Automatic Fully-Contextualized Recommendation Extraction from Radiology Reports.

Recommendations are a key component of radiology reports. Automatic extraction of recommendations would facilitate tasks such as recommendation tracking, quality improvement, and large-scale descriptive studies. Existing report-parsing systems are frequently limited to recommendations for follow-up imaging studies, operate at the sentence or document level rather than the individual recommendation level, and do not extract important contextualizing information. We present a neural network architecture capable of extracting fully contextualized recommendations from any type of radiology report. We identified six major "questions" necessary to capture the majority of context associated with a recommendation: recommendation, time period, reason, conditionality, strength, and negation. We developed a unified task representation by allowing questions to refer to answers to other questions. Our representation allows for a single system to perform named entity recognition (NER) and classification tasks. We annotated 2272 radiology reports from all specialties, imaging modalities, and multiple hospitals across our institution. We evaluated the performance of a long short-term memory (LSTM) architecture on the six-question task. The single-task LSTM model achieves a token-level performance of 89.2% at recommendation extraction, and token-level performances between 85 and 95% F1 on extracting modifying features. Our model extracts all types of recommendations, including follow-up imaging, tissue biopsies, and clinical correlation, and can operate in real time. It is feasible to extract complete contextualized recommendations of all types from arbitrary radiology reports. The approach is likely generalizable to other clinical entities referenced in radiology reports, such as radiologic findings or diagnoses.

Patient Exposure from Radiologic and Nuclear Medicine Procedures in the United States: Procedure Volume and Effective Dose for the Period 2006-2016.

Background Comprehensive assessments of the frequency and associated doses from radiologic and nuclear medicine procedures are rarely conducted. The use of these procedures and the population-based radiation dose increased remarkably from 1980 to 2006. Purpose To determine the change in per capita radiation exposure in the United States from 2006 to 2016. Materials and Methods The U.S. National Council on Radiation Protection and Measurements conducted a retrospective assessment for 2016 and compared the results to previously published data for the year 2006. Effective dose values for procedures were obtained from the literature, and frequency data were obtained from commercial, governmental, and professional society data. Results In the United States in 2006, an estimated 377 million diagnostic and interventional radiologic examinations were performed. This value remained essentially the same for 2016 even though the U.S. population had increased by about 24 million people. The number of CT scans performed increased from 67 million to 84 million, but the number of other procedures (eg, diagnostic fluoroscopy) and nuclear medicine procedures decreased from 17 million to 13.5 million. The number of dental radiographic and dental CT examinations performed was estimated to be about 320 million in 2016. Using the tissue-weighting factors from Publication 60 of the International Commission on Radiological Protection, the U.S. annual individual (per capita) effective dose from diagnostic and interventional medical procedures was estimated to have been 2.9 mSv in 2006 and 2.3 mSv in 2016, with the collective doses being 885 000 and 755 000 person-sievert, respectively. Conclusion The trend from 1980 to 2006 of increasing dose from medical radiation has reversed. Estimated 2016 total collective effective dose and radiation dose per capita dose are lower than in 2006. © RSNA, 2020 See also the editorial by Einstein in this issue.

Automated Detection of Radiology Reports that Require Follow-up Imaging Using Natural Language Processing Feature Engineering and Machine Learning Classification.

While radiologists regularly issue follow-up recommendations, our preliminary research has shown that anywhere from 35 to 50% of patients who receive follow-up recommendations for findings of possible cancer on abdominopelvic imaging do not return for follow-up. As such, they remain at risk for adverse outcomes related to missed or delayed cancer diagnosis. In this study, we develop an algorithm to automatically detect free text radiology reports that have a follow-up recommendation using natural language processing (NLP) techniques and machine learning models. The data set used in this study consists of 6000 free text reports from the author's institution. NLP techniques are used to engineer 1500 features, which include the most informative unigrams, bigrams, and trigrams in the training corpus after performing tokenization and Porter stemming. On this data set, we train naive Bayes, decision tree, and maximum entropy models. The decision tree model, with an F1 score of 0.458 and accuracy of 0.862, outperforms both the naive Bayes (F1 score of 0.381) and maximum entropy (F1 score of 0.387) models. The models were analyzed to determine predictive features, with term frequency of n-grams such as "renal neoplasm" and "evalu with enhanc" being most predictive of a follow-up recommendation. Key to maximizing performance was feature engineering that extracts predictive information and appropriate selection of machine learning algorithms based on the feature set.

Radiation protection for the echocardiographers: "To each their own".

