A deep-learning method for the denoising of ultra-low dose chest CT in coronary artery calcium score evaluation.

AIM: To evaluate a novel deep-learning denoising method for ultra-low dose CT (ULDCT) in the assessment of coronary artery calcium score (CACS). MATERIALS AND METHODS: Sixty adult patients who underwent two unenhanced chest CT examinations, a normal dose CT (NDCT) and an ULDCT, were enrolled prospectively between September 2017 to December 201. A special training set was created to learn the characteristics of the real noise affecting the ULDCT implementing a fully convolutional neural network with batch normalisation. Subsequently, the 60 ULDCTs of the evaluation set were denoised. Two blinded radiologists assessed the NDCT, ULDCT, and denoised-ULDCT (DULDCT), assigning a CACS and categorised each scan as having a score above or below 100 and presence of calcifications (score 0 versus >0). Statistical analysis was used to evaluate the agreement between the readers and differences in CACSs between each imaging method. RESULTS: After excluding one patient, the cohort included 59 patients (median age 67 years, 58% men). The ULDCT median effective radiation dose (ERD) was 0.172 mSv, which was 2.8% of the NDCT median ERD. Denoising improved the signal-to-noise ratio by 27.7% (p<0.001). Interobserver agreement was almost perfect between readers (intraclass correlation coefficient >0.993). CACSs were lower for ULDCT and DULDCT as compared to the NDCT (p ≤ 0.001). In differentiating between the presence and absence of coronary artery calcifications, DULDCT showed greater accuracy (98-100%) and positive likelihood ratio (14.29->99) compared to ULDCT (92% and 2.78, respectively). CONCLUSION: DULCT significantly reduced the image noise and better identified patients with no coronary artery calcifications than native ULDCT.

Incidental primary breast cancer detected on PET-CT.

The purpose of this study is to determine the incidence of primary breast cancer (PBC) detected on (18) F-fluorodeoxyglucose (FDG) positron emission tomography (PET)-computed tomography (CT) in patients with a known diagnosis of non-mammary malignancies. A database search was performed to identify patients with non-mammary malignancies who had undergone staging with FDG PET-CT at a single institution between September 2005 and September 2011 and with the word "breast" reported in the PET-CT dictation. Additional breast imaging studies, clinical data, and the final histopathology of the breast lesions were correlated with the PET-CT images. Of 1951 patients who underwent PET/CT, 440 incidental breast lesions were identified in 438 patients. Of these 440 lesions, 195 (45 %) were benign, 160 (37 %) malignant, and 85 (19 %) missing data. A total of 25 PBCs were diagnosed; with a median size of 1.8 cm (range 0.8-10.7 cm); and a median SUVmax of 4.4 (range 1.7-17.6). There were 19 invasive ductal cancers, 1 invasive lobular cancer, 2 papillary cancers, 1 tubular cancer, 1 sarcomatoid cancer, and 1 ductal carcinoma in situ. Eight patients had regional nodal disease. Mammography revealed the PBC in 19 of 23 tumors (83 %), sonography in 22 of 23 (96 %). Six percent (25 of 440) of incidental breast lesions identified on FDG PET-CT represent PBCs; more than half were at an early stage and potentially curable.

Consistency of response and image recognition, pulmonary nodules.

OBJECTIVE: To investigate the effect of recognition of a previously encountered radiograph on consistency of response in localized pulmonary nodules. METHODS: 13 radiologists interpreted 40 radiographs each to locate pulmonary nodules. A few days later, they again interpreted 40 radiographs. Half of the images in the second set were new. We asked the radiologists whether each image had been in the first set. We used Fisher's exact test and Kruskal-Wallis test to evaluate the correlation between recognition of an image and consistency in its interpretation. We evaluated the data using all possible recognition levels-definitely, probably or possibly included vs definitely, probably or possibly not included by collapsing the recognition levels into two and by eliminating the "possibly included" and "possibly not included" scores. RESULTS: With all but one of six methods of looking at the data, there was no significant correlation between consistency in interpretation and recognition of the image. When the possibly included and possibly not included scores were eliminated, there was a borderline statistical significance (p = 0.04) with slightly greater consistency in interpretation of recognized than that of non-recognized images. CONCLUSION: We found no convincing evidence that radiologists' recognition of images in an observer performance study affects their interpretation on a second encounter. ADVANCES IN KNOWLEDGE: Conscious recognition of chest radiographs did not result in a greater degree of consistency in the tested interpretation than that in the interpretation of images that were not recognized.

