Neural networks for the analysis of small pulmonary nodules

Harnessing the Power of Artificial Intelligence in Otolaryngology and the Communication Sciences

Use of artificial intelligence (AI) is a burgeoning field in otolaryngology and the communication sciences. A virtual symposium on the topic was convened from Duke University on October 26, 2020, and was attended by more than 170 participants worldwide. This review presents summaries of all but one of the talks presented during the symposium; recordings of all the talks, along with the discussions for the talks, are available at https://www.youtube.com/watch?v=ktfewrXvEFg and https://www.youtube.com/watch?v=-gQ5qX2v3rg . Each of the summaries is about 2500 words in length and each summary includes two figures. This level of detail far exceeds the brief summaries presented in traditional reviews and thus provides a more-informed glimpse into the power and diversity of current AI applications in otolaryngology and the communication sciences and how to harness that power for future applications.

Machine Learning in an Elderly Man with Heart Failure

Machine learning is a branch of artificial intelligence and can be used to predict important outcomes in a wide variety of medical conditions. With the widespread use of electronic medical records, the vast amount of data required for this process is now readily available. The following case demonstrates the application of machine learning to an elderly man with heart failure. The algorithms used, namely, decision tree and random forest, both correctly differentiated heart failure with preserved ejection fraction from heart failure with reduced ejection fraction. This has important treatment and prognostic ramifications and can be completed at the point of care while awaiting confirmation via echocardiogram. Viewing the machine learning process through a patient-centered lens, as in this case, highlights the key role we as physicians have in the implementation and supervision of machine learning.

The Regimen of Computed Tomography Screening for Lung Cancer: Lessons Learned Over 25 Years From the International Early Lung Cancer Action Program

We learned many unanticipated and valuable lessons since we started planning our study of low-dose computed tomography (CT) screening for lung cancer in 1991. The publication of the baseline results of the Early Lung Cancer Action Project (ELCAP) in Lancet 1999 showed that CT screening could identify a high proportion of early, curable lung cancers. This stimulated large national screening studies to be quickly started. The ELCAP design, which provided evidence about screening in the context of a clinical program, was able to rapidly expand to a 12-institution study in New York State (NY-ELCAP) and to many international institutions (International-ELCAP), ultimately working with 82 institutions, all using the common I-ELCAP protocol. This expansion was possible because the investigators had developed the ELCAP Management System for screening, capturing data and CT images, and providing for quality assurance. This advanced registry and its rapid accumulation of data and images allowed continual assessment and updating of the regimen of screening as advances in knowledge and new technology emerged. For example, in the initial ELCAP study, introduction of helical CT scanners had allowed imaging of the entire lungs in a single breath, but the images were obtained in 10 mm increments resulting in about 30 images per person. Today, images are obtained in submillimeter slice thickness, resulting in around 700 images per person, which are viewed on high-resolution monitors. The regimen provides the imaging acquisition parameters, imaging interpretation, definition of positive result, and the recommendations for further workup, which now include identification of emphysema and coronary artery calcifications. Continual updating is critical to maximize the benefit of screening and to minimize potential harms. Insights were gained about the natural history of lung cancers, identification and management of nodule subtypes, increased understanding of nodule imaging and pathologic features, and measurement variability inherent in CT scanners. The registry also provides the foundation for assessment of new statistical techniques, including artificial intelligence, and integration of effective genomic and blood-based biomarkers, as they are developed.

Artificial Intelligence for the Otolaryngologist: A State of the Art Review

Objective

To provide a state of the art review of artificial intelligence (AI), including its subfields of machine learning and natural language processing, as it applies to otolaryngology and to discuss current applications, future impact, and limitations of these technologies.

Data Sources

PubMed and Medline search engines.

Review Methods

A structured search of the current literature was performed (up to and including September 2018). Search terms related to topics of AI in otolaryngology were identified and queried to identify relevant articles.

Conclusions

AI is at the forefront of conversation in academic research and popular culture. In recent years, it has been touted for its potential to revolutionize health care delivery. Yet, to date, it has made few contributions to actual medical practice or patient care. Future adoption of AI technologies in otolaryngology practice may be hindered by misconceptions of what AI is and a fear that machine errors may compromise patient care. However, with potential clinical and economic benefits, it is vital for otolaryngologists to understand the principles and scope of AI.

Implications for Practice

In the coming years, AI is likely to have a major impact on biomedical research and the practice of medicine. Otolaryngologists are key stakeholders in the development and clinical integration of meaningful AI technologies that will improve patient care. High-quality data collection is essential for the development of AI technologies, and otolaryngologists should seek opportunities to collaborate with data scientists to guide them toward the most impactful clinical questions.

