Hybrid U-Net-based deep learning model for volume segmentation of lung nodules in CT images

OBJECTIVE

Accurate segmentation of the lung nodule in computed tomography images is a critical component of a computer-assisted lung cancer detection/diagnosis system. However, lung nodule segmentation is a challenging task due to the heterogeneity of nodules. This study is to develop a hybrid deep learning (H-DL) model for the segmentation of lung nodules with a wide variety of sizes, shapes, margins, and opacities.

MATERIALS AND METHODS

A dataset collected from Lung Image Database Consortium image collection containing 847 cases with lung nodules manually annotated by at least two radiologists with nodule diameters greater than 7 mm and less than 45 mm was randomly split into 683 training/validation and 164 independent test cases. The 50% consensus consolidation of radiologists' annotation was used as the reference standard for each nodule. We designed a new H-DL model combining two deep convolutional neural networks (DCNNs) with different structures as encoders to increase the learning capabilities for the segmentation of complex lung nodules. Leveraging the basic symmetric U-shaped architecture of U-Net, we redesigned two new U-shaped deep learning (U-DL) models that were expanded to six levels of convolutional layers. One U-DL model used a shallow DCNN structure containing 16 convolutional layers adapted from the VGG-19 as the encoder, and the other used a deep DCNN structure containing 200 layers adapted from DenseNet-201 as the encoder, while the same decoder with only one convolutional layer at each level was used in both U-DL models, and we referred to them as the shallow and deep U-DL models. Finally, an ensemble layer was used to combine the two U-DL models into the H-DL model. We compared the effectiveness of the H-DL, the shallow U-DL and the deep U-DL models by deploying them separately to the test set. The accuracy of volume segmentation for each nodule was evaluated by the 3D Dice coefficient and Jaccard index (JI) relative to the reference standard. For comparison, we calculated the median and minimum of the 3D Dice and JI over the individual radiologists who segmented each nodule, referred to as M-Dice, min-Dice, M-JI, and min-JI.

RESULTS

For the 164 test cases with 327 nodules, our H-DL model achieved an average 3D Dice coefficient of 0.750 ± 0.135 and an average JI of 0.617 ± 0.159. The radiologists' average M-Dice was 0.778 ± 0.102, and the average M-JI was 0.651 ± 0.127; both were significantly higher than those achieved by the H-DL model (p < 0.05). The radiologists' average min-Dice (0.685 ± 0.139) and the average min-JI (0.537 ± 0.153) were significantly lower than those achieved by the H-DL model (p < 0.05). The results indicated that the H-DL model approached the average performance of radiologists and was superior to the radiologist whose manual segmentation had the min-Dice and min-JI. Moreover, the average Dice and average JI achieved by the H-DL model were significantly higher than those achieved by the individual shallow U-DL model (Dice of 0.745 ± 0.139, JI of 0.611 ± 0.161; p < 0.05) or the individual deep U-DL model alone (Dice of 0.739 ± 0.145, JI of 0.604 ± 0.163; p < 0.05).

CONCLUSION

Our newly developed H-DL model outperformed the individual shallow or deep U-DL models. The H-DL method combining multilevel features learned by both the shallow and deep DCNNs could achieve segmentation accuracy comparable to radiologists' segmentation for nodules with wide ranges of image characteristics.

Recursive Training Strategy for a Deep Learning Network for Segmentation of Pathology Nuclei With Incomplete Annotation

This study developed a recursive training strategy to train a deep learning model for nuclei detection and segmentation using incomplete annotation. A dataset of 141 H&E stained breast cancer pathologic images with incomplete annotation was randomly split into training/validation set and test set of 89 and 52 images, respectively. The positive training samples were extracted at each annotated cell and augmented with affine translation. The negative training samples were selected from the non-cellular regions free of nuclei using a histogram-based semi-automatic method. A U-Net model was initially trained by minimizing a custom loss function. After the first stage of training, the trained U-Net model was applied to the images in the training set in an inference mode. The U-Net segmented objects with high quality were selected by a semi-automated method. Combining the newly selected high quality objects with the annotated nuclei and the previously generated negative samples, the U-Net model was retrained recursively until the stopping criteria were satisfied. For the 52 test images, the U-Net trained with and without using our recursive training method achieved a sensitivity of 90.3% and 85.3% for nuclei detection, respectively. For nuclei segmentation, the average Dice coefficient and average Jaccard index were 0.831±0.213 and 0.750±0.217, 0.780±0.270 and 0.697±0.264, for U-Net with and without recursive training, respectively. The improvement achieved by our proposed method was statistically significant (P < 0.05). In conclusion, our recursive training method effectively enlarged the set of annotated objects for training the deep learning model and further improved the detection and segmentation performance.

Radiographic Graft Surveillance in Lung Transplantation: Prognostic Role of Parametric Response Mapping

RATIONALE

Chronic lung allograft dysfunction (CLAD) results in significant morbidity following lung transplantation. Potential CLAD occurs when lung function declines to 80-90% of baseline. Better non-invasive tools to prognosticate at potential CLAD are needed.

OBJECTIVES

To determine if parametric response mapping (PRM), a CT voxel-wise methodology, applied to high resolution CT scans can identify patients at risk of progression to CLAD or death.

METHODS

Radiographic features and PRM-based CT metrics quantifying functional small airways disease (PRMfSAD) and parenchymal disease (PRMPD) were studied at potential CLAD (n=61). High PRMfSAD and high PRMPD were defined as ≥ 30%. Restricted mean modeling was performed to compare CLAD-free survival among groups.

MEASUREMENTS AND MAIN RESULTS

PRM metrics identified 3 unique signatures: high PRMfSAD (11.5%), high PRMPD (41%) and neither (PRMNormal; 47.5%). Patients with high PRMfSAD or PRMPD had shorter CLAD-free median survival times (0.46 years and 0.50 years) compared to patients with predominantly PRMNormal (2.03 years; p=0.004 and 0.007 compared to PRMfSAD and PRMPD groups, respectively). In multivariate modeling adjusting for single versus double lung transplant, age at transplant, BMI at potential CLAD, and time from transplant to CT, PRMfSAD or PRMPD ≥ 30% continue to be statistically significant predictors of shorter CLAD-free survival. Air trapping by radiologist interpretation was common (66%), similar across PRM groups, and was not predictive of CLAD-free survival. Ground glass opacities by radiologist read occurred in 16% of cases and was associated with decreased CLAD-free survival (p<0.001).