On the choice of acceptance radius in free-response observer performance studies.

OBJECTIVES: Choosing an acceptance radius or proximity criterion is necessary to analyse free-response receiver operating characteristic (FROC) observer performance data. This is currently subjective, with little guidance in the literature about what is an appropriate acceptance radius. We evaluated varying acceptance radii in a nodule detection task in chest radiography and suggest guidelines for determining an acceptance radius. METHODS: 80 chest radiographs were chosen, half of which contained nodules. We determined each nodule's centre. 21 radiologists read the images. We created acceptance radii bins of <5 pixels, <10 pixels, <20 pixels and onwards up to <200 and 200+ pixels. We counted lesion localisations in each bin and visually compared marks with the borders of nodules. RESULTS: Most reader marks were tightly clustered around nodule centres, with tighter clustering for smaller than for larger nodules. At least 70% of readers' marks were placed within <10 pixels for small nodules, <20 pixels for medium nodules and <30 pixels for large nodules. Of 72 inspected marks that were less than 50 pixels from the centre of a nodule, only 1 fell outside the border of a nodule. CONCLUSION: The acceptance radius should be based on the larger nodule sizes. For our data, an acceptance radius of 50 pixels would have captured all but 2 reader marks within the borders of a nodule, while excluding only 1 true-positive mark. The choice of an acceptance radius for FROC analysis of observer performance studies should be based on the size of larger abnormalities.

Quantifying the clinical relevance of a laboratory observer performance paradigm.

OBJECTIVE: Laboratory observer performance measurements, receiver operating characteristic (ROC) and free-response ROC (FROC) differ from actual clinical interpretations in several respects, which could compromise their clinical relevance. The objective of this study was to develop a method for quantifying the clinical relevance of a laboratory paradigm and apply it to compare the ROC and FROC paradigms in a nodule detection task. METHODS: The original prospective interpretations of 80 digital chest radiographs were classified by the truth panel as correct (C=1) or incorrect (C=0), depending on correlation with additional imaging, and the average of C was interpreted as the clinical figure of merit. FROC data were acquired for 21 radiologists and ROC data were inferred using the highest ratings. The areas under the ROC and alternative FROC curves were used as laboratory figures of merit. Bootstrap analysis was conducted to estimate conventional agreement measures between laboratory and clinical figures of merit. Also computed was a pseudovalue-based image-level correctness measure of the laboratory interpretations, whose association with C as measured by the area (rAUC) under an appropriately defined relevance ROC curve, is as a measure of the clinical relevance of a laboratory paradigm. RESULTS: Low correlations (e.g. κ=0.244) and near chance level rAUC values (e.g. 0.598), attributable to differences between the clinical and laboratory paradigms, were observed. The absolute width of the confidence interval was 0.38 for the interparadigm differences of the conventional measures and 0.14 for the difference of the rAUCs. CONCLUSION: The rAUC measure was consistent with the traditional measures but was more sensitive to the differences in clinical relevance. A new relevance ROC method for quantifying the clinical relevance of a laboratory paradigm is proposed.

The reversed halo sign: update and differential diagnosis.

The reversed halo sign is characterised by a central ground-glass opacity surrounded by denser air-space consolidation in the shape of a crescent or a ring. It was first described on high-resolution CT as being specific for cryptogenic organising pneumonia. Since then, the reversed halo sign has been reported in association with a wide range of pulmonary diseases, including invasive pulmonary fungal infections, paracoccidioidomycosis, pneumocystis pneumonia, tuberculosis, community-acquired pneumonia, lymphomatoid granulomatosis, Wegener granulomatosis, lipoid pneumonia and sarcoidosis. It is also seen in pulmonary neoplasms and infarction, and following radiation therapy and radiofrequency ablation of pulmonary malignancies. In this article, we present the spectrum of neoplastic and non-neoplastic diseases that may show the reversed halo sign and offer helpful clues for assisting in the differential diagnosis. By integrating the patient's clinical history with the presence of the reversed halo sign and other accompanying radiological findings, the radiologist should be able to narrow the differential diagnosis substantially, and may be able to provide a presumptive final diagnosis, which may obviate the need for biopsy in selected cases, especially in the immunosuppressed population.