Prediction models for solitary pulmonary nodules based on curvelet textural features and clinical parameters

Lung cancer, one of the leading causes of cancer-related deaths, usually appears as solitary pulmonary nodules (SPNs) which are hard to diagnose using the naked eye. In this paper, curvelet-based textural features and clinical parameters are used with three prediction models [a multilevel model, a least absolute shrinkage and selection operator (LASSO) regression method, and a support vector machine (SVM)] to improve the diagnosis of benign and malignant SPNs. Dimensionality reduction of the original curvelet-based textural features was achieved using principal component analysis. In addition, non-conditional logistical regression was used to find clinical predictors among demographic parameters and morphological features. The results showed that, combined with 11 clinical predictors, the accuracy rates using 12 principal components were higher than those using the original curvelet-based textural features. To evaluate the models, 10-fold cross validation and back substitution were applied. The results obtained, respectively, were 0.8549 and 0.9221 for the LASSO method, 0.9443 and 0.9831 for SVM, and 0.8722 and 0.9722 for the multilevel model. All in all, it was found that using curvelet-based textural features after dimensionality reduction and using clinical predictors, the highest accuracy rate was achieved with SVM. The method may be used as an auxiliary tool to differentiate between benign and malignant SPNs in CT images.

ACR appropriateness Criteria® radiographically detected solitary pulmonary nodule

The solitary pulmonary nodule (SPN) is a common medical problem for which management can be quite complex. Imaging remains at the center of management of SPNs, and computed tomography is the primary modality by which SPNs are characterized and followed up for stability. This manuscript summarizes the American College of Radiology Appropriateness Criteria for radiographically detected solitary pulmonary nodules and briefly reviews the various imaging techniques available. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed every 2 years by a multidisciplinary expert panel. The guideline development and review include an extensive analysis of current medical literature from peer reviewed journals and the application of a well-established consensus methodology (modified Delphi) to rate the appropriateness of imaging and treatment procedures by the panel. In those instances in which evidence is lacking or not definitive, expert opinion may be used to recommend imaging or treatment.

Quantifying tumour heterogeneity with CT

Heterogeneity is a key feature of malignancy associated with adverse tumour biology. Quantifying heterogeneity could provide a useful non-invasive imaging biomarker. Heterogeneity on computed tomography (CT) can be quantified using texture analysis which extracts spatial information from CT images (unenhanced, contrast-enhanced and derived images such as CT perfusion) that may not be perceptible to the naked eye. The main components of texture analysis can be categorized into image transformation and quantification. Image transformation filters the conventional image into its basic components (spatial, frequency, etc.) to produce derived subimages. Texture quantification techniques include structural-, model- (fractal dimensions), statistical- and frequency-based methods. The underlying tumour biology that CT texture analysis may reflect includes (but is not limited to) tumour hypoxia and angiogenesis. Emerging studies show that CT texture analysis has the potential to be a useful adjunct in clinical oncologic imaging, providing important information about tumour characterization, prognosis and treatment prediction and response.

Computer-aided diagnosis systems for lung cancer: challenges and methodologies

This paper overviews one of the most important, interesting, and challenging problems in oncology, the problem of lung cancer diagnosis. Developing an effective computer-aided diagnosis (CAD) system for lung cancer is of great clinical importance and can increase the patient's chance of survival. For this reason, CAD systems for lung cancer have been investigated in a huge number of research studies. A typical CAD system for lung cancer diagnosis is composed of four main processing steps: segmentation of the lung fields, detection of nodules inside the lung fields, segmentation of the detected nodules, and diagnosis of the nodules as benign or malignant. This paper overviews the current state-of-the-art techniques that have been developed to implement each of these CAD processing steps. For each technique, various aspects of technical issues, implemented methodologies, training and testing databases, and validation methods, as well as achieved performances, are described. In addition, the paper addresses several challenges that researchers face in each implementation step and outlines the strengths and drawbacks of the existing approaches for lung cancer CAD systems.

Texture analysis of non-small cell lung cancer on unenhanced computed tomography: initial evidence for a relationship with tumour glucose metabolism and stage

