Development of an automatic classification system for differentiation of obstructive lung disease using HRCT

Clustering analysis of HRCT parameters measured using a texture-based automated system: relationship with clinical outcomes of IPF

PURPOSE

The extent of honeycombing and reticulation predict the clinical prognosis of IPF. Emphysema, consolidation, and ground glass opacity are visible in HRCT scans. To date, there have been few comprehensive studies that have used these parameters. We conducted automated quantitative analysis to identify predictive parameters for clinical outcomes and then grouped the subjects accordingly.

METHODS

CT images were obtained while patients held their breath at full inspiration. Parameters were analyzed using an automated lung texture quantification system. Cluster analysis was conducted on 159 IPF patients and clinical profiles were compared between clusters in terms of survival.

RESULTS

Kaplan-Meier analysis revealed that survival rates declined as fibrosis, reticulation, honeycombing, consolidation, and emphysema scores increased. Cox regression analysis revealed that reticulation had the most significant impact on survival rate, followed by honeycombing, consolidation, and emphysema scores. Hierarchical and K-means cluster analyses revealed 3 clusters. Cluster 1 (n = 126) with the lowest values for all parameters had the longest survival duration, and relatively-well preserved FVC and DLCO. Cluster 2 (n = 15) with high reticulation and consolidation scores had the lowest FVC and DLCO values with a predominance of female, while cluster 3 (n = 18) with high honeycombing and emphysema scores predominantly consisted of male smokers. Kaplan-Meier analysis revealed that cluster 2 had the lowest survival rate, followed by cluster 3 and cluster 1.

CONCLUSION

Automated quantitative CT analysis provides valuable information for predicting clinical outcomes, and clustering based on these parameters may help identify the high-risk group for management.

CT Quantification of Interstitial Lung Abnormality and Interstitial Lung Disease: From Technical Challenges to Future Directions

ABSTRACT

Interstitial lung disease (ILD) encompasses a variety of lung disorders with varying degrees of inflammation or fibrosis, requiring a combination of clinical, imaging, and pathologic data for evaluation. Imaging is essential for the noninvasive diagnosis of the disease, as well as for assessing disease severity, monitoring its progression, and evaluating treatment response. However, traditional visual assessments of ILD with computed tomography (CT) suffer from reader variability. Automated quantitative CT offers a more objective approach by using computer-based analysis to consistently evaluate and measure ILD. Advancements in technology have significantly improved the accuracy and reliability of these measurements. Recently, interstitial lung abnormalities (ILAs), which represent potential preclinical ILD incidentally found on CT scans and are characterized by abnormalities in over 5% of any lung zone, have gained attention and clinical importance. The challenge lies in the accurate and consistent identification of ILA, given that its definition relies on a subjective threshold, making quantitative tools crucial for precise ILA evaluation. This review highlights the state of CT quantification of ILD and ILA, addressing clinical and research disparities while emphasizing how machine learning or deep learning in quantitative imaging can improve diagnosis and management by providing more accurate assessments, and finally, suggests the future directions of quantitative CT in this area.

Unraveling the interplay of image formation, data representation and learning in CT-based COPD phenotyping automation: The need for a meta-strategy

PURPOSE

In the literature on automated phenotyping of chronic obstructive pulmonary disease (COPD), there is a multitude of isolated classical machine learning and deep learning techniques, mostly investigating individual phenotypes, with small study cohorts and heterogeneous meta-parameters, e.g., different scan protocols or segmented regions. The objective is to compare the impact of different experimental setups, i.e., varying meta-parameters related to image formation and data representation, with the impact of the learning technique for subtyping automation for a variety of phenotypes. The identified associations of these parameters with automation performance and their interactions might be a first step towards a determination of optimal meta-parameters, i.e., a meta-strategy.