Reperfusion edema after lung transplantation: effect of daclizumab.

PURPOSE: To determine if daclizumab, an interleukin-2 antagonist, reduced the severity of reperfusion edema in lung transplant recipients. MATERIALS AND METHODS: Eighty-five patients who were to undergo 86 consecutive lung transplants were included; 43 (50%) received daclizumab in addition to conventional immunosuppression. Patients were assigned to one of the following groups: control, right allograft; control, left allograft; daclizumab treated, right allograft; daclizumab treated, left allograft. Radiographs obtained in the first 5 postoperative days were evaluated for degree of edema. Mean daily edema scores and curves for control and daclizumab-treated groups were compared. Differences in survival at 1, 3, 6, and 12 months after transplantation, days of mechanical ventilation, and the ratio of arterial oxygenation to inspired oxygen level at 1, 3, and 5 days after transplantation were also compared. RESULTS: Mean daily edema scores, edema curves, survival, days of mechanical ventilation, and ratio of arterial oxygenation to inspired oxygen level at 1 and 3 days after transplantation did not significantly differ between daclizumab-treated and control groups. A trend toward improved survival in the daclizumab-treated group was noted. CONCLUSION: Daclizumab had no effect on the radiographic or immediate clinical manifestations of reperfusion edema in lung transplant recipients. Additional follow-up is needed to determine if daclizumab offers any long-term benefit in terms of reduced rejection rates or survival.

Correlation of FDG-PET imaging with Glut-1 and Glut-3 expression in early-stage non-small cell lung cancer.

PURPOSE: To correlate FDG activity on PET with the expression of glucose transporter proteins Glut-1 and Glut-3 in patients with early stage non-small cell lung cancer (NSCLC). METHODS: Over a 5 year period, all patients with a PET scan and clinical stage I NSCLC underwent an immunohistochemical analysis of their tumor for Glut-1 and Glut-3 expression. The amount of FDG uptake in the primary lesion was measured by a standardized uptake ratio (SUR) and correlated with immunohistochemical results. RESULTS: Seventy-three patients with a mean age of 66 years had clinical stage I disease. The final pathologic stage showed 64 patients with stage IA/B disease, eight with stage IIA disease, and one patient with pathologic stage IIIA (T1N2) disease. Glut-1 transporter expression was significantly higher than Glut-3 (P<0.0001), and although there was some association between the SUR and Glut-1 (P=0.085) and SUR and Glut-3 (P=0.074) expression, this did not reach statistical significance. CONCLUSIONS: Glut-1 and Glut-3 transporter expression did not demonstrate a statistically significant correlation with FDG uptake in potentially resectable lung cancer. It appears that these transporters alone do not affect the variation in FDG activity in early stage NSCLC.

Lung cancer and positron emission tomography with fluorodeoxyglucose.

Over the past years, positron emission tomography (PET) with fluoro-2-deoxy-D-glucose (FDG) has emerged as an important imaging modality. In the thorax, FDG-PET has been shown to differentiate benign from malignant pulmonary lesions and stage lung cancer. Preliminary studies have shown its usefulness in assessing tumor recurrence, and assisting in radiotherapy planning. FDG-PET is often more accurate than conventional imaging studies, and has been proven to be cost-effective in evaluating lung cancer patients. This review will discuss the current applications of FDG-PET as compared with conventional imaging in diagnosing, staging, and following patients with lung cancer.

Cystic fibrosis: usefulness of thoracic CT in the examination of patients before lung transplantation.