The aim was to undertake an initial study of the relationship between texture features in computed tomography (CT) images of non-small cell lung cancer (NSCLC) and tumour glucose metabolism and stage. This retrospective pilot study comprised 17 patients with 18 pathologically confirmed NSCLC. Non-contrast-enhanced CT images of the primary pulmonary lesions underwent texture analysis in 2 stages as follows: (a) image filtration using Laplacian of Gaussian filter to differentially highlight fine to coarse textures, followed by (b) texture quantification using mean grey intensity (MGI), entropy (E) and uniformity (U) parameters. Texture parameters were compared with tumour fluorodeoxyglucose (FDG) uptake (standardised uptake value (SUV)) and stage as determined by the clinical report of the CT and FDG-positron emission tomography imaging. Tumour SUVs ranged between 2.8 and 10.4. The number of NSCLC with tumour stages I, II, III and IV were 4, 4, 4 and 6, respectively. Coarse texture features correlated with tumour SUV (E: r = 0.51, p = 0.03; U: r = −0.52, p = 0.03), whereas fine texture features correlated with tumour stage (MGI: r_s = 0.71, p = 0.001; E: r_s = 0.55, p = 0.02; U: r_s = −0.49, p = 0.04). Fine texture predicted tumour stage with a kappa of 0.7, demonstrating 100% sensitivity and 87.5% specificity for detecting tumours above stage II ( p = 0.0001). This study provides initial evidence for a relationship between texture features in NSCLC on non-contrast-enhanced CT and tumour metabolism and stage. Texture analysis warrants further investigation as a potential method for obtaining prognostic information for patients with NSCLC undergoing CT.

Computer-aided detection of metastatic brain tumors using automated three-dimensional template matching

PURPOSE

To demonstrate the efficacy of an automated three-dimensional (3D) template matching-based algorithm in detecting brain metastases on conventional MR scans and the potential of our algorithm to be developed into a computer-aided detection tool that will allow radiologists to maintain a high level of detection sensitivity while reducing image reading time.

MATERIALS AND METHODS

Spherical tumor appearance models were created to match the expected geometry of brain metastases while accounting for partial volume effects and offsets due to the cut of MRI sampling planes. A 3D normalized cross-correlation coefficient was calculated between the brain volume and spherical templates of varying radii using a fast frequency domain algorithm to identify likely positions of brain metastases.

RESULTS

Algorithm parameters were optimized on training datasets, and then data were collected on 22 patient datasets containing 79 total brain metastases producing a sensitivity of 89.9% with a false positive rate of 0.22 per image slice when restricted to the brain mass.

CONCLUSION

Study results demonstrate that the 3D template matching-based method can be an effective, fast, and accurate approach that could serve as a useful tool for assisting radiologists in providing earlier and more definitive diagnoses of metastases within the brain.

CAD (computed-aided detection) and CADx (computer aided diagnosis) systems in identifying and characterising lung nodules on chest CT: overview of research, developments and new prospects

Computer-aided detection (CAD) systems allow the automatic identification of lung nodules on chest computed tomography (CT), providing a second opinion to the radiologist's judgement and a volumetric evaluation of lesions - a very important aspect in oncological patients. The natural evolution of these systems has led to the introduction of computer-aided diagnosis (CADx) systems, which are able not only to identify nodules but also to characterise them by determining a likelihood of malignancy or benignity. The aim of this article is to describe the main technical principles of CAD and CADx systems, their applicability and influence in clinical practice and new prospects for their future development.

Artificial neural networks for prediction of response to chemoradiation in HT29 xenografts

PURPOSE

To evaluate the feasibility of using neural networks for predicting treatment response by using longitudinal measurements of apparent diffusion coefficient (ADC) obtained from diffusion-weighted magnetic resonance imaging (DWMRI).

METHODS AND MATERIALS

Mice bearing HT29 xenografts were allocated to six treatment groups receiving different combinations of daily chemotherapy and/or radiation therapy for 2 weeks. T(2)-weighted and DWMR images were acquired before treatment, twice during fractionated chemoradiation (at days 4 and 11), and four times after treatment ended (at days 18, 25, 32, and 46). A tumor doubling growth delay (T(delay)) value was found for individual xenografts. ADC values and treatment groups (1-6) were used as input to a back propagation neural network (BPNN) to predict T(delay).

RESULTS

When treatment group and ADC values from days 0, 4, 11, 18, 25, 32, and 46 were used as inputs to the BPNN, a strong correlation between measured and predicted T(delay) values was found (R = 0.731, p < 0.01). When ADC values from days 0, 4, and 11, and the treatment group were used as inputs, the correlation between predicted and measured T(delay) was 0.693 (p < 0.01).

CONCLUSIONS

BPNN was successfully used to predict T(delay) from tumor ADC values obtained from HT29 xenografts undergoing fractionated chemoradiation therapy.

Computer-aided diagnosis of lung cancer and pulmonary embolism in computed tomography-a review

Computer-aided detection (CADe) and computer-aided diagnosis (CADx) have been important areas of research in the last two decades. Significant progress has been made in the area of breast cancer detection, and CAD techniques are being developed in many other areas. Recent advances in multidetector row computed tomography have made it an increasingly common modality for imaging of lung diseases. A thoracic examination using thin-section computed tomography contains hundreds of images. Detection of lung cancer and pulmonary embolism on computed tomographic (CT) examinations are demanding tasks for radiologists because they have to search for abnormalities in a large number of images, and the lesions can be subtle. If successfully developed, CAD can be a useful second opinion to radiologists in thoracic CT interpretation. In this review, we summarize the studies that have been reported in these areas, discuss some challenges in the development of CAD, and identify areas that deserve particular attention in future research.