METHODS

A clinical cohort of 981 patients (53.8 ± 15.1 years, 554 male) was examined. The inspiratory CT images were analyzed to automate the diagnosis of 13 COPD phenotypes given by two radiologists. A benchmark feature set that integrates many quantitative criteria was extracted from the lung and trained a variety of learning algorithms on the first 654 patients (two thirds) and the respective algorithm retrospectively assessed the remaining 327 patients (one third). The automation performance was evaluated by the area under the receiver operating characteristic curve (AUC). 1717 experiments were conducted with varying meta-parameters such as reconstruction kernel, segmented regions and input dimensionality, i.e., number of extracted features. The association of the meta-parameters with the automation performance was analyzed by multivariable general linear model decomposition of the automation performance in the contributions of meta-parameters and the learning technique.

RESULTS

The automation performance varied strongly for varying meta-parameters. For emphysema-predominant phenotypes, an AUC of 93%-95% could be achieved for the best meta-configuration. The airways-predominant phenotypes led to a lower performance of 65%-85%, while smooth kernel configurations on average were unexpectedly superior to those with sharp kernels. The performance impact of meta-parameters, even that of often neglected ones like the missing-data imputation, was in general larger than that of the learning technique. Advanced learning techniques like 3D deep learning or automated machine learning yielded inferior automation performance for non-optimal meta-configurations in comparison to simple techniques with suitable meta-configurations. The best automation performance was achieved by a combination of modern learning techniques and a suitable meta-configuration.

CONCLUSIONS

Our results indicate that for COPD phenotype automation, study design parameters such as reconstruction kernel and the model input dimensionality should be adapted to the learning technique and may be more important than the technique itself. To achieve optimal automation and prediction results, the interaction between input those meta-parameters and the learning technique should be considered. This might be particularly relevant for the development of specific scan protocols for novel learning algorithms, and towards an understanding of good study design for automated phenotyping.

Hybrid Airway Segmentation Using Multi-Scale Tubular Structure Filters and Texture Analysis on 3D Chest CT Scans

Airway diseases are frequently related to morphological changes that may influence lung physiology. Accurate airway region segmentation may be useful for quantitative evaluation of disease prognosis and therapy efficacy. The information can also be applied to understand the fundamental mechanisms of various lung diseases. We present a hybrid method to automatically segment the airway regions on 3D volume chest computed tomography (CT) scans. This method uses multi-scale filtering and support vector machine (SVM) classification. The proposed scheme is comprised of two hybrid steps. First, a tubular structure-based multi-scale filter is applied to find the initial candidate airway regions. Second, for identifying candidate airway regions using the fuzzy connectedness technique, the small and disconnected branches of airway regions are detected using SVM classification trained to differentiate between airway and non-airway regions through texture analysis of user-defined landmark points. For development and evaluation of the method, two datasets were incorporated: (1) 55 lung-CT volumes from the Korean Obstructive Lung Disease (KOLD) Cohort Study and (2) 20 cases from the publicly open database (EXACT'09). The average tree-length detection rates of EXACT'09 and KOLD were 56.9 ± 11.0 and 70.5 ± 8.98, respectively. Comparison of the results for the EXACT'09 data set between the presented method and other methods revealed that our approach was a high performer. The method limitations were higher false-positive rates than those of the other methods and risk of leakage. In future studies, application of a convolutional neural network will help overcome these shortcomings.

Development of a Computer-Aided Differential Diagnosis System to Distinguish Between Usual Interstitial Pneumonia and Non-specific Interstitial Pneumonia Using Texture- and Shape-Based Hierarchical Classifiers on HRCT Images

A computer-aided differential diagnosis (CADD) system that distinguishes between usual interstitial pneumonia (UIP) and non-specific interstitial pneumonia (NSIP) using high-resolution computed tomography (HRCT) images was developed, and its results compared against the decision of a radiologist. Six local interstitial lung disease patterns in the images were determined, and 900 typical regions of interest were marked by an experienced radiologist. A support vector machine classifier was used to train and label the regions of interest of the lung parenchyma based on the texture and shape characteristics. Based on the regional classifications of the entire lung using HRCT, the distributions and extents of the six regional patterns were characterized through their CADD features. The disease division index of every area fraction combination and the asymmetric index between the left and right lungs were also evaluated. A second SVM classifier was employed to classify the UIP and NSIP, and features were selected through sequential-forward floating feature selection. For the evaluation, 54 HRCT images of UIP (n = 26) and NSIP (n = 28) patients clinically diagnosed by a pulmonologist were included and evaluated. The classification accuracy was measured based on a fivefold cross-validation with 20 repetitions using random shuffling. For comparison, thoracic radiologists assessed each case using HRCT images without clinical information or diagnosis. The accuracies of the radiologists' decisions were 75 and 87%. The accuracies of the CADD system using different features ranged from 70 to 81%. Finally, the accuracy of the proposed CADD system after sequential-forward feature selection was 91%.