This article illustrates the effectiveness of targeted radioprotective strategies for the interventional echocardiographer. The reader should recognize the importance of engagement of all team members in the multifaceted process of radiation exposure mitigation. Future efforts/studies should focus on the impact of team oriented training, lab design, and development of novel supplies and equipment to mitigate radiation exposure of all personnel in the cardiac catheterization lab, particularly during more complex interventional procedures.

Assessing cardiovascular risk in the prerenal transplant population: Comparison of myocardial perfusion imaging and coronary angiography with risk factor stratification.

INTRODUCTION: Patients with end-stage renal disease (ESRD) have a higher incidence of coronary artery disease (CAD). Hence, it is crucial to evaluate CAD before renal transplantation. This study compares the utility of pharmacologic single-photon emission computed-tomography (SPECT) imaging directly to coronary angiography for diagnosis of CAD with correlation to cardiovascular risk factors. METHOD: Retrospective review of asymptomatic renal failure patients who underwent both SPECT and coronary angiography to identify obstructive CAD between the years 2008-2016. Ninety-four ESRD subjects were evaluated. RESULTS: Myocardial perfusion SPECT study found, when compared to coronary angiography demonstrated for CAD, the sensitivity of 93.3% with a specificity of 73.4%. Importantly, the negative predictive value for coronary artery disease was 96%. With seven or more cardiac risk factors, 66.7% of patients had obstructive coronary artery disease. Among all the risk factors examined, patients with a previous history of coronary artery disease had a 68% risk of obstructive coronary artery disease. CONCLUSION: Comparing myocardial perfusion imaging SPECT findings with coronary angiography in patients with ESRD, a sensitivity of 93.3% and a specificity of 73% were observed. Of all the risk factors examined, patient with the previous history of CAD was the single most significant risk factor for CAD in 68% of cases.

Toward Complete Structured Information Extraction from Radiology Reports Using Machine Learning.

Unstructured and semi-structured radiology reports represent an underutilized trove of information for machine learning (ML)-based clinical informatics applications, including abnormality tracking systems, research cohort identification, point-of-care summarization, semi-automated report writing, and as a source of weak data labels for training image processing systems. Clinical ML systems must be interpretable to ensure user trust. To create interpretable models applicable to all of these tasks, we can build general-purpose systems which extract all relevant human-level assertions or "facts" documented in reports; identifying these facts is an information extraction (IE) task. Previous IE work in radiology has focused on a limited set of information, and extracts isolated entities (i.e., single words such as "lesion" or "cyst") rather than complete facts, which require the linking of multiple entities and modifiers. Here, we develop a prototype system to extract all useful information in abdominopelvic radiology reports (findings, recommendations, clinical history, procedures, imaging indications and limitations, etc.), in the form of complete, contextualized facts. We construct an information schema to capture the bulk of information in reports, develop real-time ML models to extract this information, and demonstrate the feasibility and performance of the system.

Reducing radiation dose: Equipment, procedure, and operator Perfecting the Trifecta.

This article illustrates the positive impact of fluoroscopic imaging equipment on radiation dose reduction in CTO PCI. The reader should recognize the importance of purchasing and maintaining the best equipment, understanding procedure/patient complexity, and assuring operator training in radiation dose reduction. Future efforts/studies should focus upon all three areas of dose reduction for best results.

Real time patient dosimetry in the cath lab: Can you see what they get?

Real time radiation dose monitoring in the cath lab may provide immediate feedback for potential dose reduction during PCI. Radiation dose monitoring to predict potential tissue injury utilizes equipment measured air Kerma at the interventional reference point (IRP) with then calculated specific tissue peak skin dose. The role of the cath lab Quality Committee is not only to assess individual high dose radiation cases but also to create processes and assess new technologies to assure radiation dose is best utilized in all cases.

Radiation-associated lens changes in the cardiac catheterization laboratory: Results from the IC-CATARACT (CATaracts Attributed to RAdiation in the CaTh lab) study.