Validation of two models to estimate the probability of malignancy in patients with solitary pulmonary nodules

BACKGROUND

Effective strategies for managing patients with solitary pulmonary nodules (SPN) depend critically on the pre-test probability of malignancy.

OBJECTIVE

To validate two previously developed models that estimate the probability that an indeterminate SPN is malignant, based on clinical characteristics and radiographic findings.

METHODS

Data on age, smoking and cancer history, nodule size, location and spiculation were collected retrospectively from the medical records of 151 veterans (145 men, 6 women; age range 39-87 years) with an SPN measuring 7-30 mm (inclusive) and a final diagnosis established by histopathology or 2-year follow-up. Each patient's final diagnosis was compared with the probability of malignancy predicted by two models: one developed by investigators at the Mayo Clinic and the other developed from patients enrolled in a VA Cooperative Study. The accuracy of each model was assessed by calculating areas under the receiver operating characteristic (ROC) curve and the models were calibrated by comparing predicted and observed rates of malignancy.

RESULTS

The area under the ROC curve for the Mayo Clinic model (0.80; 95% CI 0.72 to 0.88) was higher than that of the VA model (0.73; 95% CI 0.64 to 0.82), but this difference was not statistically significant (Delta = 0.07; 95% CI -0.03 to 0.16). Calibration curves showed that the probability of malignancy was underestimated by the Mayo Clinic model and overestimated by the VA model.

CONCLUSIONS

Two existing prediction models are sufficiently accurate to guide decisions about the selection and interpretation of subsequent diagnostic tests in patients with SPNs, although clinicians should also consider the prevalence of malignancy in their practice setting when choosing a model.

Evaluation of patients with pulmonary nodules: when is it lung cancer?: ACCP evidence-based clinical practice guidelines (2nd edition)

BACKGROUND

Pulmonary nodules are spherical radiographic opacities that measure up to 30 mm in diameter. Nodules are extremely common in clinical practice and challenging to manage, especially small, "subcentimeter" nodules. Identification of malignant nodules is important because they represent a potentially curable form of lung cancer.

METHODS

We developed evidence-based clinical practice guidelines based on a systematic literature review and discussion with a large, multidisciplinary group of clinical experts and other stakeholders.

RESULTS

We generated a list of 29 recommendations for managing the solitary pulmonary nodule (SPN) that measures at least 8 to 10 mm in diameter; small, subcentimeter nodules that measure < 8 mm to 10 mm in diameter; and multiple nodules when they are detected incidentally during evaluation of the SPN. Recommendations stress the value of risk factor assessment, the utility of imaging tests (especially old films), the need to weigh the risks and benefits of various management strategies (biopsy, surgery, and observation with serial imaging tests), and the importance of eliciting patient preferences.

CONCLUSION

Patients with pulmonary nodules should be evaluated by estimation of the probability of malignancy, performance of imaging tests to characterize the lesion(s) better, evaluation of the risks associated with various management alternatives, and elicitation of patient preferences for treatment.

Lung metastases detection in CT images using 3D template matching

The aim of this study is to demonstrate a novel, fully automatic computer detection method applicable to metastatic tumors to the lung with a diameter of 4-20 mm in high-risk patients using typical computed tomography (CT) scans of the chest. Three-dimensional (3D) spherical tumor appearance models (templates) of various sizes were created to match representative CT imaging parameters and to incorporate partial volume effects. Taking into account the variability in the location of CT sampling planes cut through the spherical models, three offsetting template models were created for each appearance model size. Lung volumes were automatically extracted from computed tomography images and the correlation coefficients between the subregions around each voxel in the lung volume and the set of appearance models were calculated using a fast frequency domain algorithm. To determine optimal parameters for the templates, simulated tumors of varying sizes and eccentricities were generated and superposed onto a representative human chest image dataset. The method was applied to real image sets from 12 patients with known metastatic disease to the lung. A total of 752 slices and 47 identifiable tumors were studied. Spherical templates of three sizes (6, 8, and 10 mm in diameter) were used on the patient image sets; all 47 true tumors were detected with the inclusion of only 21 false positives. This study demonstrates that an automatic and straightforward 3D template-matching method, without any complex training or postprocessing, can be used to detect small lung metastases quickly and reliably in the clinical setting.