Deep Learning in Radiology

As radiology is inherently a data-driven specialty, it is especially conducive to utilizing data processing techniques. One such technique, deep learning (DL), has become a remarkably powerful tool for image processing in recent years. In this work, the Association of University Radiologists Radiology Research Alliance Task Force on Deep Learning provides an overview of DL for the radiologist. This article aims to present an overview of DL in a manner that is understandable to radiologists; to examine past, present, and future applications; as well as to evaluate how radiologists may benefit from this remarkable new tool. We describe several areas within radiology in which DL techniques are having the most significant impact: lesion or disease detection, classification, quantification, and segmentation. The legal and ethical hurdles to implementation are also discussed. By taking advantage of this powerful tool, radiologists can become increasingly more accurate in their interpretations with fewer errors and spend more time to focus on patient care.

Automatic emphysema detection using weakly labeled HRCT lung images

PURPOSE

A method for automatically quantifying emphysema regions using High-Resolution Computed Tomography (HRCT) scans of patients with chronic obstructive pulmonary disease (COPD) that does not require manually annotated scans for training is presented.

METHODS

HRCT scans of controls and of COPD patients with diverse disease severity are acquired at two different centers. Textural features from co-occurrence matrices and Gaussian filter banks are used to characterize the lung parenchyma in the scans. Two robust versions of multiple instance learning (MIL) classifiers that can handle weakly labeled data, miSVM and MILES, are investigated. Weak labels give information relative to the emphysema without indicating the location of the lesions. The classifiers are trained with the weak labels extracted from the forced expiratory volume in one minute (FEV1) and diffusing capacity of the lungs for carbon monoxide (DLCO). At test time, the classifiers output a patient label indicating overall COPD diagnosis and local labels indicating the presence of emphysema. The classifier performance is compared with manual annotations made by two radiologists, a classical density based method, and pulmonary function tests (PFTs).

RESULTS

The miSVM classifier performed better than MILES on both patient and emphysema classification. The classifier has a stronger correlation with PFT than the density based method, the percentage of emphysema in the intersection of annotations from both radiologists, and the percentage of emphysema annotated by one of the radiologists. The correlation between the classifier and the PFT is only outperformed by the second radiologist.

CONCLUSIONS

The presented method uses MIL classifiers to automatically identify emphysema regions in HRCT scans. Furthermore, this approach has been demonstrated to correlate better with DLCO than a classical density based method or a radiologist, which is known to be affected in emphysema. Therefore, it is relevant to facilitate assessment of emphysema and to reduce inter-observer variability.

An Ensemble Method for Classifying Regional Disease Patterns of Diffuse Interstitial Lung Disease Using HRCT Images from Different Vendors

We propose the use of ensemble classifiers to overcome inter-scanner variations in the differentiation of regional disease patterns in high-resolution computed tomography (HRCT) images of diffuse interstitial lung disease patients obtained from different scanners. A total of 600 rectangular 20 × 20-pixel regions of interest (ROIs) on HRCT images obtained from two different scanners (GE and Siemens) and the whole lung area of 92 HRCT images were classified as one of six regional pulmonary disease patterns by two expert radiologists. Textual and shape features were extracted from each ROI and the whole lung parenchyma. For automatic classification, individual and ensemble classifiers were trained and tested with the ROI dataset. We designed the following three experimental sets: an intra-scanner study in which the training and test sets were from the same scanner, an integrated scanner study in which the data from the two scanners were merged, and an inter-scanner study in which the training and test sets were acquired from different scanners. In the ROI-based classification, the ensemble classifiers showed better (p < 0.001) accuracy (89.73%, SD = 0.43) than the individual classifiers (88.38%, SD = 0.31) in the integrated scanner test. The ensemble classifiers also showed partial improvements in the intra- and inter-scanner tests. In the whole lung classification experiment, the quantification accuracies of the ensemble classifiers with integrated training (49.57%) were higher (p < 0.001) than the individual classifiers (48.19%). Furthermore, the ensemble classifiers also showed better performance in both the intra- and inter-scanner experiments. We concluded that the ensemble classifiers provide better performance when using integrated scanner images.