OBJECTIVE: To examine the relationship between occupational exposure to ionizing radiation and the prevalence of lens changes in interventional cardiologists (ICs) and catheterization laboratory ("cath-lab") staff. BACKGROUND: Exposure to ionizing radiation is associated with the development of lens opacities. ICs and cath-lab staff can receive high doses of ionizing radiation without protection, and may thus be at risk for lens opacity formation. METHODS: We conducted a cross-sectional study at an interventional cardiology conference. Study participants completed a questionnaire pertaining to occupational exposure to radiation and potential confounders for the development of cataracts, followed by slit-lamp examination and grading of lens findings. RESULTS: A total of 117 attendees participated in the study, including 99 (85%; 49 ± 11 years-old; 82% male) with occupational exposure to ionizing radiation and 18 (15%; 39 ± 12 years-old; 61% male) unexposed controls. The prevalence of overall cortical and posterior subcapsular lens changes (including subclinical findings) was higher in exposed participants compared with controls (47 vs. 17%, P = 0.015). Occupational exposure and age over 60 were independent predictors of lens changes (odds ratio [95% CI]: 6.07 [1.38-43.45] and 7.72 [1.60-43.34], respectively). The prevalence of frank opacities was low and similar between the two groups (14 vs. 6%, P = 0.461). Most lens findings consisted of subclinical changes in the periphery of the lens without impact on visual acuity. CONCLUSIONS: Compared with unexposed controls, ICs and cath-lab staff had a higher prevalence of lens changes that may be attributable to ionizing radiation exposure. While most of these changes were subclinical, they are important due to the potential to progress to clinical symptoms, highlighting the importance of minimizing staff radiation exposure.

2018 ACC/HRS/NASCI/SCAI/SCCT Expert Consensus Document on Optimal Use of Ionizing Radiation in Cardiovascular Imaging-Best Practices for Safety and Effectiveness, Part 2: Radiological Equipment Operation, Dose-Sparing Methodologies, Patient and Medical Personnel Protection.

The stimulus to create this document was the recognition that ionizing radiation-guided cardiovascular procedures are being performed with increasing frequency, leading to greater patient radiation exposure and, potentially, to greater exposure to clinical personnel. While the clinical benefit of these procedures is substantial, there is concern about the implications of medical radiation exposure. ACC leadership concluded that it is important to provide practitioners with an educational resource that assembles and interprets the current radiation knowledge base relevant to cardiovascular procedures. By applying this knowledge base, cardiovascular practitioners will be able to select procedures optimally, and minimize radiation exposure to patients and to clinical personnel. "Optimal Use of Ionizing Radiation in Cardiovascular Imaging - Best Practices for Safety and Effectiveness" is a comprehensive overview of ionizing radiation use in cardiovascular procedures and is published online. To provide the most value to our members, we divided the print version of this document into 2 focused parts. "Part I: Radiation Physics and Radiation Biology" addresses radiation physics, dosimetry and detrimental biologic effects. "Part II: Radiologic Equipment Operation, Dose-Sparing Methodologies, Patient and Medical Personnel Protection" covers the basics of operation and radiation delivery for the 3 cardiovascular imaging modalities (x-ray fluoroscopy, x-ray computed tomography, and nuclear scintigraphy). For each modality, it includes the determinants of radiation exposure and techniques to minimize exposure to both patients and to medical personnel.

2018 ACC/HRS/NASCI/SCAI/SCCT Expert Consensus Document on Optimal Use of Ionizing Radiation in Cardiovascular Imaging-Best Practices for Safety and Effectiveness, Part 1: Radiation Physics and Radiation Biology: A Report of the American College of Cardiology Task Force on Expert Consensus Decision Pathways Developed in Collaboration With Mended Hearts.

The stimulus to create this document was the recognition that ionizing radiation-guided cardiovascular procedures are being performed with increasing frequency, leading to greater patient radiation exposure and, potentially, to greater exposure for clinical personnel. Although the clinical benefit of these procedures is substantial, there is concern about the implications of medical radiation exposure. The American College of Cardiology leadership concluded that it is important to provide practitioners with an educational resource that assembles and interprets the current radiation knowledge base relevant to cardiovascular procedures. By applying this knowledge base, cardiovascular practitioners will be able to select procedures optimally, and minimize radiation exposure to patients and to clinical personnel. Optimal Use of Ionizing Radiation in Cardiovascular Imaging: Best Practices for Safety and Effectiveness is a comprehensive overview of ionizing radiation use in cardiovascular procedures and is published online. To provide the most value to our members, we divided the print version of this document into 2 focused parts. Part I: Radiation Physics and Radiation Biology addresses the issue of medical radiation exposure, the basics of radiation physics and dosimetry, and the basics of radiation biology and radiation-induced adverse effects. Part II: Radiological Equipment Operation, Dose-Sparing Methodologies, Patient and Medical Personnel Protection covers the basics of operation and radiation delivery for the 3 cardiovascular imaging modalities (x-ray fluoroscopy, x-ray computed tomography, and nuclear scintigraphy) and will be published in the next issue of the Journal.

Fluoroscopy pulse rate reduction during diagnostic and therapeutic imaging in the cardiac catheterization laboratory: An evaluation of radiation dose, procedure complications and outcomes.