Precision of computer-aided volumetry of artificial small solid pulmonary nodules inex vivoporcine lungs

The purpose of this study was to investigate the precision of CT-based volumetric measurements of artificial small pulmonary nodules under ex vivo conditions. We implanted 322 artificial nodules in 23 inflated ex vivo porcine lungs in a dedicated chest phantom. The lungs were examined with a multislice spiral CT (20 mAs, collimation 16x0.75 mm, 1 mm slice thickness, 0.7 mm increment). A commercial volumetry software package (LungCARE VA70C-W; Siemens, Erlangen, Germany) was used for volume analysis in a semi-automatic and a manual corrected mode. After imaging, the lungs were dissected to harvest the nodules for gold standard determination. The volumes of 202 solitary, solid and well-defined lesions without contact with the pleura, greater bronchi or vessels were compared with the results of volumetry. A mean nodule diameter of 8.3 mm (+/-2.1 mm) was achieved. The mean relative deviation from the true lesion volume was -9.2% (+/-10.6%) for semi-automatic and -0.3% (+/-6.5%) for manual corrected volumetry. The subgroup of lesions from 5 mm to <10 mm in diameter showed a mean relative deviation of -8.7% (+/-10.9%) for semi-automatic volumetry and -0.3% (+/-6.9%) for manually corrected volumetry. We conclude that the presented software allowed for precise volumetry of artificial nodules in ex vivo lung tissue. This result is comparable to the findings of previous in vitro studies.

ACR Appropriateness Criteria® on Solitary Pulmonary Nodule

The solitary pulmonary nodule is traditionally defined as a relatively spherical opacity 3 cm or less in diameter surrounded by lung parenchyma. The choice of imaging test to evaluate solitary nodules is extensive. However, only 2 findings are considered to be sufficient to preclude further evaluation: calcification in a benign pattern and stability in size for more than 2 years.

Computer analysis of computed tomography scans of the lung: a survey

Current computed tomography (CT) technology allows for near isotropic, submillimeter resolution acquisition of the complete chest in a single breath hold. These thin-slice chest scans have become indispensable in thoracic radiology, but have also substantially increased the data load for radiologists. Automating the analysis of such data is, therefore, a necessity and this has created a rapidly developing research area in medical imaging. This paper presents a review of the literature on computer analysis of the lungs in CT scans and addresses segmentation of various pulmonary structures, registration of chest scans, and applications aimed at detection, classification and quantification of chest abnormalities. In addition, research trends and challenges are identified and directions for future research are discussed.

Reproducibility of computer-aided volumetry of artificial small pulmonary nodules in ex vivo porcine lungs

OBJECTIVE

The main purpose of this study was to investigate the reproducibility of computed tomography (CT)-based volumetric measurements of small pulmonary nodules.

METHODS

We implanted 70 artificial pulmonary nodules in 5 ex vivo porcine lungs in a dedicated chest phantom. The lungs were scanned 5 times consecutively with multislice-CT (collimation 16 x 0.75 mm, slice thickness 1 mm, reconstruction increment 0.7 mm). A commercial software package was used for lesion volumetry. The authors differentiated between intrascan reproducibility, interscan reproducibility, and results from semiautomatic and postprocessed volumetry.

RESULTS

Analysis of intrascan reproducibility revealed a mean variation coefficient of 6.2% for semiautomatic volumetry and of 0.7% for human adapted volumetry. For interscan reproducibility a mean variation coefficient of 9.2% and for human adapted volumetry a mean of 3.7% was detected.

CONCLUSION

The presented volumetry software showed a high reproducibility that can be expected to detect nodule growth with a high degree of certainty.

Computer aided characterization of the solitary pulmonary nodule using volumetric and contrast enhancement features

RATIONALE AND OBJECTIVES

To investigate the utility of a computer-aided diagnosis (CAD) in the task of differentiating malignant nodules from benign nodules based on quantitative features extracted from volumetric thin section CT image data acquired before and after the injection of contrast media.

MATERIALS AND METHODS

35 volumetric CT datasets of solitary pulmonary nodules (SPN) with proven diagnoses (19 malignant/16 benign) were contoured by a thoracic radiologist. All studies had at least a baseline series obtained without contrast media and at least one series following an intravenous contrast injection at 45, 90, 180, and 360 seconds. Two separate contours were created for each nodule: one including only the solid portion and another including the ground-glass component, if any, of the nodule. For each contour 31 features were calculated that measured the attenuation, shape, and enhancement of the nodule due to the injection of contrast. These features were input into a feature selection step and three different classifiers to determine if the diagnosis could be predicted from the resulting feature vector. In addition, observer input was introduced to two of the classifiers as an a priori probability of malignancy and the resulting performance was compared. Training and testing was conducted in a resubstitution and leave-one-out fashion and performance was evaluated using ROC analysis.

RESULTS

In a leave-one-out testing methodology, the classifiers achieved areas under the ROC curves AZ that ranged from 0.69 to 0.92. A classifier based on logistic regression performed the best with an AZ of 0.92 while a classifier based on quadratic discriminant analysis performed the poorest (AZ, 0.69). The AZ increased when using a priori observer input in most cases reaching a maximum of 0.95.

