Computer-aided Diagnostic Scheme Research Articles

Computer-Aided Diagnosis of COVID-19 from Chest X-ray Images Using Hybrid-Features and Random Forest Classifier

In recent years, a lot of attention has been paid to using radiology imaging to automatically find COVID-19. (1) Background: There are now a number of computer-aided diagnostic schemes that help radiologists and doctors perform diagnostic COVID-19 tests quickly, accurately, and consistently. (2) Methods: Using chest X-ray images, this study proposed a cutting-edge scheme for the automatic recognition of COVID-19 and pneumonia. First, a pre-processing method based on a Gaussian filter and logarithmic operator is applied to input chest X-ray (CXR) images to improve the poor-quality images by enhancing the contrast, reducing the noise, and smoothing the image. Second, robust features are extracted from each enhanced chest X-ray image using a Convolutional Neural Network (CNNs) transformer and an optimal collection of grey-level co-occurrence matrices (GLCM) that contain features such as contrast, correlation, entropy, and energy. Finally, based on extracted features from input images, a random forest machine learning classifier is used to classify images into three classes, such as COVID-19, pneumonia, or normal. The predicted output from the model is combined with Gradient-weighted Class Activation Mapping (Grad-CAM) visualisation for diagnosis. (3) Results: Our work is evaluated using public datasets with three different train-test splits (70-30%, 80-20%, and 90-10%) and achieved an average accuracy, F1 score, recall, and precision of 97%, 96%, 96%, and 96%, respectively. A comparative study shows that our proposed method outperforms existing and similar work. The proposed approach can be utilised to screen COVID-19-infected patients effectively. (4) Conclusions: A comparative study with the existing methods is also performed. For performance evaluation, metrics such as accuracy, sensitivity, and F1-measure are calculated. The performance of the proposed method is better than that of the existing methodologies, and it can thus be used for the effective diagnosis of the disease.

Melanoma segmentation using deep learning with test-time augmentations and conditional random fields

In a computer-aided diagnostic (CAD) system for skin lesion segmentation, variations in shape and size of the skin lesion makes the segmentation task more challenging. Lesion segmentation is an initial step in CAD schemes as it leads to low error rates in quantification of the structure, boundary, and scale of the skin lesion. Subjective clinical assessment of the skin lesion segmentation results provided by current state-of-the-art deep learning segmentation techniques does not offer the required results as per the inter-observer agreement of expert dermatologists. This study proposes a novel deep learning-based, fully automated approach to skin lesion segmentation, including sophisticated pre and postprocessing approaches. We use three deep learning models, including UNet, deep residual U-Net (ResUNet), and improved ResUNet (ResUNet++). The preprocessing phase combines morphological filters with an inpainting algorithm to eliminate unnecessary hair structures from the dermoscopic images. Finally, we used test time augmentation (TTA) and conditional random field (CRF) in the postprocessing stage to improve segmentation accuracy. The proposed method was trained and evaluated on ISIC-2016 and ISIC-2017 skin lesion datasets. It achieved an average Jaccard Index of 85.96% and 80.05% for ISIC-2016 and ISIC-2017 datasets, when trained individually. When trained on combined dataset (ISIC-2016 and ISIC-2017), the proposed method achieved an average Jaccard Index of 80.73% and 90.02% on ISIC-2017 and ISIC-2016 testing datasets. The proposed methodological framework can be used to design a fully automated computer-aided skin lesion diagnostic system due to its high scalability and robustness.

Classification of Lung Disease in Children by Using Lung Ultrasound Images and Deep Convolutional Neural Network.