Lung disease stratification using amalgamation of Riesz and Gabor transforms in machine learning framework

Lung disease risk stratification is important for both diagnosis and treatment planning, particularly in biopsies and radiation therapy. Manual lung disease risk stratification is challenging because of: (a) large lung data sizes, (b) inter- and intra-observer variability of the lung delineation and (c) lack of feature amalgamation during machine learning paradigm. This paper presents a two stage CADx cascaded system consisting of: (a) semi-automated lung delineation subsystem (LDS) for lung region extraction in CT slices followed by (b) morphology-based lung tissue characterization, thereby addressing the above shortcomings. LDS primarily uses entropy-based region extraction while ML-based lung characterization is mainly based on an amalgamation of directional transforms such as Riesz and Gabor along with texture-based features comprising of 100 greyscale features using the K-fold cross-validation protocol (K = 2, 3, 5 and 10). The lung database consisted of 96 patients: 15 normal and 81 diseased. We use five high resolution Computed Tomography (HRCT) levels representing different anatomy landmarks where disease is commonly seen. We demonstrate the amalgamated ML stratification accuracy of 99.53%, an increase of 2% against the conventional non-amalgamation ML system that uses alone Riesz-based feature embedded with feature selection based on feature strength. The robustness of the system was determined based on the reliability and stability that showed a reliability index of 0.99 and the deviation in risk stratification accuracies less than 5%. Our CADx system shows 10% better performance when compared against the mean of five other prominent studies available in the current literature covering over one decade.

Predictors of schizophrenia spectrum disorders in early-onset first episodes of psychosis: a support vector machine model

Identifying early-onset schizophrenia spectrum disorders (SSD) at a very early stage remains challenging. To assess the diagnostic predictive value of multiple types of data at the emergence of early-onset first-episode psychosis (FEP), various support vector machine (SVM) classifiers were developed. The data were from a 2-year, prospective, longitudinal study of 81 patients (age 9-17 years) with early-onset FEP and a stable diagnosis during follow-up and 42 age- and sex-matched healthy controls (HC). The input was different combinations of baseline clinical, neuropsychological, magnetic resonance imaging brain volumetric and biochemical data, and the output was the diagnosis at follow-up (SSD vs. non-SSD, SSD vs. HC, and non-SSD vs. HC). Enhanced recursive feature elimination was performed for the SSD vs. non-SSD classifier to select and rank the input variables with the highest predictive value for a diagnostic outcome of SSD. After validation with a test set and considering all baseline variables together, the SSD vs. non-SSD, SSD vs. HC and non-SSD vs. HC classifiers achieved an accuracy of 0.81, 0.99 and 0.99, respectively. Regarding the SSD vs. non-SSD classifier, a combination of baseline clinical variables (severity of negative, disorganized symptoms and hallucinations or poor insight) and neuropsychological variables (impaired attention, motor coordination, and global cognition) showed the highest predictive value for a diagnostic outcome of SSD. Neuroimaging and biochemical variables at baseline did not add to the predictive value. Thus, comprehensive clinical/cognitive assessment remains the most reliable approach for differential diagnosis during early-onset FEP. SVMs may constitute promising multivariate tools in the search for predictors of diagnostic outcome in FEP.