CONCLUSION

Based on this initial work with a limited number of nodules in our dataset, it appears that CAD using volumetric and contrast-enhanced data has the potential to assist radiologists in the task of differentiating solitary pulmonary nodules and in the management of these patients. Further studies with an increased number of patients are required to validate these results.

Lung Nodule Detection and Characterization With Multi-Slice CT

The influence of MSCT on nodule detection and characterization will be discussed. The objective is to improve understanding of the clinical issues involved in nodule detection, characterization, and management in light of technological advances. Topics to be covered are noninvasive characterization techniques, such as morphologic and density inspection on CT, nodule enhancement techniques, CT-PET, temporal nodule size assessment, and computer aided diagnosis for both detection and characterization.

Computer-aided diagnosis of the solitary pulmonary nodule

RATIONALE AND OBJECTIVES

We sought to investigate the utility of a computer-aided diagnosis in the task of differentiating malignant nodules from benign nodules based on single thin-section computed tomography image data.

MATERIALS AND METHODS

Eighty-one thin-section computed tomography data sets of solitary pulmonary nodules with proven diagnoses (48 malignant and 33 benign) were contoured manually on a single representative slice by a thoracic radiologist (>10 years of experience). Two separate contours were created for each nodule, one including only the solid portion of the nodule and one including any ground-glass components. For each contour 75 features were calculated that measured the attenuation, shape, and texture of the nodule. These features were than input into a feature selection step and four different classifiers to determine if the diagnosis could be predicted from the feature vector. Training and testing was conducted in a resubstitution and leave-one-out fashion and performance was evaluated using ROC techniques.

RESULTS

In a leave-one-out testing methodology the classifiers resulted with areas under the ROC curve (A(Z)) that ranged from 0.68 to 0.92. When evaluating with resubstitution the A(Z) ranged from 0.93 to 1.00.

CONCLUSION

Computer-aided diagnosis has the potential to assist radiologists in the task of differentiating solitary pulmonary nodules and in the management of these patients.

Pulmonary nodules 10 mm or less in diameter with ground-glass opacity component detected by high-resolution computed tomography have a high possibility of malignancy

OBJECTIVES

For the histological diagnosis of small lung cancers of 10 mm or less in diameter (< or =10), resection by video-assisted thoracic surgery (VATS) with computed tomography (CT)-guided marking is feasible. One problem is that a small number of these pulmonary nodules are malignant. We retrospectively analyzed CT images of pulmonary nodules to find better criteria to select candidates for resection among patients with small pulmonary nodules.

METHODS

Ninety-four patients with indeterminate peripheral pulmonary nodules underwent wedge resection by VATS. High-resolution CT using a 1.25 mm slice included the area of lesions. Nodules were classified by size (< or =10, 11 to 20, >20 mm) and whether they had a ground-glass opacity (GGO) component.

RESULTS

The histology of all 94 nodules showed 52 primary lung cancers, 6 metastatic tumors, 5 benign tumors, 8 intrapulmonary lymph nodes, and 23 inflammatory nodules. Ninety-three percent of nodules larger than 20 mm, 75% of nodules 10 to 20 mm, and 43% of nodules < or =10 mm were malignant. Introducing a classification according to GGO component to nodules, malignancy was detected in 88% of nodules with a GGO component and in 30% of nodules without a GGO component among nodules < or =10 mm. Nodules < or =10 mm with a GGO component showed a statistically significant (p < 0.01) correlation with malignancy.

CONCLUSIONS

Pulmonary nodules < or =10 mm with GGO should be considered to have a high possibility of malignancy and to be candidates for resection by VATS.

Radiologists' performance for differentiating benign from malignant lung nodules on high-resolution CT using computer-estimated likelihood of malignancy

OBJECTIVE

The purpose of our study was to evaluate whether a computer-aided diagnosis (CAD) scheme can assist radiologists in distinguishing small benign from malignant lung nodules on high-resolution CT (HRCT).

MATERIALS AND METHODS

We developed an automated computerized scheme for determining the likelihood of malignancy of lung nodules on multiple HRCT slices; the likelihood estimate was obtained from various objective features of the nodules using linear discriminant analysis. The data set used in this observer study consisted of 28 primary lung cancers (6-20 mm) and 28 benign nodules. Cancer cases included nodules with pure ground-glass opacity, mixed ground-glass opacity, and solid opacity. Benign nodules were selected by matching their size and pattern to the malignant nodules. Consecutive region-of-interest images for each nodule on HRCT were displayed for interpretation in stacked mode on a cathode ray tube monitor. The images were presented to 16 radiologists-first without and then with the computer output-who were asked to indicate their confidence level regarding the malignancy of a nodule. Performance was evaluated by receiver operating characteristic (ROC) analysis.

