Search

Purpose: The aim of this study was to implement and evaluate a quality assurance (QA) workflow that leverages natural language processing to rapidly resolve inadvertent discordance between radiologists and an artificial intelligence (AI) decision support system (DSS) in the interpretation of high-acuity CT studies when the radiologist does not engage with AI DSS output., Methods: All consecutive high-acuity adult CT examinations performed in a health system between March 1, 2020, and September 20, 2022, were interpreted alongside an AI DSS (Aidoc) for intracranial hemorrhage, cervical spine fracture, and pulmonary embolus. CT studies were flagged for this QA workflow if they met three criteria: (1) negative results by radiologist report, (2) a high probability of positive results by the AI DSS, and (3) unviewed AI DSS output. In these cases, an automated e-mail notification was sent to our quality team. If discordance was confirmed on secondary review-an initially missed diagnosis-addendum and communication documentation was performed., Results: Of 111,674 high-acuity CT examinations interpreted alongside the AI DSS over this 2.5-year time period, the frequency of missed diagnoses (intracranial hemorrhage, pulmonary embolus, and cervical spine fracture) uncovered by this workflow was 0.02% (n = 26). Of 12,412 CT studies prioritized as depicting positive findings by the AI DSS, 0.4% (n = 46) were discordant, unengaged, and flagged for QA. Among these discordant cases, 57% (26 of 46) were determined to be true positives. Addendum and communication documentation was performed within 24 hours of the initial report signing in 85% of these cases., Conclusions: Inadvertent discordance between radiologists and the AI DSS occurred in a small number of cases. This QA workflow leveraged natural language processing to rapidly detect, notify, and resolve these discrepancies and prevent potential missed diagnoses., (Copyright © 2023 American College of Radiology. Published by Elsevier Inc. All rights reserved.)

Background and Purpose: Communication gaps exist between radiologists and referring physicians in conveying diagnostic certainty. We aimed to explore deep learning-based bidirectional contextual language models for automatically assessing diagnostic certainty expressed in the radiology reports to facilitate the precision of communication., Materials and Methods: We randomly sampled 594 head MR imaging reports from an academic medical center. We asked 3 board-certified radiologists to read sentences from the Impression section and assign each sentence 1 of the 4 certainty categories: "Non-Definitive," "Definitive-Mild," "Definitive-Strong," "Other." Using the annotated 2352 sentences, we developed and validated a natural language-processing system based on the start-of-the-art bidirectional encoder representations from transformers (BERT), which can capture contextual uncertainty semantics beyond the lexicon level. Finally, we evaluated 3 BERT variant models and reported standard metrics including sensitivity, specificity, and area under the curve., Results: A κ score of 0.74 was achieved for interannotator agreement on uncertainty interpretations among 3 radiologists. For the 3 BERT variant models, the biomedical variant (BioBERT) achieved the best macro-average area under the curve of 0.931 (compared with 0.928 for the BERT-base and 0.925 for the clinical variant [ClinicalBERT]) on the validation data. All 3 models yielded high macro-average specificity (93.13%-93.65%), while the BERT-base obtained the highest macro-average sensitivity of 79.46% (compared with 79.08% for BioBERT and 78.52% for ClinicalBERT). The BioBERT model showed great generalizability on the heldout test data with a macro-average sensitivity of 77.29%, specificity of 92.89%, and area under the curve of 0.93., Conclusions: A deep transfer learning model can be developed to reliably assess the level of uncertainty communicated in a radiology report., (© 2021 by American Journal of Neuroradiology.)

Within an everchanging healthcare system, continuous evaluation of standard operating procedures must be performed to ensure optimization of system level organization, communication, and efficiency. Using the Lean management approach, our institution introduced modifications to our musculoskeletal (MSK) radiology workflow in order to facilitate beneficial change that improved clinical workflow efficiency, reduced moonlighting costs, and improved radiologist satisfaction without sacrificing quality of care. The scope of our study included the MSK division of adult inpatient and outpatient populations at three hospitals in a single academic medical center. A root cause analysis was executed to determine the causative factors contributing to clinical inefficiency. Five main factors were identified, and appropriate countermeasures were introduced. Efficiency was measured via the turnaround time (TAT) for radiographic examinations, measured from exam completion to final report submission. Moonlighting expenses were monitored for the fiscal year in which the modifications were implemented. Surveys were administered to MSK radiologists before and after the countermeasures were introduced to determine subjective ratings of efficiency and satisfaction. The average TAT within our MSK division decreased from 40 h to 12 h after introducing changes to our workflow. During one fiscal year, moonlighting expenses decreased from $26,000 to $5000. Post-study survey results indicated increased efficiency of and satisfaction with our implemented modifications to the scheduling and clinical workflow. Optimization of our radiology department's workflow led to increased productivity, efficiency, and radiologist satisfaction, as well as a reduction in moonlighting costs. This project leveraged Lean management principles to combat clinical inefficiency, waste time, and high costs.