Three-dimensional SVM with latent variable: application for detection of lung lesions in CT images

The study aims to improve the performance of current computer-aided schemes for the detection of lung lesions, especially the low-contrast in gray density or irregular in shape. The relative position between suspected lesion and whole lung is, for the first time, added as a latent feature to enrich current Three-dimensional (3D) features such as shape, texture. Subsequently, 3D matrix patterns-based Support Vector Machine (SVM) with the latent variable, referred to as L-SVM3Dmatrix, was constructed accordingly. A CT image database containing 750 abnormal cases with 1050 lesions was used to train and evaluate several similar computer-aided detection (CAD) schemes: traditional features-based SVM (SVMfeature), 3D matrix patterns-based SVM (SVM3Dmatrix) and L-SVM3Dmatrix. The classifier performances were evaluated by computing the area under the ROC curve (AUC), using a 5-fold cross-validation. The L-SVM3Dmatrix sensitivity was 93.0 with 1.23% percentage of False Positive (FP), the SVM3Dmatrix sensitivity was 88.4 with 1.49% percentage of FP, and the SVMfeature sensitivity was 87.2 with 1.78% percentage of FP. The L-SVM3Dmatrix outperformed other current lung CAD schemes, especially regarding the difficult lesions.

A support vector machine classifier reduces interscanner variation in the HRCT classification of regional disease pattern in diffuse lung disease: comparison to a Bayesian classifier

PURPOSE

To investigate the effect of using different computed tomography (CT) scanners on the accuracy of high-resolution CT (HRCT) images in classifying regional disease patterns in patients with diffuse lung disease, support vector machine (SVM) and Bayesian classifiers were applied to multicenter data.

METHODS

Two experienced radiologists marked sets of 600 rectangular 20 × 20 pixel regions of interest (ROIs) on HRCT images obtained from two scanners (GE and Siemens), including 100 ROIs for each of local patterns of lungs-normal lung and five of regional pulmonary disease patterns (ground-glass opacity, reticular opacity, honeycombing, emphysema, and consolidation). Each ROI was assessed using 22 quantitative features belonging to one of the following descriptors: histogram, gradient, run-length, gray level co-occurrence matrix, low-attenuation area cluster, and top-hat transform. For automatic classification, a Bayesian classifier and a SVM classifier were compared under three different conditions. First, classification accuracies were estimated using data from each scanner. Next, data from the GE and Siemens scanners were used for training and testing, respectively, and vice versa. Finally, all ROI data were integrated regardless of the scanner type and were then trained and tested together. All experiments were performed based on forward feature selection and fivefold cross-validation with 20 repetitions.

RESULTS

For each scanner, better classification accuracies were achieved with the SVM classifier than the Bayesian classifier (92% and 82%, respectively, for the GE scanner; and 92% and 86%, respectively, for the Siemens scanner). The classification accuracies were 82%/72% for training with GE data and testing with Siemens data, and 79%/72% for the reverse. The use of training and test data obtained from the HRCT images of different scanners lowered the classification accuracy compared to the use of HRCT images from the same scanner. For integrated ROI data obtained from both scanners, the classification accuracies with the SVM and Bayesian classifiers were 92% and 77%, respectively. The selected features resulting from the classification process differed by scanner, with more features included for the classification of the integrated HRCT data than for the classification of the HRCT data from each scanner. For the integrated data, consisting of HRCT images of both scanners, the classification accuracy based on the SVM was statistically similar to the accuracy of the data obtained from each scanner. However, the classification accuracy of the integrated data using the Bayesian classifier was significantly lower than the classification accuracy of the ROI data of each scanner.

CONCLUSIONS

The use of an integrated dataset along with a SVM classifier rather than a Bayesian classifier has benefits in terms of the classification accuracy of HRCT images acquired with more than one scanner. This finding is of relevance in studies involving large number of images, as is the case in a multicenter trial with different scanners.

