Imaging of Lung Transplantation.

Lung transplantation is the only curative treatment for end-stage lung disease, which is caused by a wide variety of pathologies and encountered in a diverse range of patients. Potential recipients, as well as donors are carefully evaluated by imaging prior to transplant for contraindications to the transplant. After transplantation, recipients are imaged in the immediate, early, intermediate, and late periods for complications that may arise and require intervention. Radiography and computed tomography are the 2 most commonly used imaging modalities used to evaluate the chest after lung transplantation.

Management of Incidental Thyroid Nodules on Chest CT: Using Natural Language Processing to Assess White Paper Adherence and Track Patient Outcomes.

OBJECTIVE: The purpose of this study was to develop a natural language processing (NLP) pipeline to identify incidental thyroid nodules (ITNs) meeting criteria for sonographic follow-up and to assess both adherence rates to white paper recommendations and downstream outcomes related to these incidental findings. METHODS: 21583 non-contrast chest CT reports from 2017 and 2018 were retrospectively evaluated to identify reports which included either an explicit recommendation for thyroid ultrasound, a description of a nodule ≥ 1.5 cm, or description of a nodule with suspicious features. Reports from 2018 were used to train an NLP algorithm called fastText for automated identification of such reports. Algorithm performance was then evaluated on the 2017 reports. Next, any patient from 2017 with a report meeting criteria for ultrasound follow-up was further evaluated with manual chart review to determine follow-up adherence rates and nodule-related outcomes. RESULTS: NLP identified reports with ITNs meeting criteria for sonographic follow-up with an accuracy of 96.5% (95% CI 96.2-96.7) and sensitivity of 92.1% (95% CI 89.8-94.3). In 10006 chest CTs from 2017, ITN follow-up ultrasound was indicated according to white paper criteria in 81 patients (0.8%), explicitly recommended in 46.9% (38/81) of patients, and obtained in less than half of patients in which it was appropriately recommended (17/35, 48.6%). DISCUSSION: NLP accurately identified chest CT reports meeting criteria for ITN ultrasound follow-up. Radiologist adherence to white paper guidelines and subsequent referrer adherence to radiologist recommendations showed room for improvement.

Acute myocarditis after a second dose of the mRNA COVID-19 vaccine: a report of two cases.

We report two cases of myocarditis, in two young and previously healthy individuals, temporally related to the second dose of the mRNA-COVID-19 vaccine. Both patients developed acute chest pain, changes on electrocardiogram (ECG), and elevated serum troponin within two days of receiving their second dose. Cardiac magnetic resonance (CMR) findings were consistent with acute myocarditis.

Computed Tomography for Abdominal Pain: Do Radiology Reports Answer the Clinical Question?

RATIONALE AND OBJECTIVES: To assess whether abdominopelvic computed tomography (CT) radiology reports directly address a cause for pain when pain is included in the scan indication. MATERIALS AND METHODS: Four hundred and ninety-five random abdominopelvic CT reports containing "pain" in the indication were retrospectively reviewed. The position of pain descriptors within the indication, the presence of an oncology-related indication in addition to pain and whether a cause for pain was addressed in the impression were recorded. Linguistic analysis of indication and impression sections was performed. Comparisons between reports that addressed pain and those that did not were conducted using Chi-square, Fisher exact, and two-tailed t-tests. RESULTS: A cause for pain was addressed in 454 of 495 (91.7%) report impressions. Indications with both oncology-related and pain-related descriptors were less likely to have pain directly addressed (χ2 (1, N = 495) = 16.4, p < .001). There was no significant association between where pain appeared within the indication and whether pain was addressed (χ2 (1, N = 495) = 3.2, p = .07). Whether an impression conveyed a normal result did not influence if pain was addressed (p = .49). Impression word count and complexity were higher in the addressed group compared to the not addressed group (word count 66.6 vs. 51.9, p= .02, Composite grade level 30.1 vs. 25.3, p= .02). CONCLUSION: Radiologists at our institution consistently addressed a cause for pain on abdominopelvic CTs when pain was in the indication. However, oncology patients who also had an indication of pain were less likely to have a cause for pain addressed. Impression complexity was high for all reports, though higher in those where pain was addressed.