The primary purpose of this study is to determine if the implementation of an actionable findings communication system (PeerVue) with explicitly defined criteria for the classification of critical results, leads to an increase in the number of actionable findings reported by radiologists. Secondary goals are to 1) analyze the adoption rate of PeerVue and 2) assess the accuracy of the classification of actionable findings within this system. Over a two-year period, 890,204 radiology reports were analyzed retrospectively in order to identify the number of actionable findings communicated before (Year 1) and after the implementation of PeerVue (Year 2) at a tertiary care academic medical center. A sub-sample of 145 actionable findings over a two-month period in Year 2 was further analyzed to assess the degree of concordance with our reporting policy. Before PeerVue, 4623/423,070 (1.09%) of radiology reports contained an actionable finding. After its implementation, this number increased to 6825/467,134 (1.46%) (p < 0.0001). PeerVue was used in 3886/6825 (56.9%) cases with actionable findings. The remaining 2939/6825 (43.1%) were reported using the legacy tagging system. From the sub-sample taken from PeerVue, 104/145 (71.7%) were consistent with the updated reporting policy. A software program (PeerVue) utilized for the communication of actionable findings contributed to a 34% (p < 0.0001) increase in the reporting rate of actionable findings. A sub-analysis within the new system indicated a 56.9% adoption rate and a 71.7% accuracy rate in reporting of actionable findings.

All radiology departments are now expected to create organized and comprehensive quality and safety (QS) programs. No matter the department size, there are fundamental principles that should be at the core of each program. "Just culture" and culture of safety are essential principles in establishing effective programs. Physician leaders of QS programs must involve stakeholders, define program goals, and develop an effective program structure. QS programs should establish reliable quality assurance and patient safety systems. Integrating continuous quality improvement and learning into the department and prioritizing patient and referring clinician experiences will enhance outcomes. Physician QS leaders will face common obstacles in program development and management. Heightened awareness of these obstacles and understanding potential solutions will help programs succeed. Leveraging resources offered by professional societies and engaging in radiology QS community networks will provide ongoing support for program leaders. © RSNA, 2018.

The appropriate communication and management of incidental findings on emergency department (ED) radiology studies is an important component of patient safety. Guidelines have been issued by the ACR and other medical associations that best define incidental findings across various modalities and imaging studies. However, there are few examples of health care facilities designing ways to manage incidental findings. Our institution aimed to improve communication and follow-up of incidental radiology findings in ED patients through the collaborative development and implementation of system-level process changes including a standardized loop-closure method. We assembled a multidisciplinary team to address the nature of these incidental findings and designed new workflows and operational pathways for both radiology and ED staff to properly communicate incidental findings. Our results are based on all incidental findings received and acknowledged between November 1, 2016, and May 30, 2017. The total number of incidental findings discovered was 1,409. Our systematic compliance fluctuated between 45% and 95% initially after implementation. However, after overcoming various challenges through optimization, our system reached a compliance rate of 93% to 95%. Through the implementation of our new, standardized communication system, a high degree of compliance with loop closure for ED incidental radiology findings was achieved at our institution., (Copyright © 2017 American College of Radiology. Published by Elsevier Inc. All rights reserved.)

Stress fractures, including both fatigue and insufficiency types, are frequently encountered in clinical practice as a source of pain in both athletes and patients with predisposing conditions. Radiography is the imaging modality of choice for baseline diagnosis. MRI has greatly improved our ability to diagnose radiographically occult stress fractures. Tc-99m bone scan and CT may also be useful as diagnostic tools. Although fatigue and insufficiency fractures can be self-limited and go onto healing even without diagnosis, there is usually value in initiating prompt therapeutic measures as incomplete stress fractures have the potential of progressing to completion and requiring more invasive treatment or delay in return to activity. This is particularly important in the setting of stress fractures of the femoral neck. Accuracy in the identification of these injuries is also relevant because the differential diagnosis includes entities that would otherwise be treated significantly different (ie, osteoid osteoma, osteomyelitis, and metastasis). The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer-reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment., (Copyright © 2017 American College of Radiology. Published by Elsevier Inc. All rights reserved.)

Chronic hip pain is a common clinical problem whose cause is often elucidated by imaging. The ACR Appropriateness Criteria for chronic hip pain define best practices of image ordering. Clinical scenarios are followed by the imaging choices and their appropriateness. The information is in ordered tables with an accompanying narrative explanation to guide physicians to order the right test. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer-reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment., (Copyright © 2017 American College of Radiology. Published by Elsevier Inc. All rights reserved.)