Comparative evaluation of support vector machines for computer aided diagnosis of lung cancer in CT based on a multi-dimensional data set

Lung cancer is one of the most common forms of cancer resulting in over a million deaths per year worldwide. In this paper, the usage of support vector machine (SVM) classification for lung cancer is investigated, presenting a systematic quantitative evaluation against Boosting, Decision trees, k-nearest neighbor, LASSO regressions, neural networks and random forests. A large database of 5984 regions of interest (ROIs) and 488 input features (including textural features, patient characteristics, and morphological features) were used to train the classifiers and evaluate for their performance. The evaluation for classifiers' performance was based on a tenfold cross validation framework, receiver operating characteristic curve (ROC), and Matthews correlation coefficient. Area under curve (AUC) of SVM, Boosting, Decision trees, k-nearest neighbor, LASSO, neural networks, random forests were 0.94, 0.86, 0.73, 0.72, 0.91, 0.92, and 0.85, respectively. It was proved that SVM classification offered significantly increased classification performance compared to the reference methods. This scheme may be used as an auxiliary tool to differentiate between benign and malignant SPNs of CT images in future.

Fast and efficient lung disease classification using hierarchical one-against-all support vector machine and cost-sensitive feature selection

To improve time and accuracy in differentiating diffuse interstitial lung disease for computer-aided quantification, we introduce a hierarchical support vector machine which selects a class by training a binary classifier at each node in a hierarchy, thus allowing each classifier to use a class-specific quasi-optimal feature set. In addition, the computational cost-sensitive group-feature selection criterion combined with the sequential forward selection is applied in order to obtain a useful and computationally inexpensive quasi-optimal feature set for the purpose of accelerating the classification time. The classification time was reduced by up to 57% and the overall accuracy was significantly improved in comparison with the one-against-all and one-against-one support vector machine methods with sequential forward selection (paired t-test, p<0.001). The reduction of classification time as well as the improvement of overall accuracy demonstrates promise for the proposed classification method to be adopted in various real-time and on-line image-based clinical applications.

Automated texture-based quantification of centrilobular nodularity and centrilobular emphysema in chest CT images

RATIONALE AND OBJECTIVES

Characterization of smoking-related lung disease typically consists of visual assessment of chest computed tomographic (CT) images for the presence and extent of emphysema and centrilobular nodularity (CN). Quantitative analysis of emphysema and CN may improve the accuracy, reproducibility, and efficiency of chest CT scoring. The purpose of this study was to develop a fully automated texture-based system for the detection and quantification of centrilobular emphysema (CLE) and CN in chest CT images.

MATERIALS AND METHODS

A novel approach was used to prepare regions of interest (ROIs) within the lung parenchyma for representation by texture features associated with the gray-level run-length and gray-level gap-length methods. These texture features were used to train a multiple logistic regression classifier to discriminate between normal lung tissue, CN or "smoker's lung," and CLE. This classifier was trained and evaluated on 24 and 71 chest CT scans, respectively.

RESULTS

During training, the classifier correctly classified 89% of ROIs depicting normal lung tissue, 74% of ROIs depicting CN, and 95% of ROIs manifesting CLE. When the performance of the classifier in quantifying extent of CN and CLE was evaluated on 71 chest CT scans, 65% of ROIs in smokers without CLE were classified as CN, compared to 31% in nonsmokers (P < .001) and 28% in smokers with CLE (P < .001).

CONCLUSIONS

The texture-based framework described herein facilitates successful discrimination among normal lung tissue, CN, and CLE and can be used for the automated quantification of smoking-related lung disease.

Quantitative assessment of change in regional disease patterns on serial HRCT of fibrotic interstitial pneumonia with texture-based automated quantification system

OBJECTIVES

To evaluate the usefulness of a texture-based automated quantification system (AQS) for evaluating the extent and interval change of regional disease patterns on initial and follow-up high-resolution computed tomographies (HRCTs) of fibrotic interstitial pneumonia (FIP).

METHODS

Eighty-nine patients with clinically and/or biopsy confirmed usual interstitial pneumonia (UIP) (n = 71) and non-specific interstitial pneumonia (NSIP) (n = 18) were included. An AQS to quantify five disease patterns (ground-glass opacity [GGO], reticular opacity [RO], honeycombing [HC], emphysema [EMPH], consolidation [CONS]) and normal lung was developed. The extent and interval changes of each disease pattern, FS (fibrosis score), TA (total abnormal lung fraction) of entire lung on initial and 1-year follow-up HRCTs were quantified. The agreement between the results of AQS and two readers was assessed. Results of AQS were correlated with forced vital capacity (FVC) and carbon monoxide diffusing capacity (DLco).