CONCLUSIONS

PRM analysis offers valuable prognostic information at potential CLAD, identifying patients most at risk of developing CLAD or death.

Variabilities in Reference Standard by Radiologists and Performance Assessment in Detection of Pulmonary Embolism in CT Pulmonary Angiography

Annotating lesion locations by radiologists' manual marking is a key step to provide reference standard for the training and testing of a computer-aided detection system by supervised machine learning. Inter-reader variability is not uncommon in readings even by expert radiologists. This study evaluated the variability of the radiologist-identified pulmonary emboli (PEs) to demonstrate the importance of improving the reliability of the reference standard by a multi-step process for performance evaluation. In an initial reading of 40 CTPA PE cases, two experienced thoracic radiologists independently marked the PE locations. For markings from the two radiologists that did not agree, each radiologist re-read the cases independently to assess the discordant markings. Finally, for markings that still disagreed after the second reading, the two radiologists read together to reach a consensus. The variability of radiologists was evaluated by analyzing the agreement between two radiologists. For the 40 cases, 475 and 514 PEs were identified by radiologists R1 and R2 in the initial independent readings, respectively. For a total of 545 marks by the two radiologists, 81.5% (444/545) of the marks agreed but 101 marks in 36 cases differed. After consensus, 65 (64.4%) and 36 (35.6%) of the 101 marks were determined to be true PEs and false positives (FPs), respectively. Of these, 48 and 17 were false negatives (FNs) and 14 and 22 were FPs by R1 and R2, respectively. Our study demonstrated that there is substantial variability in reference standards provided by radiologists, which impacts the performance assessment of a lesion detection system. Combination of multiple radiologists' readings and consensus is needed to improve the reliability of a reference standard.

Improved detection of air trapping on expiratory computed tomography using deep learning

BACKGROUND

Radiologic evidence of air trapping (AT) on expiratory computed tomography (CT) scans is associated with early pulmonary dysfunction in patients with cystic fibrosis (CF). However, standard techniques for quantitative assessment of AT are highly variable, resulting in limited efficacy for monitoring disease progression.

OBJECTIVE

To investigate the effectiveness of a convolutional neural network (CNN) model for quantifying and monitoring AT, and to compare it with other quantitative AT measures obtained from threshold-based techniques.

MATERIALS AND METHODS

Paired volumetric whole lung inspiratory and expiratory CT scans were obtained at four time points (0, 3, 12 and 24 months) on 36 subjects with mild CF lung disease. A densely connected CNN (DN) was trained using AT segmentation maps generated from a personalized threshold-based method (PTM). Quantitative AT (QAT) values, presented as the relative volume of AT over the lungs, from the DN approach were compared to QAT values from the PTM method. Radiographic assessment, spirometric measures, and clinical scores were correlated to the DN QAT values using a linear mixed effects model.

RESULTS

QAT values from the DN were found to increase from 8.65% ± 1.38% to 21.38% ± 1.82%, respectively, over a two-year period. Comparison of CNN model results to intensity-based measures demonstrated a systematic drop in the Dice coefficient over time (decreased from 0.86 ± 0.03 to 0.45 ± 0.04). The trends observed in DN QAT values were consistent with clinical scores for AT, bronchiectasis, and mucus plugging. In addition, the DN approach was found to be less susceptible to variations in expiratory deflation levels than the threshold-based approach.

CONCLUSION

The CNN model effectively delineated AT on expiratory CT scans, which provides an automated and objective approach for assessing and monitoring AT in CF patients.

Abdominal CT in COVID-19 patients: incidence, indications, and findings

Purpose

The purpose of this study was to evaluate the frequency, indications, and findings of abdominal CTs ordered in the initial evaluation of patients who had a positive COVID-19 test performed in our institution.

Methods

Retrospective chart review was performed on all patients who had a positive test for COVID-19 performed at a single quaternary care center from 1/20/2020 through 5/8/2020. In a subset of patients undergoing abdominal CT as part of the initial evaluation, the demographics, suspected COVID-19 status at the time of scan, presenting complaints, and abdominal CT findings were recorded. Cardiothoracic radiologists reviewed and scored the visualized lung bases for the likelihood of COVID-19.

Results

Only 43 (4.1%) of 1057 COVID-19 patients presented with abdominal complaints sufficient to warrant an abdominal CT. Of these 43 patients, the vast majority (39, 91%) were known or suspected to have COVID-19 at the time of the scan. Most (27/43, 63%) scans showed no acute abdominal abnormality, and those that were positive did not share a discernable pattern of abnormalities. Lung base abnormalities were common, and there was moderate inter-reviewer reliability.

Conclusion

A minority of COVID-19 patients present with abdominal complaints sufficient to warrant a dedicated CT of the abdomen, and most of these studies will be negative or have abdominal findings not associated with COVID-19. Appropriate lung base findings are a more consistent indication of COVID-19 infection than abdominal findings.

Case Report: Successful Treatment of Pulmonary Talaromyces marneffei Infection with Posaconazole in a Renal Transplant Recipient

Talaromyces marneffei (T. marneffei), formerly Penicillium marneffei, is a dimorphic fungus prevalent in Southeast Asia that can cause severe systemic infection, especially in immunocompromised patients. There are few reports about the use of posaconazole in T. marneffei infection. Here, we present a case of pulmonary T. marneffei infection in a renal transplant recipient. The patient responded rapidly to oral posaconazole administration but experienced serum creatinine fluctuation because of the interaction between posaconazole and immunosuppressants. Seven months after adjusting the dose of immunosuppressants, the patient recovered completely. Posaconazole is a potentially promising therapy for T. marneffei infection, but it should be administered under close monitoring.

Pathologic categorization of lung nodules: Radiomic descriptors of CT attenuation distribution patterns of solid and subsolid nodules in low-dose CT

PURPOSE

Develop a quantitative image analysis method to characterize the heterogeneous patterns of nodule components for the classification of pathological categories of nodules.

MATERIALS AND METHODS

With IRB approval and permission of the National Lung Screening Trial (NLST) project, 103 subjects with low dose CT (LDCT) were used in this study. We developed a radiomic quantitative CT attenuation distribution descriptor (qADD) to characterize the heterogeneous patterns of nodule components and a hybrid model (qADD+) that combined qADD with subject demographic data and radiologist-provided nodule descriptors to differentiate aggressive tumors from indolent tumors or benign nodules with pathological categorization as reference standard. The classification performances of qADD and qADD + were evaluated and compared to the Brock and the Mayo Clinic models by analysis of the area under the receiver operating characteristic curve (AUC).