Bronchiolitis is the most common cause of hospitalization of children in the first year of life and pneumonia is the leading cause of infant mortality worldwide. Lung ultrasound technology (LUS) is a novel imaging diagnostic tool for the early detection of respiratory distress and offers several advantages due to its low-cost, relative safety, portability, and easy repeatability. More precise and efficient diagnostic and therapeutic strategies are needed. Deep-learning-based computer-aided diagnosis (CADx) systems, using chest X-ray images, have recently demonstrated their potential as a screening tool for pulmonary disease (such as COVID-19 pneumonia). We present the first computer-aided diagnostic scheme for LUS images of pulmonary diseases in children. In this study, we trained from scratch four state-of-the-art deep-learning models (VGG19, Xception, Inception-v3 and Inception-ResNet-v2) for detecting children with bronchiolitis and pneumonia. In our experiments we used a data set consisting of 5,907 images from 33 healthy infants, 3,286 images from 22 infants with bronchiolitis, and 4,769 images from 7 children suffering from bacterial pneumonia. Using four-fold cross-validation, we implemented one binary classification (healthy vs. bronchiolitis) and one three-class classification (healthy vs. bronchiolitis vs. bacterial pneumonia) out of three classes. Affine transformations were applied for data augmentation. Hyperparameters were optimized for the learning rate, dropout regularization, batch size, and epoch iteration. The Inception-ResNet-v2 model provides the highest classification performance, when compared with the other models used on test sets: for healthy vs. bronchiolitis, it provides 97.75% accuracy, 97.75% sensitivity, and 97% specificity whereas for healthy vs. bronchiolitis vs. bacterial pneumonia, the Inception-v3 model provides the best results with 91.5% accuracy, 91.5% sensitivity, and 95.86% specificity. We performed a gradient-weighted class activation mapping (Grad-CAM) visualization and the results were qualitatively evaluated by a pediatrician expert in LUS imaging: heatmaps highlight areas containing diagnostic-relevant LUS imaging-artifacts, e.g., A-, B-, pleural-lines, and consolidations. These complex patterns are automatically learnt from the data, thus avoiding hand-crafted features usage. By using LUS imaging, the proposed framework might aid in the development of an accessible and rapid decision support-method for diagnosing pulmonary diseases in children using LUS imaging.

Diagnosis of common pulmonary diseases in children by X-ray images and deep learning

Acute lower respiratory infection is the leading cause of child death in developing countries. Current strategies to reduce this problem include early detection and appropriate treatment. Better diagnostic and therapeutic strategies are still needed in poor countries. Artificial-intelligence chest X-ray scheme has the potential to become a screening tool for lower respiratory infection in child. Artificial-intelligence chest X-ray schemes for children are rare and limited to a single lung disease. We need a powerful system as a diagnostic tool for most common lung diseases in children. To address this, we present a computer-aided diagnostic scheme for the chest X-ray images of several common pulmonary diseases of children, including bronchiolitis/bronchitis, bronchopneumonia/interstitial pneumonitis, lobar pneumonia, and pneumothorax. The study consists of two main approaches: first, we trained a model based on YOLOv3 architecture for cropping the appropriate location of the lung field automatically. Second, we compared three different methods for multi-classification, included the one-versus-one scheme, the one-versus-all scheme and training a classifier model based on convolutional neural network. Our model demonstrated a good distinguishing ability for these common lung problems in children. Among the three methods, the one-versus-one scheme has the best performance. We could detect whether a chest X-ray image is abnormal with 92.47% accuracy and bronchiolitis/bronchitis, bronchopneumonia, lobar pneumonia, pneumothorax, or normal with 71.94%, 72.19%, 85.42%, 85.71%, and 80.00% accuracy, respectively. In conclusion, we provide a computer-aided diagnostic scheme by deep learning for common pulmonary diseases in children. This scheme is mostly useful as a screening for normal versus most of lower respiratory problems in children. It can also help review the chest X-ray images interpreted by clinicians and may remind possible negligence. This system can be a good diagnostic assistance under limited medical resources.

Retinal Mosaicking with Vascular Bifurcations Detected on Vessel Mask by a Convolutional Network.

Mosaicking of retinal images is potentially useful for ophthalmologists and computer-aided diagnostic schemes. Vascular bifurcations can be used as features for matching and stitching of retinal images. A fully convolutional network model is employed to segment vascular structures in retinal images to detect vascular bifurcations. Then, bifurcations are extracted as feature points on the vascular mask by a robust and efficient approach. Transformation parameters for stitching can be estimated from the correspondence of vascular bifurcations. The proposed feature detection and mosaic method is evaluated on retinal images of 14 different eyes, 62 retinal images. The proposed method achieves a considerably higher average recall rate of matching for paired images compared with speeded-up robust features and scale-invariant feature transform. The running time of our method was also lower than other methods. Results produced by the proposed method superior to that of AutoStitch, photomerge function in Photoshop cs6 and ICE, demonstrate that accurate matching of detected vascular bifurcations could lead to high-quality mosaic of retinal images.