OBJECTIVES: To evaluate radiation reduction by reducing fluoroscopy pulse rate in diagnostic cardiac catheterizations and percutaneous coronary interventions (PCI) as well as outcomes at 30 days and six months. BACKGROUND: Radiation exposure to the public at large has increased dramatically over the past three decades, and the cardiac catheterization laboratory is a large contributor. Fluoroscopy pulse rate is one way to decrease radiation exposure. METHODS: Fluoroscopy pulse rate was reduced from 10 pulses/sec (p/s) to 7.5 p/s as part of an internal quality improvement project. A retrospective analysis of all cardiac catheterizations was performed, evaluating Air KERMA at the interventional reference point (Ka, r ), Air KERMA area product (PKA ), procedural complications and major adverse cardiac events at 30 days and 6 months. RESULTS: In diagnostic catheterization median PKA (µGy·m2 ) and Ka,r (mGy) were significantly reduced (PKA - 5,613.3 vs. 4,400, P < 0.001; Ka,r - 703.0 vs. 621.0, P = 0.041). In PCI, median PKA and Ka,r were further reduced (PKA - 13,481.6 vs. 10,648.0, P < 0.001; Ka,r - 1787.0 vs. 1,459.0, P = 0.002). There was no difference in complications, fluoroscopy time or number of stents placed. There was no difference in MACE after adjustment for number of STEMIs. CONCLUSIONS: Reducing fluoroscopy pulse rates to 7.5 from 10 is an effective way to reduce patient radiation exposure across meaningful dose indices. A pulse rate of 7.5 p/s is safe, with no difference in complications or outcomes. A fluoroscopy pulse rate of 7.5 p/s should be given strong consideration for a new standard. © 2016 Wiley Periodicals, Inc.

Optimizing Radiation Safety in the Cardiac Catheterization Laboratory: A Practical Approach.

Reducing radiation exposure during cardiovascular catheterization is of paramount importance for both patient and staff safety. Over the years, advances in equipment and application of radiation safety protocols have significantly reduced patient dose and operator exposure. This review examines the current status of radiation protection in the cardiac and vascular catheterization laboratory and summarizes best practices for minimizing radiation exposure.

Determinants of operator and patient radiation exposure during cardiac catheterization: Insights from the RadiCure (RADIation reduction during cardiac catheterization using real-timE monitoring) trial.

BACKGROUND: In the RadiCure study 505 catheterization procedures were 1:1 randomized to use or no use of real-time radiation monitoring. Use of the Bleeper Sv monitor resulted in a significant reduction in operator radiation exposure. METHODS: We examined the association between several baseline and procedural parameters with operator and patient radiation exposure using univariable and multivariable analysis in the 505 patients that were enrolled in RadiCure. All baseline demographic and procedure characteristics recorded were included in the univariable analysis. RESULTS: Median fluoroscopy time was 6.2 (2.5-12.5) minutes, median patient air kerma dose was 0.908 (0.602-1.636) Gray and median first operator exposure was 10 (5-22) µSv. For analysis purposes, the 505 procedures were dichotomized based on the median operator exposure (10 µSv) and median patient radiation dose (0.908 Gray). On multivariable analysis, factors associated with high (above median or >10 µSv) first operator radiation exposure included radial access (odds ratio [OR] 5.44, 95% Confidence Interval [CI] 2.88-10.76), chronic total occlusion (CTO) intervention (OR 12.78, 95% CI 4.42-43.60), real-time radiation monitoring (OR 0.42, 95% CI 0.26-0.66), and use of a radioabsorbent drape (OR 0.53, 95% CI 0.28-0.96). High patient radiation dose (above median or >0.908 Gray) was associated with body mass index>30 kg/m2 (OR 3.22, 95% CI 1.99-5.29), prior MI (OR 2.26, 95% CI 1.29-4.04), prior cerebrovascular disease (OR 0.34, 95% CI 0.15-0.75), hypertension (OR 2.40, 95% CI 1.05-5.82), prior coronary artery bypass graft surgery (OR 2.46, 95% CI 1.40-4.39) and CTO intervention (OR 12.93, 95% CI 3.28-87.31), but was not associated with real-time radiation monitoring and use of a radioabsorbent drape. CONCLUSIONS: Several clinical and procedural factors are associated with higher patient and operator radiation exposure. Real-time radiation monitoring and use of disposable radiation shields were associated with lower operator, but not patient, radiation dose. © 2015 Wiley Periodicals, Inc.

Impact of Operators' Height on Individual Radiation Exposure Measurements During Catheter-Based Cardiovascular Interventions.