RESULTS

The radiomic features were consistently selected into qADDs to differentiate pathological invasive nodules from (1) preinvasive nodules, (2) benign nodules, and (3) the group of preinvasive and benign nodules, achieving test AUCs of 0.847 ± 0.002, 0.842 ± 0.002 and 0.810 ± 0.001, respectively. The qADD + obtained test AUCs of 0.867 ± 0.002, 0.888 ± 0.001 and 0.852 ± 0.001, respectively, which were higher than both the Brock and the Mayo Clinic models.

CONCLUSION

The pathologic invasiveness of lung tumors could be categorized according to the CT attenuation distribution patterns of the nodule components manifested on LDCT images, and the majority of invasive lung cancers could be identified at baseline LDCT scans.

Comparison of clinical and epidemiological characteristics of asymptomatic and symptomatic SARS-CoV-2 infection: A multi-center study in Sichuan Province, China

Objectives

Asymptomatic infection of SARS-CoV-2 has become a concern worldwide. This study aims to compare the epidemiology and the clinical characteristics of SARS-CoV-2 infection in asymptomatic and symptomatic individuals.

Methods

A total of 511 confirmed SARS-CoV-2 infection cases, including 100 asymptomatic (by the time of the pathogenic tests) and 411 symptomatic individuals were consecutively enrolled from January 25 to February 20, 2020 from hospitals in 21 cities and 47 counties or districts in Sichuan Province. Epidemiological and clinical characteristics were compared.

Results

Compared to the symptomatic patients, the asymptomatic cases were younger (P < 0.001), had similar co-morbidity percentages (P = 0.609), and came from higher altitude areas with lower population mobility (P < 0.001) with better defined epidemiological history (P < 0.001). 27.4% of well-documented asymptomatic cases developed delayed symptoms after the pathogenic diagnosis. 60% of asymptomatic cases demonstrated findings of pneumonia on the initial chest CT, including well-recognized features of coronavirus disease-19. None of the asymptomatic individuals died. Two elderly individuals with initially asymptomatic infection developed severe symptoms during hospitalization. One case of possible virus transmission by a patient during the incubation period was highly suspected.

Conclusions

The epidemiological and clinical findings highlight the significance of asymptomatic infection with SARS-CoV-2. Inspecting the epidemiological history would facilitate the identification of asymptomatic cases. Evidence supports the chest 10.13039/100004811CT scans for confirmed asymptomatic cases to evaluate the extent of lung involvement.

•

The asymptomatic cases were younger, with better defined epidemiological history.

•

60% of asymptomatic cases showed positive findings of pneumonia on the initial CT.

•

At least one case of possible transmission during incubation period was suspected.

Hypersensitivity Pneumonitis: Radiologic Phenotypes Are Associated With Distinct Survival Time and Pulmonary Function Trajectory

BACKGROUND

Hypersensitivity pneumonitis (HP) is an interstitial lung disease with a better prognosis, on average, than idiopathic pulmonary fibrosis (IPF). We compare survival time and pulmonary function trajectory in patients with HP and IPF by radiologic phenotype.

METHODS

HP (n = 117) was diagnosed if surgical/transbronchial lung biopsy, BAL, and exposure history results suggested this diagnosis. IPF (n = 152) was clinically and histopathologically diagnosed. All participants had a baseline high-resolution CT (HRCT) scan and FVC % predicted. Three thoracic radiologists documented radiologic features. Survival time is from HRCT scan to death or lung transplant. Cox proportional hazards models identify variables associated with survival time. Linear mixed models compare post-HRCT scan FVC % predicted trajectories.

RESULTS

Subjects were grouped by clinical diagnosis and three mutually exclusive radiologic phenotypes: honeycomb present, non-honeycomb fibrosis (traction bronchiectasis and reticulation) present, and nonfibrotic. Nonfibrotic HP had the longest event-free median survival (> 14.73 years) and improving FVC % predicted (1.92%; 95% CI, 0.49-3.35; P = .009). HP with non-honeycomb fibrosis had longer survival than IPF (> 7.95 vs 5.20 years), and both groups experienced a significant decline in FVC % predicted. Subjects with HP and IPF with honeycombing had poor survival (2.76 and 2.81 years, respectively) and significant decline in FVC % predicted.

CONCLUSIONS

Three prognostically distinct, radiologically defined phenotypes are identified among patients with HP. The importance of pursuing a specific diagnosis (eg, HP vs IPF) among patients with non-honeycomb fibrosis is highlighted. When radiologic honeycombing is present, invasive diagnostic testing directed at determining the diagnosis may be of limited value given a uniformly poor prognosis.

Development and validation of a radiological diagnosis model for hypersensitivity pneumonitis

High-resolution computed tomography (HRCT) may be useful for diagnosing hypersensitivity pneumonitis. Here, we develop and validate a radiological diagnosis model and model-based points score.Patients with interstitial lung disease seen at the University of Michigan Health System (derivation cohort) or enrolling in the Lung Tissue Research Consortium (validation cohort) were included. A thin-section, inspiratory HRCT scan was required. Thoracic radiologists documented radiological features.The derivation cohort comprised 356 subjects (33.9% hypersensitivity pneumonitis) and the validation cohort comprised 424 subjects (15.5% hypersensitivity pneumonitis). An age-, sex- and smoking status-adjusted logistic regression model identified extent of mosaic attenuation or air trapping greater than that of reticulation ("MA-AT>Reticulation"; OR 6.20, 95% CI 3.53-10.90; p<0.0001) and diffuse axial disease distribution (OR 2.33, 95% CI 1.31-4.16; p=0.004) as hypersensitivity pneumonitis predictors (area under the receiver operating characteristic curve 0.814). A model-based score >2 (1 point for axial distribution, 2 points for "MA-AT>Reticulation") has specificity 90% and positive predictive value (PPV) 74% in the derivation cohort and specificity 96% and PPV 44% in the validation cohort. Similar model performance is seen with population restriction to those reporting no exposure (score >2: specificity 91%).When radiological mosaic attenuation or air trapping are more extensive than reticulation and disease has diffuse axial distribution, hypersensitivity pneumonitis specificity is high and false diagnosis risk low (<10%), but PPV is diminished in a low-prevalence setting.