Osteoporosis is a considerable public health risk, with 50% of women and 20% of men >50 years of age experiencing fracture, with mortality rates of 20% within the first year. Dual x-ray absorptiometry (DXA) is the primary diagnostic modality by which to screen women >65 years of age and men >70 years of age for osteoporosis. In postmenopausal women <65 years of age with additional risk factors for fracture, DXA is recommended. Some patients with bone mineral density above the threshold for treatment may qualify for treatment on the basis of vertebral body fractures detected through a vertebral fracture assessment scan, a lateral spine equivalent generated from a commercial DXA machine. Quantitative CT is useful in patients with advanced degenerative bony changes in their spines. New technologies such as trabecular bone score represent an emerging role for qualitative assessment of bone in clinical practice. It is critical that both radiologists and referring providers consider osteoporosis in their patients, thereby reducing substantial morbidity, mortality, and cost to the health care system. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment., (Copyright © 2017 American College of Radiology. Published by Elsevier Inc. All rights reserved.)

Evaluation for suspected inflammatory arthritis as a cause for chronic extremity joint pain often relies on imaging. This review first discusses the characteristic osseous and soft tissue abnormalities seen with inflammatory arthritis and how they may be imaged. It is essential that imaging results are interpreted in the context of clinical and serologic results to add specificity as there is significant overlap of imaging findings among the various types of arthritis. This review provides recommendations for imaging evaluation of specific types of inflammatory arthritis, including rheumatoid arthritis, seronegative spondyloarthropathy, gout, calcium pyrophosphate dihydrate disease (or pseudogout), and erosive osteoarthritis. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer-reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment., (Copyright © 2017 American College of Radiology. Published by Elsevier Inc. All rights reserved.)

There has been a rapid increase in the number of shoulder arthroplasties, including partial or complete humeral head resurfacing, hemiarthroplasty, total shoulder arthroplasty, and reverse total shoulder arthroplasty, performed in the United States over the past two decades. Imaging can play an important role in diagnosing the complications that can occur in the setting of these shoulder arthroplasties. This review is divided into two parts. The first part provides a general discussion of various imaging modalities, comprising radiography, CT, MRI, ultrasound, and nuclear medicine, and their role in providing useful, treatment-guiding information. The second part focuses on the most appropriate imaging algorithms for shoulder arthroplasty complications such as aseptic loosening, infection, fracture, rotator cuff tendon tear, and nerve injury. The evidence-based ACR Appropriateness Criteria guidelines offered in this report were reached via an extensive analysis of current medical literature from peer-reviewed journals and the application of well-established methodologies (the RAND/UCLA Appropriateness Method and the Grading of Recommendations Assessment, Development, and Evaluation) for rating the appropriateness of imaging and treatment procedures for specific clinical scenarios. Further analysis and review of the guidelines were performed by a multidisciplinary expert panel. In those instances in which there was insufficient or equivocal data for recommending the appropriate imaging algorithm, expert opinion may have supplemented the available evidence., (Copyright © 2016 American College of Radiology. Published by Elsevier Inc. All rights reserved.)

Purpose: The purpose of this study was to improve the transcription of patient information from imaging study requisitions to the radiology information database at a single institution., Methods: Five hundred radiology reports from adult outpatient radiographic examinations were chosen randomly from the radiology information system (RIS) and categorized according to their degree of concordance with their corresponding clinical order indications. The number and types of grammatical errors and types of order forms were also recorded. Countermeasures centered on the education of the technical staff and referring physician offices and the implementation of a checklist. Another sample of 500 reports was taken after the implementation of the countermeasures and compared with the baseline data using a χ(2) test., Results: The number of RIS indications perfectly concordant with their corresponding clinical order indications increased from 232 (46.4%) to 314 (62.8%) after the implementation of the countermeasures (P < .0001). The number of partially concordant matches due to inadequate RIS indications dropped from 162 (32.4%) to 114 (22.8%) (P < .001), whereas the number of partially concordant matches due to inadequate clinical order indications increased from 22 (4.4%) to 57 (11.4%) (P < .0001). The number of discordant pairings dropped from 84 (16.8%) to 15 (3%) (P < .0001). Technologists began to input additional patient information obtained from the patients (not present in the image requisitions) in the RIS after the implementation of the countermeasures., Conclusions: The education of technical staff members and the implementation of a checklist markedly improved the information provided to radiologists on image requisitions from referring providers., (Copyright © 2016 American College of Radiology. Published by Elsevier Inc. All rights reserved.)

Sarcoid-like reaction has been documented in association with several types of malignancy, including renal cell carcinoma. We report the case of a 41-year-old man with nonmetastatic renal cell carcinoma and concomitant non-caseating granulomas distributed diffusely throughout the bone marrow. The granulomas resolved after nephrectomy. As far as we know, this is the first reported case of a sarcoid-like reaction primarily involving the bone marrow in association with renal cell carcinoma.