RESULTS

The Intraclass correlation coefficient (ICC) study revealed acceptable agreement between visual assessment and AQS (r = 0.78, 0.66 for HC; 0.76, 0.61 for FS; 0.64, 0.68 for TA, initial and follow-up HRCTs, respectively). Linear regression analysis revealed the extent of HC, TA on initial CT, interval changes of FS contributed negatively to DLco, and interval changes of FS, TA contributed negatively to FVC.

CONCLUSIONS

Our AQS is comparable with visual assessment for evaluating the disease extent and the interval changes of FIP on HRCT.

Near-affine-invariant texture learning for lung tissue analysis using isotropic wavelet frames

We propose near-affine-invariant texture descriptors derived from isotropic wavelet frames for the characterization of lung tissue patterns in high-resolution computed tomography (HRCT) imaging. Affine invariance is desirable to enable learning of nondeterministic textures without a priori localizations, orientations, or sizes. When combined with complementary gray-level histograms, the proposed method allows a global classification accuracy of 76.9% with balanced precision among five classes of lung tissue using a leave-one-patient-out cross validation, in accordance with clinical practice.

Regional context-sensitive support vector machine classifier to improve automated identification of regional patterns of diffuse interstitial lung disease

We propose the use of a context-sensitive support vector machine (csSVM) to enhance the performance of a conventional support vector machine (SVM) for identifying diffuse interstitial lung disease (DILD) in high-resolution computerized tomography (HRCT) images. Nine hundred rectangular regions of interest (ROIs), each 20 × 20 pixels in size and consisting of 150 ROIs representing six regional disease patterns (normal, ground-glass opacity, reticular opacity, honeycombing, emphysema, and consolidation), were marked by two experienced radiologists using consensus HRCT images of various DILD. Twenty-one textual and shape features were evaluated to characterize the ROIs. The csSVM classified an ROI by simultaneously using the decision value of each class and information from the neighboring ROIs, such as neighboring region feature distances and class differences. Sequential forward-selection was used to select the relevant features. To validate our results, we used 900 ROIs with fivefold cross-validation and 84 whole lung images categorized by a radiologist. The accuracy of the proposed method for ROI and whole lung classification (89.88 ± 0.02%, and 60.30 ± 13.95%, respectively) was significantly higher than that provided by the conventional SVM classifier (87.39 ± 0.02%, and 57.69 ± 13.31%, respectively; paired t test, p < 0.01, and p < 0.01, respectively). We conclude that our csSVM provides better overall quantification of DILD.

Automatic detection and quantification of tree-in-bud (TIB) opacities from CT scans

This study presents a novel computer-assisted detection (CAD) system for automatically detecting and precisely quantifying abnormal nodular branching opacities in chest computed tomography (CT), termed tree-in-bud (TIB) opacities by radiology literature. The developed CAD system in this study is based on 1) fast localization of candidate imaging patterns using local scale information of the images, and 2) Möbius invariant feature extraction method based on learned local shape and texture properties of TIB patterns. For fast localization of candidate imaging patterns, we use ball-scale filtering and, based on the observation of the pattern of interest, a suitable scale selection is used to retain only small size patterns. Once candidate abnormality patterns are identified, we extract proposed shape features from regions where at least one candidate pattern occupies. The comparative evaluation of the proposed method with commonly used CAD methods is presented with a dataset of 60 chest CTs (laboratory confirmed 39 viral bronchiolitis human parainfluenza CTs and 21 normal chest CTs). The quantitative results are presented as the area under the receiver operator characteristics curves and a computer score (volume affected by TIB) provided as an output of the CAD system. In addition, a visual grading scheme is applied to the patient data by three well-trained radiologists. Interobserver and observer-computer agreements are obtained by the relevant statistical methods over different lung zones. Experimental results demonstrate that the proposed CAD system can achieve high detection rates with an overall accuracy of 90.96%. Moreover, correlations of observer-observer (R(2)=0.8848, and observer-CAD agreements (R(2)=0.824, validate the feasibility of the use of the proposed CAD system in detecting and quantifying TIB patterns.