Coronary artery analysis: Computer-assisted selection of best-quality segments in multiple-phase coronary CT angiography

PURPOSE

The authors are developing an automated method to identify the best-quality coronary arterial segment from multiple-phase coronary CT angiography (cCTA) acquisitions, which may be used by either interpreting physicians or computer-aided detection systems to optimally and efficiently utilize the diagnostic information available in multiple-phase cCTA for the detection of coronary artery disease.

METHODS

After initialization with a manually identified seed point, each coronary artery tree is automatically extracted from multiple cCTA phases using our multiscale coronary artery response enhancement and 3D rolling balloon region growing vessel segmentation and tracking method. The coronary artery trees from multiple phases are then aligned by a global registration using an affine transformation with quadratic terms and nonlinear simplex optimization, followed by a local registration using a cubic B-spline method with fast localized optimization. The corresponding coronary arteries among the available phases are identified using a recursive coronary segment matching method. Each of the identified vessel segments is transformed by the curved planar reformation (CPR) method. Four features are extracted from each corresponding segment as quality indicators in the original computed tomography volume and the straightened CPR volume, and each quality indicator is used as a voting classifier for the arterial segment. A weighted voting ensemble (WVE) classifier is designed to combine the votes of the four voting classifiers for each corresponding segment. The segment with the highest WVE vote is then selected as the best-quality segment. In this study, the training and test sets consisted of 6 and 20 cCTA cases, respectively, each with 6 phases, containing a total of 156 cCTA volumes and 312 coronary artery trees. An observer preference study was also conducted with one expert cardiothoracic radiologist and four nonradiologist readers to visually rank vessel segment quality. The performance of our automated method was evaluated by comparing the automatically identified best-quality segments identified by the computer to those selected by the observers.

RESULTS

For the 20 test cases, 254 groups of corresponding vessel segments were identified after multiple phase registration and recursive matching. The AI-BQ segments agreed with the radiologist's top 2 ranked segments in 78.3% of the 254 groups (Cohen's kappa 0.60), and with the 4 nonradiologist observers in 76.8%, 84.3%, 83.9%, and 85.8% of the 254 groups. In addition, 89.4% of the AI-BQ segments agreed with at least two observers' top 2 rankings, and 96.5% agreed with at least one observer's top 2 rankings. In comparison, agreement between the four observers' top ranked segment and the radiologist's top 2 ranked segments were 79.9%, 80.7%, 82.3%, and 76.8%, respectively, with kappa values ranging from 0.56 to 0.68.

CONCLUSIONS

The performance of our automated method for selecting the best-quality coronary segments from a multiple-phase cCTA acquisition was comparable to the selection made by human observers. This study demonstrates the potential usefulness of the automated method in clinical practice, enabling interpreting physicians to fully utilize the best available information in cCTA for diagnosis of coronary disease, without requiring manual search through the multiple phases and minimizing the variability in image phase selection for evaluation of coronary artery segments across the diversity of human readers with variations in expertise.

How to perform a critical appraisal of diagnostic tests: 7 steps

The critically appraised topic (CAT) is a format in evidence-based practice for sharing information. A CAT is a standardized way of summarizing the most current research evidence focused on a pertinent clinical question. Its aim is to provide both a critique of the most up-to-date retrieved research and an indication of the clinical relevance of results. A clinical question is initially generated following a patient encounter, which leads to and directs a literature search to answer the clinical question. Studies obtained from the literature search are assigned a level of evidence. This allows the most valid and relevant articles to be selected and to be critically appraised. The results are summarized, and this information is translated into clinically useful procedures and processes.

Computerized detection of noncalcified plaques in coronary CT angiography: evaluation of topological soft gradient prescreening method and luminal analysis

PURPOSE

The buildup of noncalcified plaques (NCPs) that are vulnerable to rupture in coronary arteries is a risk for myocardial infarction. Interpretation of coronary CT angiography (cCTA) to search for NCP is a challenging task for radiologists due to the low CT number of NCP, the large number of coronary arteries, and multiple phase CT acquisition. The authors conducted a preliminary study to develop machine learning method for automated detection of NCPs in cCTA.

METHODS

With IRB approval, a data set of 83 ECG-gated contrast enhanced cCTA scans with 120 NCPs was collected retrospectively from patient files. A multiscale coronary artery response and rolling balloon region growing (MSCAR-RBG) method was applied to each cCTA volume to extract the coronary arterial trees. Each extracted vessel was reformatted to a straightened volume composed of cCTA slices perpendicular to the vessel centerline. A topological soft-gradient (TSG) detection method was developed to prescreen for NCP candidates by analyzing the 2D topological features of the radial gradient field surface along the vessel wall. The NCP candidates were then characterized by a luminal analysis that used 3D geometric features to quantify the shape information and gray-level features to evaluate the density of the NCP candidates. With machine learning techniques, useful features were identified and combined into an NCP score to differentiate true NCPs from false positives (FPs). To evaluate the effectiveness of the image analysis methods, the authors performed tenfold cross-validation with the available data set. Receiver operating characteristic (ROC) analysis was used to assess the classification performance of individual features and the NCP score. The overall detection performance was estimated by free response ROC (FROC) analysis.

RESULTS

With our TSG prescreening method, a prescreening sensitivity of 92.5% (111/120) was achieved with a total of 1181 FPs (14.2 FPs/scan). On average, six features were selected during the tenfold cross-validation training. The average area under the ROC curve (AUC) value for training was 0.87 ± 0.01 and the AUC value for validation was 0.85 ± 0.01. Using the NCP score, FROC analysis of the validation set showed that the FP rates were reduced to 3.16, 1.90, and 1.39 FPs/scan at sensitivities of 90%, 80%, and 70%, respectively.

CONCLUSIONS

The topological soft-gradient prescreening method in combination with the luminal analysis for FP reduction was effective for detection of NCPs in cCTA, including NCPs causing positive or negative vessel remodeling. The accuracy of vessel segmentation, tracking, and centerline identification has a strong impact on NCP detection. Studies are underway to further improve these techniques and reduce the FPs of the CADe system.

Computerized analysis of coronary artery disease: performance evaluation of segmentation and tracking of coronary arteries in CT angiograms

PURPOSE

The authors are developing a computer-aided detection system to assist radiologists in analysis of coronary artery disease in coronary CT angiograms (cCTA). This study evaluated the accuracy of the authors' coronary artery segmentation and tracking method which are the essential steps to define the search space for the detection of atherosclerotic plaques.