PURPOSE: To evaluate the usefulness of thoracic computed tomography (CT) in the pre-lung transplantation examination of patients with cystic fibrosis (CF). MATERIALS AND METHODS: Fifty-six patients (age range, 12-42 years) with CF were evaluated for possible lung transplantation from 1991 to 1997. Twenty-six of these patients underwent bilateral lung transplantation, 19 were awaiting transplantation at the time of the study, seven died before transplantation, and four were excluded for psychosocial concerns. Preoperative chest radiographic and CT findings were reviewed and correlated with clinical, operative, and pathology records. RESULTS: In seven patients, discrete, 1-2-cm pulmonary nodules were detected at CT. Five of these patients underwent transplantation; the nodules were found to be mucous impactions. No malignancy was found in any of the patients who underwent transplantation. Pretransplantation sputum cultures grew Aspergillus fumigatus in seven patients, none of whom had radiologic findings suggestive of Aspergillus infection. Radiographic or CT findings were suggestive of mycetoma in five cases, but no such tumors were found at transplantation. The accuracies of chest radiography and CT for the detection of pleural disease in 48 hemithoraces were 81% (n = 39) and 69% (n = 33), respectively. The radiologic findings of pleural thickening did not influence the surgical approach in any patient. CONCLUSION: Thoracic CT has little utility in the routine pre-lung transplantation examination of patients with CF.

Radiologic findings of bronchogenic carcinoma with pulmonary metastases at presentation.

PURPOSE: To describe the radiologic findings in patients with primary bronchogenic carcinoma and pulmonary metastases at presentation. MATERIALS AND METHODS: A retrospective review of patients with bronchogenic carcinoma who at presentation had pulmonary metastases. RESULTS: Fourteen (52%) men and 13 (48%) women with a mean age of 60 years were identified. Adenocarcinoma was the most common histology (70%). The number of nodules varied, although 78% of patients had greater than 50 nodules. Nodules size ranged from 2 to 30 mm, but 82% of patients had nodules less than 10 mm in diameter. Mediastinal lymphadenopathy was seen in 41% of patients, and pleural disease in 44% of patients. Only 37% had radiologic evidence of extrathoracic disease, with bone metastases (30%) being the most common. CONCLUSION: Multiple pulmonary nodules may be the presenting thoracic manifestation of primary bronchogenic carcinoma, with patterns of metastases and survival rates similar to other stage IV patients.

Staging non-small cell lung cancer with whole-body PET.

PURPOSE: To compare the accuracies of whole-body 2-[fluorine 18]fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET) and conventional imaging (thoracic computed tomography [CT], bone scintigraphy, and brain CT or magnetic resonance [MR] imaging) in staging bronchogenic carcinoma. MATERIALS AND METHODS: Within 20 months, 100 patients with newly diagnosed bronchogenic carcinoma underwent whole-body FDG PET and chest CT. Ninety of these patients underwent radionuclide bone scintigraphy, and 70 patients underwent brain CT or MR imaging. For each patient, all examinations were completed within 1 month. A radiologic stage was assigned by using PET and conventional imaging independently and was compared with the pathologic stage. The accuracy, sensitivity, specificity, and negative and positive predictive values were calculated. RESULTS: PET staging was accurate in 83 (83%) patients; conventional imaging staging was accurate in 65 (65%) patients (P < .005). Staging with mediastinal lymph nodes was correct by using PET in 67 (85%) patients and by using CT in 46 (58%) patients (P < .001). Nine (9%) patients had metastases demonstrated by using PET that were not found with conventional imaging, whereas 10 (10%) patients suspected of having metastases because of conventional imaging findings were correctly shown with PET to not have metastases. CONCLUSION: Whole-body PET was more accurate than thoracic CT, bone scintigraphy, and brain CT or MR imaging in staging bronchogenic carcinoma.

Lung cancer staging and management: comparison of contrast-enhanced and nonenhanced helical CT of the thorax.