Comparison of usual interstitial pneumonia and nonspecific interstitial pneumonia: quantification of disease severity and discrimination between two diseases on HRCT using a texture-based automated system

Objective

To evaluate the usefulness of an automated system for quantification and discrimination of usual interstitial pneumonia (UIP) and nonspecific interstitial pneumonia (NSIP).

Materials and Methods

An automated system to quantify six regional high-resolution CT (HRCT) patterns: normal, NL; ground-glass opacity, GGO; reticular opacity, RO; honeycombing, HC; emphysema, EMPH; and consolidation, CONS, was developed using texture and shape features. Fifty-four patients with pathologically proven UIP (n = 26) and pathologically proven NSIP (n = 28) were included as part of this study. Inter-observer agreement in measuring the extent of each HRCT pattern between the system and two thoracic radiologists were assessed in 26 randomly selected subsets using an interclass correlation coefficient (ICC). A linear regression analysis was used to assess the contribution of each disease pattern to the pulmonary function test parameters. The discriminating capacity of the system between UIP and NSIP was evaluated using a binomial logistic regression.

Results

The overall ICC showed acceptable agreement among the system and the two radiologists (r = 0.895 for the abnormal lung volume fraction, 0.706 for the fibrosis fraction, 0.895 for NL, 0.625 for GGO, 0.626 for RO, 0.893 for HC, 0.800 for EMPH, and 0.430 for CONS). The volumes of NL, GGO, RO, and EMPH contribute to forced expiratory volume during one second (FEV₁) (r = 0.72, β values, 0.84, 0.34, 0.34 and 0.24, respectively) and forced vital capacity (FVC) (r = 0.76, β values, 0.82, 0.28, 0.21 and 0.34, respectively). For diffusing capacity (DL_co), the volumes of NL and HC were independent contributors in opposite directions (r = 0.65, β values, 0.64, -0.21, respectively). The automated system can help discriminate between UIP and NSIP with an accuracy of 82%.

Conclusion

The automated quantification system of regional HRCT patterns can be useful in the assessment of disease severity and may provide reliable agreement with the radiologists' results. In addition, this system may be useful in differentiating between UIP and NSIP.

Feasibility of automated quantification of regional disease patterns depicted on high-resolution computed tomography in patients with various diffuse lung diseases

OBJECTIVE

This study was designed to develop an automated system for quantification of various regional disease patterns of diffuse lung diseases as depicted on high-resolution computed tomography (HRCT) and to compare the performance of the automated system with human readers.

MATERIALS AND METHODS

A total of 600 circular regions-of-interest (ROIs), 10 pixels in diameter, were utilized. The 600 ROIs comprised 100 ROIs that represented six typical regional patterns (normal, ground-glass opacity, reticular opacity, honeycombing, emphysema, and consolidation). The ROIs were used to train the automated classification system based on the use of a Support Vector Machine classifier and 37 features of texture and shape. The performance of the classification system was tested with a 5-fold cross-validation method. An automated quantification system was developed with a moving ROI in the lung area, which helped classify each pixel into six categories. A total of 92 HRCT images obtained from patients with different diseases were used to validate the quantification system. Two radiologists independently classified lung areas of the same CT images into six patterns using the manual drawing function of dedicated software. Agreement between the automated system and the readers and between the two individual readers was assessed.

RESULTS

The overall accuracy of the system to classify each disease pattern based on the typical ROIs was 89%. When the quantification results were examined, the average agreement between the system and each radiologist was 52% and 49%, respectively. The agreement between the two radiologists was 67%.

CONCLUSION

An automated quantification system for various regional patterns of diffuse interstitial lung diseases can be used for objective and reproducible assessment of disease severity.