METHODS

The heart region in cCTA is segmented and the vascular structures are enhanced using the authors' multiscale coronary artery response (MSCAR) method that performed 3D multiscale filtering and analysis of the eigenvalues of Hessian matrices. Starting from seed points at the origins of the left and right coronary arteries, a 3D rolling balloon region growing (RBG) method that adapts to the local vessel size segmented and tracked each of the coronary arteries and identifies the branches along the tracked vessels. The branches are queued and subsequently tracked until the queue is exhausted. With Institutional Review Board approval, 62 cCTA were collected retrospectively from the authors' patient files. Three experienced cardiothoracic radiologists manually tracked and marked center points of the coronary arteries as reference standard following the 17-segment model that includes clinically significant coronary arteries. Two radiologists visually examined the computer-segmented vessels and marked the mistakenly tracked veins and noisy structures as false positives (FPs). For the 62 cases, the radiologists marked a total of 10191 center points on 865 visible coronary artery segments.

RESULTS

The computer-segmented vessels overlapped with 83.6% (8520/10191) of the center points. Relative to the 865 radiologist-marked segments, the sensitivity reached 91.9% (795/865) if a true positive is defined as a computer-segmented vessel that overlapped with at least 10% of the reference center points marked on the segment. When the overlap threshold is increased to 50% and 100%, the sensitivities were 86.2% and 53.4%, respectively. For the 62 test cases, a total of 55 FPs were identified by radiologist in 23 of the cases.

CONCLUSIONS

The authors' MSCAR-RBG method achieved high sensitivity for coronary artery segmentation and tracking. Studies are underway to further improve the accuracy for the arterial segments affected by motion artifacts, severe calcified and noncalcified soft plaques, and to reduce the false tracking of the veins and other noisy structures. Methods are also being developed to detect coronary artery disease along the tracked vessels.

Coronary CT angiography (cCTA): automated registration of coronary arterial trees from multiple phases

Coronary computed tomography angiography (cCTA) is a commonly used imaging modality for the evaluation of coronary artery disease. cCTA is generally reconstructed in multiple cardiac phases because different coronary arteries may be better visualized in some phases than in others due to the periodic cardiac motion. We are developing an automated registration method for coronary arterial trees from multiple-phase cCTA that has potential application in building a 'best-quality' tree to facilitate image analysis and detection of stenotic plaques. Given the segmented left or right coronary arterial (LCA or RCA) trees from the multiple phases as input, the adjacent phase pairs, where displacements are relatively small, are registered by a specifically designed method based on a cubic B-spline with fast localized optimization (CBSO). For the phase pairs with large displacements, a global registration using an affine transform with quadratic terms and nonlinear simplex optimization (AQSO) is followed by a local registration using CBSO to refine the AQSO registered volumes. 26 LCA and 26 RCA trees with six cCTA phases from 26 patients were used for registration evaluation. The average distances for the tree pairs between the adjacent phases with small displacements before and after CBSO registration were 0.96  ±  0.79 and 0.76  ±  0.61 mm respectively for LCA, and 0.93  ±  0.97 and 0.64  ±  0.43 mm, respectively for RCA. The average distance differences before and after registration were statistically significant (p < 0.001) for both LCA and RCA trees. The average distances for the distant phases with large displacements before registration, after AQSO registration, and finally after the CBSO registration were 2.85  ±  1.46, 1.62  ±  0.76, and 0.97  ±  0.43 mm, respectively for LCA, and 4.03  ±  2.36, 2.18  ±  1.11, and 0.97  ±  0.44 mm, respectively for RCA. The average distance differences between every two consecutive stages of registration were statistically significant. The corresponding phases of LCA and RCA trees were aligned to an average of less than 1 mm, providing a basis for a best-quality tree construction.

Automated iterative neutrosophic lung segmentation for image analysis in thoracic computed tomography

PURPOSE

Lung segmentation is a fundamental step in many image analysis applications for lung diseases and abnormalities in thoracic computed tomography (CT). The authors have previously developed a lung segmentation method based on expectation-maximization (EM) analysis and morphological operations (EMM) for our computer-aided detection (CAD) system for pulmonary embolism (PE) in CT pulmonary angiography (CTPA). However, due to the large variations in pathology that may be present in thoracic CT images, it is difficult to extract the lung regions accurately, especially when the lung parenchyma contains extensive lung diseases. The purpose of this study is to develop a new method that can provide accurate lung segmentation, including those affected by lung diseases.

METHODS

An iterative neutrosophic lung segmentation (INLS) method was developed to improve the EMM segmentation utilizing the anatomic features of the ribs and lungs. The initial lung regions (ILRs) were extracted using our previously developed EMM method, in which the ribs were extracted using 3D hierarchical EM segmentation and the ribcage was constructed using morphological operations. Based on the anatomic features of ribs and lungs, the initial EMM segmentation was refined using INLS to obtain the final lung regions. In the INLS method, the anatomic features were mapped into a neutrosophic domain, and the neutrosophic operation was performed iteratively to refine the ILRs. With IRB approval, 5 and 58 CTPA scans were collected retrospectively and used as training and test sets, of which 2 and 34 cases had lung diseases, respectively. The lung regions manually outlined by an experienced thoracic radiologist were used as reference standard for performance evaluation of the automated lung segmentation. The percentage overlap area (POA), the Hausdorff distance (Hdist), and the average distance (AvgDist) of the lung boundaries relative to the reference standard were used as performance metrics.

RESULTS

The proposed method achieved larger POAs and smaller distance errors than the EMM method. For the 58 test cases, the average POA, Hdist, and AvgDist were improved from 85.4±18.4%, 22.6±29.4 mm, and 3.5±5.4 mm using EMM to 91.2±6.7%, 16.0±11.3 mm, and 2.5±1.0 mm using INLS, respectively. The improvements were statistically significant (p<0.05). To evaluate the accuracy of the INLS method in the identification of the lung boundaries affected by lung diseases, the authors separately analyzed the performance of the proposed method on the cases with versus without the lung diseases. The results showed that the cases without lung diseases were segmented more accurately than the cases with lung diseases by both the EMM and the INLS methods, but the INLS method achieved better performance than the EMM method in both cases.

CONCLUSIONS

The new INLS method utilizing the anatomic features of the rib and lung significantly improved the accuracy of lung segmentation, especially for the cases affected by lung diseases. Improvement in lung segmentation will facilitate many image analysis tasks and CAD applications for lung diseases and abnormalities in thoracic CT, including automated PE detection.