Chest CT and Coronavirus Disease (COVID-19): A Critical Review of the Literature to Date.

OBJECTIVE. Coronavirus disease (COVID-19) is a global pandemic. Studies in the radiology literature have suggested that CT might be sufficiently sensitive and specific in diagnosing COVID-19 when used in lieu of a reverse transcription-polymerase chain reaction test; however, this suggestion runs counter to current society guidelines. The purpose of this article is to critically review some of the most frequently cited studies on the use of CT for detecting COVID-19. CONCLUSION. To date, the radiology literature on COVID-19 has consisted of limited retrospective studies that do not substantiate the use of CT as a diagnostic test for COVID-19.

Using Artificial Intelligence to Revise ACR TI-RADS Risk Stratification of Thyroid Nodules: Diagnostic Accuracy and Utility.

Background Risk stratification systems for thyroid nodules are often complicated and affected by low specificity. Continual improvement of these systems is necessary to reduce the number of unnecessary thyroid biopsies. Purpose To use artificial intelligence (AI) to optimize the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS). Materials and Methods A total of 1425 biopsy-proven thyroid nodules from 1264 consecutive patients (1026 women; mean age, 52.9 years [range, 18-93 years]) were evaluated retrospectively. Expert readers assigned points based on five ACR TI-RADS categories (composition, echogenicity, shape, margin, echogenic foci), and a genetic AI algorithm was applied to a training set (1325 nodules). Point and pathologic data were used to create an optimized scoring system (hereafter, AI TI-RADS). Performance of the systems was compared by using a test set of the final 100 nodules with interpretations from the expert reader, eight nonexpert readers, and an expert panel. Initial performance of AI TI-RADS was calculated by using a test for differences between binomial proportions. Additional comparisons across readers were conducted by using bootstrapping; diagnostic performance was assessed by using area under the receiver operating curve. Results AI TI-RADS assigned new point values for eight ACR TI-RADS features. Six features were assigned zero points, which simplified categorization. By using expert reader data, the diagnostic performance of ACR TI-RADS and AI TI-RADS was area under the receiver operating curve of 0.91 and 0.93, respectively. For the same expert, specificity of AI TI-RADS (65%, 55 of 85) was higher (P < .001) than that of ACR TI-RADS (47%, 40 of 85). For the eight nonexpert radiologists, mean specificity for AI TI-RADS (55%) was also higher (P < .001) than that of ACR TI-RADS (48%). An interactive AI TI-RADS calculator can be viewed at http://deckard.duhs.duke.edu/∼ai-ti-rads . Conclusion An artificial intelligence-optimized Thyroid Imaging Reporting and Data System (TI-RADS) validates the American College of Radiology TI-RADS while slightly improving specificity and maintaining sensitivity. Additionally, it simplifies feature assignments, which may improve ease of use. © RSNA, 2019 Online supplemental material is available for this article.

Comprehensive Word-Level Classification of Screening Mammography Reports Using a Neural Network Sequence Labeling Approach.