A CAD SCHEME FOR EARLY LUNG CANCER DETECTION IN CHEST RADIOGRAPHY

The purpose of this work is to describe a chest radiography computer-aided diagnostic (CAD) scheme designed to analyze the chest radiographs performed in the framework of the Galician (Spain) Health Service (GHS), including the radiographs that are not reported by the radiologists. The final goal of this CAD system is its integration in the GHS daily clinical environment, with a feasible RIS-PACS-CAD and EHR-CAD integration model. The database of the study included 55 chest radiographies with 64 nodules/lung cancer. This database was used to develop and test the CAD system in our research laboratory. Free-Response Receiver Operating Characteristic (FROC) curves were employed to evaluate the performance of the CAD system. An independent database was employed to evaluate the performance of the CAD system by external radiologists. After the application of a linear classifier, our CAD system achieved a sensitivity of 70% with a false positive rate between 4 and 6 per image depending on the testing database. When compared with other commercial systems, our CAD scheme achieved similar performance results. Therefore, our CAD scheme could be utilized to help radiologists in the detection of lung nodules in chest radiography, and therefore, it can be integrated in the clinical practice.

Cascade of multi-scale convolutional neural networks for bone suppression of chest radiographs in gradient domain.

Suppression of bony structures in chest radiographs (CXRs) is potentially useful for radiologists and computer-aided diagnostic schemes. In this paper, we present an effective deep learning method for bone suppression in single conventional CXR using deep convolutional neural networks (ConvNets) as basic prediction units. The deep ConvNets were adapted to learn the mapping between the gradients of the CXRs and the corresponding bone images. We propose a cascade architecture of ConvNets (called CamsNet) to refine progressively the predicted bone gradients in which the ConvNets work at successively increased resolutions. The predicted bone gradients at different scales from the CamsNet are fused in a maximum-a-posteriori framework to produce the final estimation of a bone image. This estimation of a bone image is subtracted from the original CXR to produce a soft-tissue image in which the bone components are eliminated. Our method was evaluated on a dataset that consisted of 504 cases of real two-exposure dual-energy subtraction chest radiographs (404 cases for training and 100 cases for test). The results demonstrate that our method can produce high-quality and high-resolution bone and soft-tissue images. The average relative mean absolute error of the produced bone images and peak signal-to-noise ratio of the produced soft-tissue images were 3.83% and 38.7dB, respectively. The average bone suppression ratio of our method was 83.8% for the CXRs with pixel sizes of nearly 0.194mm. Furthermore, we apply the trained CamsNet model on the CXRs acquired by various types of X-ray machines, including scanned films, and our method can also produce visually appealing bone and soft-tissue images.

Recent Progress in Development of Computer-Aided Diagnostic (CAD) Schemes in Radiology

Recent Progress in Development of Computer-Aided Diagnostic (CAD) Schemes in Radiology

Hybrid detection of lung nodules on CT scan images.

The diversity of lung nodules poses difficulty for the current computer-aided diagnostic (CAD) schemes for lung nodule detection on computed tomography (CT) scan images, especially in large-scale CT screening studies. We proposed a novel CAD scheme based on a hybrid method to address the challenges of detection in diverse lung nodules. The hybrid method proposed in this paper integrates several existing and widely used algorithms in the field of nodule detection, including morphological operation, dot-enhancement based on Hessian matrix, fuzzy connectedness segmentation, local density maximum algorithm, geodesic distance map, and regression tree classification. All of the adopted algorithms were organized into tree structures with multi-nodes. Each node in the tree structure aimed to deal with one type of lung nodule. The method has been evaluated on 294 CT scans from the Lung Image Database Consortium (LIDC) dataset. The CT scans were randomly divided into two independent subsets: a training set (196 scans) and a test set (98 scans). In total, the 294 CT scans contained 631 lung nodules, which were annotated by at least two radiologists participating in the LIDC project. The sensitivity and false positive per scan for the training set were 87% and 2.61%. The sensitivity and false positive per scan for the testing set were 85.2% and 3.13%. The proposed hybrid method yielded high performance on the evaluation dataset and exhibits advantages over existing CAD schemes. We believe that the present method would be useful for a wide variety of CT imaging protocols used in both routine diagnosis and screening studies.