Automated coronary artery tree extraction in coronary CT angiography using a multiscale enhancement and dynamic balloon tracking (MSCAR-DBT) method

RATIONAL AND OBJECTIVES

To evaluate our prototype method for segmentation and tracking of the coronary arterial tree, which is the foundation for a computer-aided detection (CADe) system to be developed to assist radiologists in detecting non-calcified plaques in coronary CT angiography (cCTA) scans.

MATERIALS AND METHODS

The heart region was first extracted by a morphological operation and an adaptive thresholding method based on expectation-maximization (EM) estimation. The vascular structures within the heart region were enhanced and segmented using a multiscale coronary response (MSCAR) method that combined 3D multiscale filtering, analysis of the eigenvalues of Hessian matrices and EM estimation segmentation. After the segmentation of vascular structures, the coronary arteries were tracked by a 3D dynamic balloon tracking (DBT) method. The DBT method started at two manually identified seed points located at the origins of the left and right coronary arteries (LCA and RCA) for extraction of the arterial trees. The coronary arterial trees of a data set containing 20 ECG-gated contrast-enhanced cCTA scans were extracted by our MSCAR-DBT method and a clinical GE Advantage workstation. Two experienced thoracic radiologists visually examined the coronary arteries on the original cCTA scans and the rendered volume of segmented vessels to count the untracked false-negative (FN) segments and false positives (FPs) for both methods.

RESULTS

For the visible coronary arterial segments in the 20 cases, the radiologists identified that 25 segments were missed by our MSCAR-DBT method, ranging from 0 to 5 FN segments in individual cases, and that 55 artery segments were missed by the GE software, ranging from 0 to 7 FN segments in individual cases. 19 and 15 FPs were identified in our and the GE coronary trees, ranging from 0 to 4 FPs for both methods in individual cases, respectively.

CONCLUSION

The preliminary study demonstrates the feasibility of our MSCAR-DBT method for segmentation and tracking coronary artery trees. The results indicated that both our method and GE software can extract coronary artery trees reasonably well and the performance of our method is superior to that of GE software in this small data set. Further studies are underway to develop methods for improvement of the segmentation and tracking accuracy.

Development and validation of a heart atlas to study cardiac exposure to radiation following treatment for breast cancer

PURPOSE

Cardiac toxicity is an important sequela of breast radiotherapy. However, the relationship between dose to cardiac structures and subsequent toxicity has not been well defined, partially due to variations in substructure delineation, which can lead to inconsistent dose reporting and the failure to detect potential correlations. Here we have developed a heart atlas and evaluated its effect on contour accuracy and concordance.

METHODS AND MATERIALS

A detailed cardiac computed tomography scan atlas was developed jointly by cardiology, cardiac radiology, and radiation oncology. Seven radiation oncologists were recruited to delineate the whole heart, left main and left anterior descending interventricular branches, and right coronary arteries on four cases before and after studying the atlas. Contour accuracy was assessed by percent overlap with gold standard atlas volumes. The concordance index was also calculated. Standard radiation fields were applied. Doses to observer-contoured cardiac structures were calculated and compared with gold standard contour doses. Pre- and post-atlas values were analyzed using a paired t test.

RESULTS

The cardiac atlas significantly improved contour accuracy and concordance. Percent overlap and concordance index of observer-contoured cardiac and gold standard volumes were 2.3-fold improved for all structures (p < 0.002). After application of the atlas, reported mean doses to the whole heart, left main artery, left anterior descending interventricular branch, and right coronary artery were within 0.1, 0.9, 2.6, and 0.6 Gy, respectively, of gold standard doses.

CONCLUSIONS

This validated University of Michigan cardiac atlas may serve as a useful tool in future studies assessing cardiac toxicity and in clinical trials which include dose volume constraints to the heart.

Computer-aided diagnosis of lung nodules on CT scans: ROC study of its effect on radiologists' performance

RATIONALE AND OBJECTIVES

The aim of this study was to evaluate the effect of computer-aided diagnosis (CAD) on radiologists' estimates of the likelihood of malignancy of lung nodules on computed tomographic (CT) imaging.

METHODS AND MATERIALS

A total of 256 lung nodules (124 malignant, 132 benign) were retrospectively collected from the thoracic CT scans of 152 patients. An automated CAD system was developed to characterize and provide malignancy ratings for lung nodules on CT volumetric images. An observer study was conducted using receiver-operating characteristic analysis to evaluate the effect of CAD on radiologists' characterization of lung nodules. Six fellowship-trained thoracic radiologists served as readers. The readers rated the likelihood of malignancy on a scale of 0% to 100% and recommended appropriate action first without CAD and then with CAD. The observer ratings were analyzed using the Dorfman-Berbaum-Metz multireader, multicase method.

RESULTS

The CAD system achieved a test area under the receiver-operating characteristic curve (A(z)) of 0.857 +/- 0.023 using the perimeter, two nodule radii measures, two texture features, and two gradient field features. All six radiologists obtained improved performance with CAD. The average A(z) of the radiologists improved significantly (P < .01) from 0.833 (range, 0.817-0.847) to 0.853 (range, 0.834-0.887).

CONCLUSION

CAD has the potential to increase radiologists' accuracy in assessing the likelihood of malignancy of lung nodules on CT imaging.

Computer-aided diagnosis of pulmonary nodules on CT scans: improvement of classification performance with nodule surface features

The purpose of this work is to develop a computer-aided diagnosis (CAD) system to differentiate malignant and benign lung nodules on CT scans. A fully automated system was designed to segment the nodule from its surrounding structured background in a local volume of interest (VOI) and to extract image features for classification. Image segmentation was performed with a 3D active contour method. The initial contour was obtained as the boundary of a binary object generated by k-means clustering within the VOI and smoothed by morphological opening. A data set of 256 lung nodules (124 malignant and 132 benign) from 152 patients was used in this study. In addition to morphological and texture features, the authors designed new nodule surface features to characterize the lung nodule surface smoothness and shape irregularity. The effects of two demographic features, age and gender, as adjunct to the image features were also investigated. A linear discriminant analysis (LDA) classifier built with features from stepwise feature selection was trained using simplex optimization to select the most effective features. A two-loop leave-one-out resampling scheme was developed to reduce the optimistic bias in estimating the test performance of the CAD system. The area under the receiver operating characteristic curve, A(z), for the test cases improved significantly (p < 0.05) from 0.821 +/- 0.026 to 0.857 +/- 0.023 when the newly developed image features were included with the original morphological and texture features. A similar experiment performed on the data set restricted to primary cancers and benign nodules, excluding the metastatic cancers, also resulted in an improved test A(z), though the improvement did not reach statistical significance (p = 0.07). The two demographic features did not significantly affect the performance of the CAD system (p > 0.05) when they were added to the feature space containing the morphological, texture, and new gradient field and radius features. To investigate if a support vector machine (SVM) classifier can achieve improved performance over the LDA classifier, we compared the performance of the LDA and SVMs with various kernels and parameters. Principal component analysis was used to reduce the dimensionality of the feature space for both the LDA and the SVM classifiers. When the number of selected principal components was varied, the highest test A(z) among the SVMs of various kernels and parameters was slightly higher than that of the LDA in one-loop leave-one-case-out resampling. However, no SVM with fixed architecture consistently performed better than the LDA in the range of principal components selected. This study demonstrated that the authors' proposed segmentation and feature extraction techniques are promising for classifying lung nodules on CT images.