Radiology reports contain a large amount of potentially valuable unstructured data. Recently, neural networks have been employed to perform classification of radiology reports over a few classes at the document level. The success of neural networks in sequence-labeling problems such as named entity recognition and part of speech tagging suggests that they could be used to classify radiology report text with greater granularity. We employed a neural network architecture to comprehensively classify mammography report text at the word level using a sequence labeling approach. Two radiologists devised a comprehensive classification system for screening mammography reports. Each word in each report was manually categorized by a radiologist into one of 33 categories according to the classification system. Tagged words referencing the same finding were grouped into unique sets. We pre-labeled reports with a rule-based algorithm and then manually edited these annotations for 6705 screening mammography reports (25.1%, 66.8%, and 8.1% BI-RADS 0, 1, and 2, respectively). A combined convolutional and recurrent neural network model was used to label words in each sentence of the individual reports. A siamese recurrent neural network was then used to group findings into sets. Performance of the neural network-based method was compared to a rule-based algorithm and a conditional random field (CRF) model. Global accuracy (percentage of documents where all word tags were predicted correctly) and keyword accuracy (percentage of all words that were labeled correctly, excluding words tagged as unimportant) were calculated on an unseen 519 report test set. Two-tailed t tests were used to assess differences between algorithm performance, and p < 0.05 was used to determine statistical significance. The neural network-based approach showed significantly higher global accuracy compared to both the rule-based algorithm (88.3 vs 57.0%, p < 0.001) and the CRF model (88.3% vs. 75.8%, p < 0.001). The neural network also showed significantly higher keyword level accuracy compared to the rule-based algorithm (95.5% vs. 80.9% p < 0.001) and CRF model (95.5% vs. 76.9%, p < 0.001). We demonstrate the potential of neural networks to accurately perform word-level multilabel classification of free text radiology reports across 33 classes, thus showing the utility of a sequence labeling approach to NLP of radiology reports. We found that a neural network classifier outperforms a rule-based algorithm and a CRF classifier for comprehensive multilabel classification of free text screening mammography reports at the word level. By approaching radiology report classification as a sequence-labeling problem, we demonstrate the ability of neural networks to extract data from free text radiology reports at a level of granularity not previously reported.

Case 260: Endobronchial Posttransplantation Lymphoproliferative Disease.

History A 31-year-old woman with a history of bilateral orthotopic lung transplantation performed 10 months earlier for cystic fibrosis presented for a routine follow-up appointment, with her chief symptom being a cough. The cough started approximately 1 month prior to this appointment and was minimally productive of clear to yellow phlegm. In addition to her cough, she reported increased sinus congestion and a sensation of "something in her upper chest." She denied shortness of breath, wheezing, hemoptysis, or cigarette smoking. Review of systems was negative for fever, chills, or night sweats. At physical examination, the patient was afebrile, borderline tachycardic (heart rate, 99 beats per minute), and mildly hypertensive (blood pressure, 138/99 mm Hg). Oxygen saturation was 96% on room air. Laboratory evaluation revealed a white blood cell count of 3.5 × 109/L (normal range, [3.2-9.8] × 109/L). Pulmonary function testing was notable for a newly decreased ratio of forced expiratory volume in 1 second (FEV1) to forced vital capacity (FVC) of 64% (2.0 and 3.4 L, respectively) (normal FEV1-to-FVC ratio, 80%), suggesting an obstructive lung process. One month prior to presentation, the patient's sputum cultures grew Pseudomonas and methicillin-resistant Staphylococcus aureus. The patient showed no evidence of active infection at the time of bronchoscopy. Thus, the bacteria were favored to reflect colonization, and antibiotic therapy was not administered at that time. The patient was taking an immunosuppression regimen of mycophenolate mofetil (CellCept; Genentech, San Francisco, Calif) (1 g twice daily), prednisone (10 mg daily), and tacrolimus (Prograf; Astellas Pharma US, Northbrook, Ill) (goal therapeutic range, 12-14 ng/mL). The patient was sent for posteroanterior and lateral chest radiography followed by chest CT ( Figs 1 - 3 ) and fluorine 18 (18F) fluorodeoxyglucose (FDG) PET/CT ( Fig 4 ).

Case 260.