AIM: This study is aimed to evaluate the impact of an operators' height on personal radiation exposure measurements during cardiovascular interventional procedures. Based upon both clinical data and phantom simulation, a new approach for monitoring an individual's exposure is proposed. METHODS: The clinical component of this study was composed of the operators and staff in a single center full service cardiovascular laboratory being divided into 2 groups based upon their height: group A included all individuals whose height was <165 cm; group B included the individuals >165 cm. All operators wore a standard TLD dosimeter at all times with doses recorded for 12 months. To support these clinical findings, a second investigation was performed utilizing a phantom. Measurements were obtained at 100 and 135 cm from the radiation source during simulation of different cardiovascular interventional procedures. RESULTS: The radiation dose measured from the personal dosimeters identified that Group A, operators <165 cm, had significantly higher doses than those recorded in Group B, operators >165 cm, when compared among individuals performing similar tasks (physicians, technicians, and nurses): 4.55 ± 4.0 (Group A) versus 1.95 ± 1.0 (Group B) mSv (P < 0.01). During procedure simulation with the phantom, the doses measured were similarly significantly higher if measured at 100 cm than at 135 cm from the radiation source. CONCLUSION: This study suggests that the height from radiation source does impact the measured dose from an operator worn personal TLD. This was operator specific, consistent thought-out multiple procedures, and confined with phantom measurements.

Radiation dose variation in fluoroscopic imaging: Is What You Get, What You See?

Variations in dose exist amongst various imaging systems and in various acquisition angulations; this is not a new finding. Assessment of Image quality should accompany an assessment of dose, as these complement each other. Routine assessment of dose and image quality with potential comparisons amongst operators, laboratories, and imaging systems may prove beneficial for radiation safety in interventional cardiology practice.

Occupational health hazards of interventional cardiologists in the current decade: Results of the 2014 SCAI membership survey.

BACKGROUND: Interventional cardiologists and staff are subject to unique physical demands that predispose them to distinct occupational health hazards not seen in other medical disciplines. METHODS: To characterize the prevalence of these occupational health problems, The Society for Cardiovascular Angiography and Interventions (SCAI) surveyed its members by email. Inquiries included age, years of invasive practice, and diagnostic and interventional cases per year. Questions focused on orthopedic (spine, hips, knees, and ankles) and radiation-associated problems (cataracts and cancers). RESULTS: There were 314 responses. Responders were on average busy and experienced, performing a mean of 380±249 diagnostic and 200±129 interventional cases annually. Of the responders, 6.9% of operators have had to limit their caseload because of radiation exposure and 9.3% have had a health-related period of absence. Furthermore, 153 (49.4%) operators reported at least one orthopedic injury: 24.7% cervical spine disease, 34.4% lumbar spine problems, and 19.6% hip, knee or ankle joint problems. Age was most significantly correlated with orthopedic illnesses: cervical injuries (χ2=150.7, P<0.0001); hip/knee or ankle injuries (χ2=80.9, P<0.0001); lumbar injuries (χ2=147.0, P<0.0001); and any orthopedic illness (χ2= 241.2, P<0.0001). Annual total caseload was also associated: the estimated change in the odds of orthopedic illness for each additional total caseload quintile is 1.0013 (1.0001, 1.0026). There is a small but substantial incidence of cancer. CONCLUSIONS: These findings are consistent with, and extend the findings, of a prior 2004 SCAI survey, in documenting a substantial prevalence of orthopedic complications among active interventional cardiologists, which persists despite increased awareness.

Myocardial perfusion imaging is an effective screening test for coronary artery disease in liver transplant candidates.

A reliable screening test for coronary artery disease (CAD) in liver transplant (LT) candidates with end-stage liver disease is essential because a high percentage of perioperative mortality and morbidity is CAD-related. In this study, the effectiveness of myocardial perfusion imaging (MPI) for identification of significant CAD in LT candidates was evaluated. Records of 244 patients meeting criteria for MPI were evaluated: 74 met inclusion criteria; 40 had a positive MPI and cardiology follow-up; 27 had a negative MPI and underwent LT; and seven had a negative MPI and then had coronary angiography or a significant cardiac event. A selective MPI interpretation strategy was established where MPI-positive patients were divided into high, intermediate, and low CAD risk groups. The overall incidence of CAD in this study population was 5.1% and our strategy resulted in PPV 20%, NPV 94%, sensitivity 80%, and specificity 50% for categorizing CAD risk. When applied only to the subset of patients categorized as high CAD risk, the strategy was more effective, with PPV 67%, NPV 97%, sensitivity 80%, and specificity 94%. We determined that renal dysfunction was an independent predictive factor for CAD (p < 0.0001, odds ratio = 8.1), and grades of coronary occlusion correlated significantly with chronic renal dysfunction (p = 0.0079).