RESULTS

The area under the ROC curve (Az value) of the CAD scheme alone was 0.831 for distinguishing benign from malignant nodules. The average Az value for radiologists was improved with the aid of the CAD scheme from 0.785 to 0.853 by a statistically significant level (p = 0.016). The radiologists' diagnostic performance with the CAD scheme was more accurate than that of the CAD scheme alone (p < 0.05) and also that of radiologists alone.

CONCLUSION

CAD has the potential to improve radiologists' diagnostic accuracy in distinguishing small benign nodules from malignant ones on HRCT.

Lung image database consortium: developing a resource for the medical imaging research community

To stimulate the advancement of computer-aided diagnostic (CAD) research for lung nodules in thoracic computed tomography (CT), the National Cancer Institute launched a cooperative effort known as the Lung Image Database Consortium (LIDC). The LIDC is composed of five academic institutions from across the United States that are working together to develop an image database that will serve as an international research resource for the development, training, and evaluation of CAD methods in the detection of lung nodules on CT scans. Prior to the collection of CT images and associated patient data, the LIDC has been engaged in a consensus process to identify, address, and resolve a host of challenging technical and clinical issues to provide a solid foundation for a scientifically robust database. These issues include the establishment of (a) a governing mission statement, (b) criteria to determine whether a CT scan is eligible for inclusion in the database, (c) an appropriate definition of the term qualifying nodule, (d) an appropriate definition of "truth" requirements, (e) a process model through which the database will be populated, and (f) a statistical framework to guide the application of assessment methods by users of the database. Through a consensus process in which careful planning and proper consideration of fundamental issues have been emphasized, the LIDC database is expected to provide a powerful resource for the medical imaging research community. This article is intended to share with the community the breadth and depth of these key issues.

Automated lung nodule classification following automated nodule detection on CT: a serial approach

We have evaluated the performance of an automated classifier applied to the task of differentiating malignant and benign lung nodules in low-dose helical computed tomography (CT) scans acquired as part of a lung cancer screening program. The nodules classified in this manner were initially identified by our automated lung nodule detection method, so that the output of automated lung nodule detection was used as input to automated lung nodule classification. This study begins to narrow the distinction between the "detection task" and the "classification task." Automated lung nodule detection is based on two- and three-dimensional analyses of the CT image data. Gray-level-thresholding techniques are used to identify initial lung nodule candidates, for which morphological and gray-level features are computed. A rule-based approach is applied to reduce the number of nodule candidates that correspond to non-nodules, and the features of remaining candidates are merged through linear discriminant analysis to obtain final detection results. Automated lung nodule classification merges the features of the lung nodule candidates identified by the detection algorithm that correspond to actual nodules through another linear discriminant classifier to distinguish between malignant and benign nodules. The automated classification method was applied to the computerized detection results obtained from a database of 393 low-dose thoracic CT scans containing 470 confirmed lung nodules (69 malignant and 401 benign nodules). Receiver operating characteristic (ROC) analysis was used to evaluate the ability of the classifier to differentiate between nodule candidates that correspond to malignant nodules and nodule candidates that correspond to benign lesions. The area under the ROC curve for this classification task attained a value of 0.79 during a leave-one-out evaluation.

The challenge of the solitary pulmonary nodule: diagnostic assessment with multislice spiral CT

The advent of fast multiscale computed tomography (MSCT) technology has sparked new interest in the noninvasive assessment of the solitary pulmonary nodule (SPN). Fast scanning within a single breath-hold period, simultaneous acquisition of multiple thin slices with subsequent morphologic characterization of the nodule, determination of perfusion patterns as well as growth rates has led to unprecedented improvements in this emerging field. This article reviews the capabilities of MSCT in the diagnostic assessment of the SPN.

The solitary pulmonary nodule

More than 150,00 patients a year present to their physicians with the diagnostic dilemma of a solitary pulmonary nodule (SPN) found either on chest radiography or chest CT. A thoughtful and timely workup of this finding is essential if lung cancer is to be recognized early and the chance for cure optimized. Based on the literature to date, recommendations are made for appropriate imaging modalities and diagnostic testing, as well as indications for obtaining preoperative tissue diagnosis for the patient with an SPN.

Usefulness of an artificial neural network for differentiating benign from malignant pulmonary nodules on high-resolution CT: evaluation with receiver operating characteristic analysis

OBJECTIVE

The purpose of our study was to use an artificial neural network to differentiate benign from malignant pulmonary nodules on high-resolution CT findings and to evaluate the effect of artificial neural network output on the performance of radiologists using receiver operating characteristic analysis.