History A 31-year-old woman with a history of bilateral orthotopic lung transplantation performed 10 months earlier for cystic fibrosis presented for a routine follow-up appointment, with her chief symptom being a cough. The cough started approximately 1 month prior to this appointment and was minimally productive of clear to yellow phlegm. In addition to her cough, she reported increased sinus congestion and a sensation of "something in her upper chest." She denied shortness of breath, wheezing, hemoptysis, or cigarette smoking. Review of systems was negative for fever, chills, or night sweats. At physical examination, the patient was afebrile, borderline tachycardic (heart rate, 99 beats per minute), and mildly hypertensive (blood pressure, 138/99 mm Hg). Oxygen saturation was 96% on room air. Laboratory evaluation revealed a white blood cell count of 3.5 × 109/L (normal range, 3.2-9.8 × 109/L). Pulmonary function testing was notable for a newly decreased ratio of forced expiratory volume in 1 second (FEV1) to forced vital capacity (FVC) of 64% (2.0 and 3.4 L, respectively) (normal FEV1-to-FVC ratio, 80%), suggesting an obstructive lung process. One month prior to presentation, the patient's sputum cultures grew Pseudomonas and methicillin-resistant Staphylococcus aureus. The patient showed no evidence of active infection at the time of bronchoscopy. Thus, the bacteria were favored to reflect colonization, and antibiotic therapy was not administered at that time. The patient was taking an immunosuppression regimen of mycophenalate mofetil (CellCept; Genentech, San Francisco, Calif) (1 g twice daily), prednisone (10 mg daily), and tacrolimus (Prograf; Astellas Pharma US, Northbrook, Ill) (goal therapeutic range, 12-14 ng/mL). The patient was sent for posteroanterior and lateral chest radiography followed by chest CT ( Figs 1 - 3 ) and fluorine 18 fluorodeoxyglucose PET/CT ( Fig 4 ). [Figure: see text][Figure: see text][Figure: see text][Figure: see text].

A Normal Thyroid by Any Other Name: Linguistic Analysis of Statements Describing a Normal Thyroid Gland from Noncontrast Chest CT Reports.

PURPOSE: To determine the variability and readability of language used in chest CT reports to describe a "normal" thyroid gland. METHODS: Using a semi-automated process, we identified unique sentences or phrases describing a "normal" thyroid gland in 11,357 noncontrast chest CT reports. Readability metrics were computed for each descriptor, including sentence length and grade-level indices quantifying the education necessary for text comprehension. Grade-level indices included the Flesch-Kincaid (FK) grade level, Gunning Fog index (GF), Coleman-Liau (CL) index, automated readability index (ARI), and a computed composite grade level (CGL) calculated as (FK + GF + CL + ARI)/4. RESULTS: There were 342 unique "normal" thyroid descriptors identified among 6,957 noncontrast chest CT reports characterizing a normal thyroid gland. For these 342 unique descriptors, sentence length varied 23-fold, with a mean sentence length of 8.3 ± 5.1 words. CGL varied 4-fold, with a mean of 16.4 ± 4.5, suggesting that descriptors for a normal thyroid gland, on average, require an advanced college-level education for comprehension. CONCLUSIONS: The language used by radiologists to describe a normal thyroid gland in chest CT reports is variable and complex. The linguistic characteristics observed herein may be a surrogate for the broader readability of radiology reports. With the growing role of the radiology report in patient communication, further linguistic analysis of reporting language may provide valuable insight for optimizing radiology communication.

Sequential CT Findings in Patients With Non-small-cell Lung Cancer Receiving Nivolumab.

BACKGROUND: Nivolumab is a novel immunotherapy that was recently approved for treatment of advanced non-small-cell lung cancer (NSCLC). Patients treated with checkpoint inhibitors may show variable computed tomography (CT) features on follow-up imaging, and it is unclear how reliable conventional response criteria are to determine patient management and outcomes. We report the spectrum of sequential CT findings in patients with advanced stage lung cancer who received nivolumab in an effort to better inform appropriate imaging strategies. MATERIALS AND METHODS: We identified all patients at our institution with advanced NSCLC who received nivolumab. Pre- and posttreatment CT scans were reviewed and categorized based on radiographic response to therapy. Demographic data as well as survival data were recorded. RESULTS: There were 34 patients with advanced NSCLC who received nivolumab with sufficient follow-up data. Nineteen patients were classified as responders to treatment; 6 (32%) of 19 showed improvement on their initial follow-up CT and had an average survival of 11.2 months, whereas 13 (68%) of 19 responders initially had stable or progressive disease on CT with an average survival of 11.6 months. Fifteen patients were classified as nonresponders to treatment with an average survival of 3.4 months. CONCLUSION: Novel immunotherapies such as nivolumab mechanistically differ from conventional chemotherapy. Some patients have improved survival despite initial radiographic progression of disease. Our findings underscore the heterogeneous radiographic appearance at follow-up CT in patients with lung cancer who ultimately respond to nivolumab.