Computer-aided detection of pulmonary embolism in computed tomographic pulmonary angiography (CTPA): performance evaluation with independent data sets

The authors are developing a computer-aided detection system for pulmonary emboli (PE) in computed tomographic pulmonary angiography (CTPA) scans. The pulmonary vessel tree is extracted using a 3D expectation-maximization segmentation method based on the analysis of eigen-values of Hessian matrices at multiple scales. A parallel multiprescreening method is applied to the segmented vessels to identify volume of interests (VOIs) that contained suspicious PE. A linear discriminant analysis (LDA) classifier with feature selection is designed to reduce false positives (FPs). Features that characterize the contrast, gray level, and size of PE are extracted as input predictor variables to the LDA classifier. With the IRB approval, 59 CTPA PE cases were collected retrospectively from the patient files (UM cases). With access permission, 69 CTPA PE cases were randomly selected from the data set of the prospective investigation of pulmonary embolism diagnosis (PIOPED) II clinical trial. Extensive lung parenchymal or pleural diseases were present in 22/59 UM and 26/69 PIOPED cases. Experienced thoracic radiologists manually marked 595 and 800 PE as the reference standards in the UM and PIOPED data sets, respectively. PE occlusion of arteries ranged from 5% to 100%, with PE located from the main pulmonary artery to the subsegmental artery levels. Of the 595 PE identified in the UM cases, 245 and 350 PE were located in the subsegmental arteries and the more proximal arteries, respectively. The detection performance was assessed by free response ROC (FROC) analysis. The FROC analysis indicated that the PE detection system could achieve an overall sensitivity of 80% at 18.9 FPs/case for the PIOPED cases when the LDA classifier was trained with the UM cases. The test sensitivity with the UM cases was 80% at 22.6 FPs/cases when the LDA classifier was trained with the PIOPED cases. The detection performance depended on the arterial level where the PE was located and on the percentage of occlusion. The sensitivity was lower for PE in the subsegmental arteries than in more proximal arteries and was lower for PE with less than 20% occlusion. The results indicate that the PE detection system achieves high sensitivity for PE detection on independent CTPA scans for both the PIOPED and UM data sets and demonstrate the potential that the automated PE detection approach can be generalized to unknown cases.

Multidetector CT of thoracic aortic aneurysms

Thoracic aortic aneurysms (TAAs) can be broadly divided into true aneurysms and false aneurysms (pseudoaneurysms). True aneurysms contain all three layers of the aortic wall (intima, media, and adventitia), whereas false aneurysms have fewer than three layers and are contained by the adventitia or periadventitial tissues. Multidetector computed tomographic (CT) angiography allows the comprehensive evaluation of TAAs in terms of morphologic features and extent, presence of thrombus, relationship to adjacent structures and branches, and signs of impending or acute rupture, and is routinely used in this setting. Knowledge of the causes, significance, imaging appearances, and potential complications of both common and uncommon aortic aneurysms, as well as of the normal postoperative appearance of the thoracic aorta, is essential for prompt and accurate diagnosis. Supplemental material available at http://radiographics.rsnajnls.org/cgi/content/full/29/2/537/DC1.

Hemodynamic and functional assessment of mechanical aortic valves using combined echocardiography and multidetector computed tomography

BACKGROUND

Limitations are found in the ability of transthoracic echocardiography to evaluate mechanical aortic valve replacements (AVR). We evaluated the ability of combined echocardiography and computed tomography (CT) to enhance the hemodynamic and functional evaluation of AVR.

METHODS

We performed a retrospective evaluation of 41 consecutive patients with AVR (27 bileaflet, 14 single disc) and both transthoracic echocardiography and 64-detector electrocardiographic-gated CT. Each study was interpreted by 2 independent, blinded readers. The effective orifice area was compared with the corrected energy-loss coefficient area and the geometric orifice area. Patients with an elevated mean pressure gradient (>15 mm Hg) were assessed for potential abnormal findings, including patient-prosthesis mismatch, elevated cardiac index, valve dysfunction, significant regurgitation, or pressure recovery effect.

RESULTS

Significant differences (P<0.05) and moderate-to-high correlations (r=0.55-0.98) were observed between the effective orifice area (2.2+/-0.8 cm(2)), corrected energy-loss coefficient area (3.0+/-1.5 cm(2)), and geometric orifice area (3.6+/-0.9 cm(2)). At least one abnormality was observed in 7 of 25 patients with normal gradients and in 14 of 16 patients with elevated gradients (P<0.001). In 16 patients with elevated mean pressure gradient, a potential cause could be determined in 4 with echocardiography alone and in 14 patients with combined echocardiography and CT (P=0.001).

CONCLUSION

CT aids in the interrogation of prosthetic valve function, enhancing evaluation for patient prosthesis mismatch, and correction for pressure recovery by the corrected energy-loss coefficient. CT is additive to the assessment of mechanical AVR with transthoracic echocardiography, and the combination permits a more complete assessment of both AVR function and hemodynamics.