PURPOSE: To determine whether contrast material-enhanced helical computed tomography (CT) of the thorax and upper abdomen changes the tumor stage and management compared with nonenhanced helical CT in patients with newly diagnosed lung cancer. MATERIALS AND METHODS: During 15 months, any patient in whom lung cancer was strongly suspected or newly diagnosed and who was scheduled for thoracic CT was considered eligible for the study. All patients underwent nonenhanced thoracic helical CT from the lung apices through the adrenal glands and then contrast-enhanced thoracic helical CT from the lung apices through the entire liver. Each study was read independently, and the thoracic radiologic stage was determined. Tissue sampling was performed and the final pathologic stage assigned. RESULTS: Ninety-six patients had a final pathologic diagnosis of lung cancer. There was agreement in stage between the nonenhanced and contrast-enhanced examinations in 92 of the 96 patients. In three patients, the tumor stage at nonehanced CT increased at contrast-enhanced CT, from IA to IIA (n = 1), IIB to IV (n = 1), and IIIB to IV (n = 1). In one patient, the tumor stage decreased from IIIB to IIB. There was no substantial change in management of any patient. CONCLUSION: The results suggest that contrast-enhanced thoracic CT through the liver for staging lung cancer rarely changes the tumor stage determined with nonenhanced CT through the adrenal glands and does not substantially influence management decisions.

Fibrolipomatous hamartoma: pathognomonic on MR imaging.

PURPOSE: To assess the MR imaging characteristics, presenting symptoms, age and nerve distribution of fibrolipomatous hamartoma. DESIGN: A computer search was performed of the term fibrolipomatous hamartoma through the musculoskeletal section MR imaging results at our institution from June 7, 1996 to January 21, 1998 followed by a search of the terms lipomatous hamartoma, median nerve, surrounding fat, increased fatty signal, coaxial, and neuroma. MR images and medical files were retrospectively reviewed by two experienced musculoskeletal radiologists for imaging characteristics, nerve and age distribution as well as for history of trauma. In addition three consultation cases from outside institutions were added for determination of image characteristics. RESULTS: Ten fibrolipomatous hamartomas were identified: eight in the median nerve, one in the ulnar nerve and one in the sciatic nerve. Mean age was 32.3 years (range 4-75 years, SD 21 years). Imaging characteristics were serpiginous low-intensity structures representing thickened nerve fascicles, surrounded by evenly distributed fat, high signal intensity on T1-weighted sequences and low signal intensity on T2-weighted sequences. The amount of fat varied; however, distribution in eight cases (80%) was predominantly between nerve fibers rather than surrounding them peripherally. All had a coaxial-cable-like appearance on axial planes and a spaghetti-like appearance on coronal planes that was not seen in any other type of median nerve abnormality imaged during the study period. CONCLUSION: The MR imaging characteristics of fibrolipomatous hamartoma are pathognomonic, obviating the need for biopsy for diagnosis.

Fine-needle aspiration cytology of "ancient" schwannoma.

The term "ancient" schwannoma was proposed for a group of neural tumors showing degenerative changes and marked nuclear atypia. Prior to the realization that the observed atypia was a regressive phenomenon, many of these lesions were erroneously diagnosed as sarcomas. Fine-needle aspiration (FNA) cytologic material from five patients is included in this study. Tissue examined histologically included four resected tumors and 18 gauge core biopsies of one tumor. Aspirates of ancient schwannoma showed many of the same features as FNA of regular schwannoma: aggregates of spindled cells with indistinct cytoplasm and elongate nuclei with blunt point ends. The feature unique to these lesions was nuclear pleomorphism, which was identified in all aspirates. Nuclear inclusions were identified in all but one case. Cystic degeneration, xanthomatous changes, and perivascular sclerosis were identified in excised lesions. Ancient schwannomas show most of the FNA features of benign schwannomas but can demonstrate marked nuclear atypia. The FNA features of ancient schwannoma are important to note because of the potential to confuse this lesion with a more serious one such as sarcoma on FNA.

The many faces of pulmonary aspiration.

The clinical and radiologic manifestations of aspiration are truly protean. The diagnosis is straightforward in patients with known risk factors and new radiographic findings in dependent portions of lung. However, diagnosis is difficult in patients without known risk factors or who present with less specific radiologic findings. Careful history-taking skills and knowledge of the diverse radiologic appearances of aspiration are thus required to facilitate diagnosis and expedite treatment.