Patient-Centered Radiology Reporting: Using Online Crowdsourcing to Assess the Effectiveness of a Web-Based Interactive Radiology Report.

PURPOSE: The aim of this study was to evaluate the effectiveness of a patient-centered web-based interactive mammography report. METHODS: A survey was distributed on Amazon Mechanical Turk, an online crowdsourcing platform. One hundred ninety-three US women ≥18 years of age were surveyed and then randomized to one of three simulated BI-RADS® 0 report formats: standard report, Mammography Quality Standards Act-modeled patient letter, or web-based interactive report. Survey questions assessed participants' report comprehension, satisfaction with and perception of the interpreting radiologist, and experience with the presented report. Two-tailed t tests and χ2 tests were used to evaluate differences among groups. RESULTS: Participants in the interactive web-based group spent more than double the time viewing the report than the standard report group (160.0 versus 64.2 seconds, P < .001). Report comprehension scores were significantly higher for the interactive web-based and patient letter groups than the standard report group (P < .05). Scores of satisfaction with the interpreting radiologist were significantly higher for the web-based interactive report and patient letter groups than the standard report group (P < .01). There were no significant differences between the patient letter and web-based interactive report groups. CONCLUSIONS: Radiology report format likely influences communication effectiveness. For result communication to a non-medical patient audience, patient-centric report formats, such as a Mammography Quality Standards Act-modeled patient letter or web-based interactive report, may offer advantages over the standard radiology report. Future work is needed to determine if these findings are reproducible in patient care settings and to determine how best to optimize radiology result communication to patients.

Potential of F-18 PET/CT in the Detection of Leptomeningeal Metastasis.

Leptomeningeal metastasis (LM) is a rare but increasingly common condition in which malignant cells migrate to the meninges. The gold standard for diagnosing LM is detection of cancer cells in the cerebrospinal fluid (CSF). Contrast enhanced-magnetic resonance imaging (CE-MRI) is also used to diagnose LM. We describe a case of LM in which CE-MRI of the neuroaxis was initially negative for meningeal enhancement but F-18 fluorodeoxyglucose positron-emission tomography/computed tomography (F-18 FDG PET/CT) revealed hypermetabolism within the lumbar spinal canal. Positive F-18 FDG PET findings have rarely been reported in LM and, to our knowledge, have never been reported in the context of initially negative CE-MRI scanning of the neuroaxis. F-18 FDG PET/CT may represent an alternative modality for diagnosing LM in patients who are unable to undergo CE-MRI and/or LP or in patients for whom initial CE-MRI and/or LP are negative for LM.

18F-FDG PET/CT in a 16-year-old patient with hydranencephaly.

A 16-year-old male patient with a past medical history of congenital brain malformation presented for surveillance F-FDG PET/CT scan of left parotid acinic cell adenocarcinoma. F-FDG PET/CT revealed absence of gray matter activity in the bilateral cerebral hemispheres most consistent with hydranencephaly. Hydranencephaly is a rare congenital condition characterized by absent cerebral hemispheres replaced by cerebrospinal fluid-filled sacs. The etiology is hypothesized to be secondary to intrauterine bilateral internal carotid artery compromise. Most affected individuals die in utero or within weeks of birth; however, there are rare reported cases of prolonged survival as in our patient.

Cross-sectional evidence of a signaling pathway from bone homeostasis to glucose metabolism.