Effect of CT scanning parameters on volumetric measurements of pulmonary nodules by 3D active contour segmentation: a phantom study

The purpose of this study is to investigate the effects of CT scanning and reconstruction parameters on automated segmentation and volumetric measurements of nodules in CT images. Phantom nodules of known sizes were used so that segmentation accuracy could be quantified in comparison to ground-truth volumes. Spherical nodules having 4.8, 9.5 and 16 mm diameters and 50 and 100 mg cc(-1) calcium contents were embedded in lung-tissue-simulating foam which was inserted in the thoracic cavity of a chest section phantom. CT scans of the phantom were acquired with a 16-slice scanner at various tube currents, pitches, fields-of-view and slice thicknesses. Scans were also taken using identical techniques either within the same day or five months apart for study of reproducibility. The phantom nodules were segmented with a three-dimensional active contour (3DAC) model that we previously developed for use on patient nodules. The percentage volume errors relative to the ground-truth volumes were estimated under the various imaging conditions. There was no statistically significant difference in volume error for repeated CT scans or scans taken with techniques where only pitch, field of view, or tube current (mA) were changed. However, the slice thickness significantly (p < 0.05) affected the volume error. Therefore, to evaluate nodule growth, consistent imaging conditions and high resolution should be used for acquisition of the serial CT scans, especially for smaller nodules. Understanding the effects of scanning and reconstruction parameters on volume measurements by 3DAC allows better interpretation of data and assessment of growth. Tracking nodule growth with computerized segmentation methods would reduce inter- and intraobserver variabilities.

CT and MRI of acute thoracic cardiovascular emergencies

A wide spectrum of acute cardiovascular disorders is seen in patients who are hospitalized in a critical care setting. Imaging plays a central role in the diagnosis and management of these conditions. The most frequently used imaging remains chest radiography; however, more advanced modalities, including coronary angiography, echocardiography, and radioisotope scintigraphy, have well established roles in the assessment of patients in the critical care setting. More recently, multidetector row CT (MDCT) and MRI are being used increasingly for evaluation of coronary artery disease, cardiac structure and function, coronary artery anomalies, cardiac masses, pericardial disease, valvular disease, postoperative cardiovascular abnormalities, venous thromboembolism and acute aortic syndromes, often with other ancillary findings that can provide important clinical information. The three most common life-threatening cardiovascular processes in which advanced imaging plays a role, particularly CT, are discussed, including pulmonary embolism, aortic dissection, and coronary artery disease.

Automatic multiscale enhancement and segmentation of pulmonary vessels in CT pulmonary angiography images for CAD applications

The authors are developing a computerized pulmonary vessel segmentation method for a computer-aided pulmonary embolism (PE) detection system on computed tomographic pulmonary angiography (CTPA) images. Because PE only occurs inside pulmonary arteries, an automatic and accurate segmentation of the pulmonary vessels in 3D CTPA images is an essential step for the PE CAD system. To segment the pulmonary vessels within the lung, the lung regions are first extracted using expectation-maximization (EM) analysis and morphological operations. The authors developed a 3D multiscale filtering technique to enhance the pulmonary vascular structures based on the analysis of eigenvalues of the Hessian matrix at multiple scales. A new response function of the filter was designed to enhance all vascular structures including the vessel bifurcations and suppress nonvessel structures such as the lymphoid tissues surrounding the vessels. An EM estimation is then used to segment the vascular structures by extracting the high response voxels at each scale. The vessel tree is finally reconstructed by integrating the segmented vessels at all scales based on a "connected component" analysis. Two CTPA cases containing PEs were used to evaluate the performance of the system. One of these two cases also contained pleural effusion disease. Two experienced thoracic radiologists provided the gold standard of pulmonary vessels including both arteries and veins by manually tracking the arterial tree and marking the center of the vessels using a computer graphical user interface. The accuracy of vessel tree segmentation was evaluated by the percentage of the "gold standard" vessel center points overlapping with the segmented vessels. The results show that 96.2% (2398/2494) and 96.3% (1910/1984) of the manually marked center points in the arteries overlapped with segmented vessels for the case without and with other lung diseases. For the manually marked center points in all vessels including arteries and veins, the segmentation accuracy are 97.0% (4546/4689) and 93.8% (4439/4732) for the cases without and with other lung diseases, respectively. Because of the lack of ground truth for the vessels, in addition to quantitative evaluation of the vessel segmentation performance, visual inspection was conducted to evaluate the segmentation. The results demonstrate that vessel segmentation using our method can extract the pulmonary vessels accurately and is not degraded by PE occlusion to the vessels in these test cases.

Metastatic shoulder liposarcoma to the right ventricle: CT findings

Metastatic cardiac liposarcoma is rare. A right ventricular liposarcoma metastasis is described in a 46-year-old man, who was admitted with significant shortness of breath and fatigue, and in whom a large lobulated low attenuation mass occupying most of the right ventricular cavity, with extension through the right ventricular apex and a small-to-moderate pericardial effusion was detected by electrocardiogram-gated cardiac computed tomography. The patient had an antecedent history of a left upper arm liposarcoma treated with surgical resection, chemotherapy, and postoperative radiotherapy 3 years earlier. Surgical resection was performed with the majority of the neoplasm removed though; the right ventricular apex and epicardial extension of tumor could not be fully resected. The histopathologic analysis revealed a liposarcoma, similar to the one resected in the left arm 3 years earlier. Electrocardiogram-gated cardiac computed tomography was able to visualize the metastatic tumor within the heart, accurately evaluate cardiac function and allow for prompt surgical treatment that produced relief of symptoms, and assess for further metastatic disease within the thorax.

Cardiac pseudosarcomatous fibromyxoid tumor: a review of the literature

A cardiac pseudosarcomatous fibromyxoid tumor (PFT) is described in a previously healthy 35-year-old man, together with a review of the literature. Pseudosarcomatous fibromyxoid tumor is within the spectrum of inflammatory (myofibroblastic) pseudotumors. It has previously been described predominantly within the genitourinary tract and respiratory tract. Inflammatory pseudotumor is rare as a cardiac tumor, and cardiac PFT is not previously reported. No recurrence or metastatic disease has been reported after resection of PFTs elsewhere in the body, and this tumor seems to have a benign clinical course.

Plasma cell granuloma: maxillary sinuses

Inflammatory pseudotumours or plasma cell granulomas are space-occupying, rounded masses appearing in various locations usually as an aftermath of recurrent infections. Recurrent pain and mass have been the most common presentations (Durst et al., 1977). These have been confused with tumours but radical surgical treatment is unnecessary, in the absence of life-threatening complications. The prognosis is excellent after excision. The nature and pathogenesis of the inflammatory pseudotumours, their presentation as fever and anaemia in some cases, are unknown. This report is probably the first case of a plasma cell granuloma involving the maxillary sinus causing focal erosion of the orbital wall, simulating a malignant tumour clinically.