominent imaging finding.

CONCLUSION

Our study showed that imaging findings in children were often milder and more focal than adults, typically as ground-glass opacities and consolidations with unilateral lower-lobe predominance, which have been regressed during the recovery time. A balance must be struck between the risk of radiation and the need for chest CT. If still necessary, low-dose CT is more appropriate in this age group. Albeit, due to the limited number of reported pediatrics with COVID-19, and the lack of consistency in CT descriptors, further work is still needed in this regard.

Chest Computed Tomography Manifestation of Coronavirus Disease 2019 (COVID-19) in Patients With Cardiothoracic Conditions

The coronavirus disease 2019 (COVID-19) pandemic is a serious public health concern, with an exponentially growing number of patients worldwide. Computed tomography (CT) has been suggested as a highly sensitive modality for the diagnosis of pulmonary involvement in the early stages of COVID-19. The typical features of COVID-19 in chest CT include bilateral, peripheral, and multifocal ground-glass opacities with or without superimposed consolidations. Patients with underlying medical conditions are at higher risks of complications and mortality. The diagnosis of COVID-19 on the basis of the imaging features may be more challenging in patients with preexisting cardiothoracic conditions, such as chronic obstructive pulmonary disease, interstitial lung disease, cardiovascular disease, and malignancies with cardiothoracic involvement. The extensive pulmonary involvement in some of these pathologies may obscure the typical manifestation of COVID-19, whereas other preexisting pathologies may resemble the atypical or rare CT manifestations of this viral pneumonia. Thus, understanding the specific CT manifestations in these special subgroups is essential for a prompt diagnosis.

Coronavirus disease 2019 (COVID-19) imaging reporting and data system (COVID-RADS) and common lexicon: a proposal based on the imaging data of 37 studies

BACKGROUND

In the vast majority of the laboratory-confirmed coronavirus disease 2019 (COVID-19) patients, computed tomography (CT) examinations yield a typical pattern and the sensitivity of this modality has been reported to be 97% in a large-scale study. Structured reporting systems simplify the interpretation and reporting of imaging examinations, serve as a framework for consistent generation of recommendations, and improve the quality of patient care.

PURPOSE

To compose a comprehensive lexicon for description of the imaging findings and propose a grading system and structured reporting format for CT findings in COVID-19.

MATERIAL AND METHODS

We updated our published systematic review on imaging findings in COVID-19 to include 37 published studies pertaining to diagnostic features of COVID-19 in chest CT. Using the reported imaging findings of 3647 patients, we summarized the typical chest CT findings, atypical features, and temporal changes of COVID-19 in chest CT. Subsequently, we extracted a list of descriptive terms and mapped it to the terminology that is commonly used in imaging literature.

RESULTS

We composed a comprehensive lexicon that can be used for documentation and reporting of typical and atypical CT imaging findings in COVID-19 patients. Using the same data, we propose a grading system with five COVID-RADS categories. Each COVID-RADS grade corresponds to a low, moderate, or high level of suspicion for pulmonary involvement of COVID-19.

CONCLUSION

The proposed COVID-RADS and common lexicon would improve the communication of findings to other healthcare providers, thus facilitating the diagnosis and management of COVID-19 patients.

KEY POINTS

• Chest CT has high sensitivity in diagnosing the coronavirus disease 2019 (COVID-19). • Structured reporting systems simplify the interpretation and reporting of imaging examinations, serve as a framework for consistent generation of recommendations, and improve the quality of patient care. • The proposed COVID-RADS and common lexicon would improve the communication of findings to other healthcare providers, thus facilitating the diagnosis and management of COVID-19 patients.

Deep learning, reusable and problem-based architectures for detection of consolidation on chest X-ray images

BACKGROUND AND OBJECTIVE

In most patients presenting with respiratory symptoms, the findings of chest radiography play a key role in the diagnosis, management, and follow-up of the disease. Consolidation is a common term in radiology, which indicates focally increased lung density. When the alveolar structures become filled with pus, fluid, blood cells or protein subsequent to a pulmonary pathological process, it may result in different types of lung opacity in chest radiograph. This study aims at detecting consolidations in chest x-ray radiographs, with a certain precision, using artificial intelligence and especially Deep Convolutional Neural Networks to assist radiologist for better diagnosis.

METHODS

Medical image datasets usually are relatively small to be used for training a Deep Convolutional Neural Network (DCNN), so transfer learning technique with well-known DCNNs pre-trained with ImageNet dataset are used to improve the accuracy of the models. ImageNet feature space is different from medical images and in the other side, the well-known DCNNs are designed to achieve the best performance on ImageNet. Therefore, they cannot show their best performance on medical images. To overcome this problem, we designed a problem-based architecture which preserves the information of images for detecting consolidation in Pediatric Chest X-ray dataset. We proposed a three-step pre-processing approach to enhance generalization of the models. To demonstrate the correctness of numerical results, an occlusion test is applied to visualize outputs of the model and localize the detected appropriate area. A different dataset as an extra validation is used in order to investigate the generalization of the proposed model.

RESULTS

The best accuracy to detect consolidation is 94.67% obtained by our problem based architecture for the understudy dataset which outperforms the previous works and the other architectures.

CONCLUSIONS

The designed models can be employed as computer aided diagnosis tools in real practice. We critically discussed the datasets and the previous works based on them and show that without some considerations the results of them may be misleading. We believe, the output of AI should be only interpreted as focal consolidation. The clinical significance of the finding can not be interpreted without integration of clinical data.