CONTEXT: Preclinical studies suggested the existence of a signaling pathway connecting bone and glucose metabolisms. Supposedly leptin modulates osteocalcin bioactivity, which in turn stimulates insulin and adiponectin secretion, and ß-cell proliferation. OBJECTIVE: The objective of the investigation was to study the reciprocal relationships of adiponectin, leptin, osteocalcin, insulin resistance, and insulin secretion to verify whether such relationships are consistent with a signaling pathway connecting bone homeostasis and glucose metabolism. DESIGN: This was a cross-sectional analysis. SETTING: The study was conducted with community-dwelling volunteers participating in the Baltimore Longitudinal Study of Aging. PARTICIPANTS: Two hundred eighty women and 300 men with complete data on fasting plasma adiponectin, leptin, and osteocalcin, oral glucose tolerance test (plasma glucose and insulin values available at t = 0, 20, and 120 min), and anthropometric measures participated in the study. MAIN OUTCOME MEASURES: Linear regression models were used to test independent associations of adiponectin, osteocalcin, and leptin with the indices of insulin resistance and secretion. The expected reciprocal relationship between different biomarkers was verified by structural equation modeling. RESULTS: In linear regression models, leptin was strongly associated with indices of both insulin resistance and secretion. Both adiponectin and osteocalcin were negatively associated with insulin resistance. Structural equation modeling revealed a direct inverse association of leptin with osteocalcin; a direct positive association of osteocalcin with adiponectin; and an inverse relationship of osteocalcin with insulin resistance and adiponectin with insulin resistance and secretion, which is cumulatively consistent with the hypothesized model. CONCLUSIONS: Bone and glucose metabolisms are probably connected through a complex pathway that involves leptin, osteocalcin, and adiponectin. The clinical relevance of such a pathway for bone pathology in diabetes should be further investigated.

High-dose isoflavones do not improve metabolic and inflammatory parameters in androgen-deprived men with prostate cancer.

The profound hypogonadism that occurs with androgen deprivation therapy (ADT) for prostate cancer (PCa) results in complications such as diabetes and metabolic syndrome that predispose to cardiovascular disease. Because phytoestrogens have been associated with an improvement in metabolic parameters, we evaluated their role in men undergoing ADT. Our objective was to evaluate the effects of high-dose isoflavones on metabolic and inflammatory parameters in men undergoing ADT. This was a randomized, double-blind, placebo-controlled, 12-week pilot study. Participants were randomly assigned to receive 20 g of soy protein containing 160 mg of total isoflavones vs taste-matched placebo (20 g whole milk protein). The study was conducted at a tertiary care center in the United States. Thirty-three men (isoflavones = 17, placebo = 16) undergoing ADT for PCa completed this pilot study. Mean age in the 2 groups was 69 years and the majority of men were Caucasians. Mean duration of ADT in both groups was approximately 2 years (P = .70). The 2 groups were well matched at baseline. After 12 weeks of intervention, there was no significant difference in either metabolic or inflammatory parameters between the 2 groups. We found that high-dose isoflavones over a course of 12 weeks do not improve metabolic or inflammatory parameters in androgen-deprived men.

A cellular model of inflammation for identifying TNF-alpha synthesis inhibitors.

Neuroinflammation is a common facet of both acute and chronic neurodegenerative conditions, exemplified by stroke and by Alzheimer's and Parkinson's disease, and the presence of elevated levels of the proinflammatory cytokine, tumor necrosis factor-alpha (TNF-alpha), has been documented in each. Although initial TNF-alpha generation is associated with a protective compensatory response, its unregulated chronic elevation is generally detrimental and can drive the disease process. In such circumstances, therapeutic strategies that can both gain access to the brain and target the production of TNF-alpha are predicted to be of clinical benefit. An in vitro mouse macrophage-like cellular screen, utilizing RAW 264.7 cells, was hence developed to identify novel TNF-alpha lowering agents incorporating lipophilic physicochemical characteristics predicted to allow penetration of the blood-brain barrier. Cultured RAW 264.7 cells exposed to lipopolysaccharide (LPS) induced a rapid, marked and concentration-dependent cellular release of TNF-alpha into the cell culture media, which was readily detected by enzyme linked immunosorbent assay (ELISA). The effects of four characterized thalidomide-based TNF-alpha lowering agents were assessed alongside 10 novel uncharacterized compounds synthesized on the same backbone. One of these new analogs possessed activity of sufficient magnitude to warrant further investigation. Activity determined in the cellular model translated to an in vivo rodent model of acute LPS-induced TNF-alpha elevation. The utility of the TNF-alpha cellular assay lies in its simplicity and robust nature, providing a tool for initial pharmacological screening to allow for the rapid identification novel TNF-alpha lowering agents.