MATERIALS AND METHODS

We selected 155 cases with pulmonary nodules less than 3 cm (99 malignant nodules and 56 benign nodules). An artificial neural network was used to distinguish benign from malignant nodules on the basis of seven clinical parameters and 16 radiologic findings that were extracted by attending radiologists using subjective rating scales. In the observer test, 12 radiologists (four attending radiologists, four radiology fellows, and four radiology residents) were presented with high-resolution CT images, first without and then with the artificial neural network output. Observer performance was evaluated by means of receiver operating characteristic analysis using a continuous rating scale.

RESULTS

The artificial neural network showed a high performance in differentiating benign from malignant pulmonary nodules (A(z) = 0.951). The average A(z) value for all radiologists increased by a statistically significant level, from 0.831 to 0.959, with the use of the artificial neural network output.

CONCLUSION

Our computerized scheme using the artificial neural network can improve the diagnostic accuracy of radiologists who are differentiating benign from malignant pulmonary nodules on high-resolution CT.

Chest CT: automated nodule detection and assessment of change over time--preliminary experience

The authors developed a computer system that automatically identifies nodules at chest computed tomography, quantifies their diameter, and assesses for change in size at follow-up. The automated nodule detection system identified 318 (86%) of 370 nodules in 16 studies (eight initial and eight follow-up studies) obtained in eight oncology patients with known nodules. Assessment of change in nodule size by the computer matched that by the thoracic radiologist (Spearman rank correlation coefficient, 0.932).

Automatic detection and quantification of ground-glass opacities on high-resolution CT using multiple neural networks: comparison with a density mask

OBJECTIVE

We compared multiple neural networks with a density mask for the automatic detection and quantification of ground-glass opacities on high-resolution CT under clinical conditions.

SUBJECTS AND METHODS

Eighty-four patients (54 men and 30 women; age range, 18-82 years; mean age, 49 years) with a total of 99 consecutive high-resolution CT scans were enrolled in the study. The neural network was designed to detect ground-glass opacities with high sensitivity and to omit air-tissue interfaces to increase specificity. The results of the neural network were compared with those of a density mask (thresholds, -750/-300 H), with a radiologist serving as the gold standard.

RESULTS

The neural network classified 6% of the total lung area as ground-glass opacities. The density mask failed to detect 1.3%, and this percentage represented the increase in sensitivity that was achieved by the neural network. The density mask identified another 17.3% of the total lung area to be ground-glass opacities that were not detected by the neural network. This area represented the increase in specificity achieved by the neural network. Related to the extent of the ground-glass opacities as classified by the radiologist, the neural network (density mask) reached a sensitivity of 99% (89%), specificity of 83% (55%), positive predictive value of 78% (18%), negative predictive value of 99% (98%), and accuracy of 89% (58%).

CONCLUSION

Automatic segmentation and quantification of ground-glass opacities on high-resolution CT by a neural network are sufficiently accurate to be implemented for the preinterpretation of images in a clinical environment; it is superior to a double-threshold density mask.

Comparison of biopsy techniques in assessment of solitary pulmonary nodules

A wide variety of diagnostic tests are available to evaluate solitary pulmonary nodules, ranging from noninvasive to invasive. Given the virulence of lung cancer, those techniques that can provide cytological and pathological information are often chosen. However, the choice of which procedure to perform is complicated by numerous factors, including the sensitivity and specificity of the test, as well as the prevalence of disease. Additional considerations also include complications, availability and expertise in performing procedures, and overall cost of the diagnostic algorithm. Rather than make specific recommendations for diagnostic workup, it is more appropriate to consider that this will vary from institution to institution based on the above factors.

Artificial neural networks in chest radiography: application to the differential diagnosis of interstitial lung disease

RATIONALE AND OBJECTIVES

The authors evaluated the usefulness of artificial neural networks (ANNs) in the differential diagnosis of interstitial lung disease.

MATERIALS AND METHODS

The authors used three-layer, feed-forward ANNs with a back-propagation algorithm. The ANNs were designed to distinguish between 11 interstitial lung diseases on the basis of 10 clinical parameters and 16 radiologic findings extracted by chest radiologists. Thus, the ANNs consisted of 26 input units and 11 output units. One hundred fifty actual clinical cases, 110 cases from previously published articles, and 110 hypothetical cases were used for training and testing the ANNs by using a round-robin (or leave-one-out) technique. ANN performance was evaluated with receiver operating characteristic (ROC) analysis.

RESULTS

The Az (area under the ROC curve) obtained with actual clinical cases was 0.947, and both the sensitivity and specificity of the ANNs were approximately 90% in terms of indicating the correct diagnosis with the two largest output values among the 11 diseases.

CONCLUSION

ANNs using clinical parameters and radiologic findings may be useful for making the differential diagnosis of interstitial lung disease on chest radiographs.