Breast elastography diagnosis based on dynamic sequence features

Each dynamic elastography contains multiple images; therefore, physicians need to determine the most representative image from the scanned sequence to make diagnosis. To eliminate interobserver variations on diagnoses and help doctors providing correct treatments, the authors developed an objective computer-aided diagnostic scheme without requiring manually selecting representative images for diagnosis. About 112 histological-proven lesions including 66 benign and 46 malignant tumors were involved as the material database. Suspicious lesions were automatically segmented on the first B-mode images on each captured dynamic elastography. Tissue strains inside lesions on elastograms were classified by utilizing the fuzzy c-means algorithm. In order to reduce the influence of image quality, important tumor characteristics were computed from every strain images in elastography and regressed to a probability of being malignant. Since tumor boundaries changed slightly between adjacent slices, a tumor boundary tracking scheme based on template matching was applied on slices excepting the first one. The diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value are 85.71% (96∕112), 86.96% (40∕46), 84.85% (56∕66), 80.00% (40∕50), and 90.32% (56∕62). In addition, the area under the receiver operating characteristic curve is 0.9016. The authors' proposed study provides a reliable computer-aided system which helps physicians to make diagnoses according to features computed from entire elastographic sequence. Experimental results illustrate that the diagnostic scheme is sufficient in distinguishing benign and malignant tumors. It is not necessary for physicians to spend a lot of time to determine the most suitable image for diagnosis. The tumor boundary tracking mechanism effectively eliminates the computation time since it slightly adjusts tumor boundaries between neighboring slices instead of segmenting the tumor contour on each image in the dynamic elastography. The system sufficiently reduces the variations of diagnosis caused by operator dependencies and image qualities and furthermore save physicians' workloads.

MR画像における転移性脳腫瘍検出のコンピュータ支援診断

The fact that accurate detection of metastatic brain tumors is important for making decisions on the treatment course of patients prompted us to develop a computer-aided diagnostic scheme for detecting metastatic brain tumors. In this paper, we first describe how we extracted the cerebral parenchyma region using a standard deviation filter. Second, initial candidates for tumors were decided by sphericity and cross-correlation value with a simulated ring template. Third, we made true positive and false positive templates obtained from actual clinical images and applied the template matching technique to them. Finally, we detected metastatic tumors using these two characteristics. Our improved method was applied to 13 cases with 97 brain metastases. Sensitivity of detection of metastatic brain tumors was 80.4%, with 5.6 false positives per patient. Our proposed method has potential for detection of metastatic brain tumors in brain magnetic resonance (MR) images.

High performance lung nodule detection schemes in CT using local and global information

A key issue in current computer-aided diagnostic (CAD) schemes for nodule detection in CT is the large number of false positives, because current schemes use only global three-dimensional (3D) information to detect nodules and discard useful local two-dimensional (2D) information. Thus, the authors integrated local and global information to markedly improve the performance levels of CAD schemes. Our database was obtained from the standard CT lung nodule database created by the Lung Image Database Consortium (LIDC). It consisted of 85 CT scans with 111 nodules of 3 mm or larger in diameter. The 111 nodules were confirmed by at least two of the four radiologists participated in the LIDC. Twenty-six nodules were missed by two of the four radiologists and were thus very difficult to detect. The authors developed five CAD schemes for nodule detection in CT using global 3D information (3D scheme), local 2D information (2D scheme), and both local and global information (2D + 3D scheme, 2D - 3D scheme, and 3D - 2D scheme). The 3D scheme, which was developed previously, used only global 3D information and discarded local 2D information, as other CAD schemes did. The 2D scheme used a uniform viewpoint reformation technique to decompose a 3D nodule candidate into a set of 2D reformatted images generated from representative viewpoints, and selected and used "effective" 2D reformatted images to remove false positives. The 2D + 3D scheme, 2D - 3D scheme, and 3D - 2D scheme used complementary local and global information in different ways to further improve the performance of lung nodule detection. The authors employed a leave-one-scan-out testing method for evaluation of the performance levels of the five CAD schemes. At the sensitivities of 85%, 80%, and 75%, the existing 3D scheme reported 17.3, 7.4, and 2.8 false positives per scan, respectively; the 2D scheme improved the detection performance and reduced the numbers of false positives to 7.6, 2.5, and 1.3 per scan; the 2D + 3D scheme further reduced those to 2.7, 1.9, and 0.6 per scan; the 2D - 3D scheme reduced those to 7.6, 2.1, and 0.8 per scan; and the 3D - 2D scheme reduced those to 17.3, 1.6, and 1.0 per scan. The local 2D information appears to be more useful than the global 3D information for nodule detection, particularly, when it is integrated with 3D information.

A computerized scheme for lung nodule detection in multiprojection chest radiography

Our previous study indicated that multiprojection chest radiography could significantly improve radiologists' performance for lung nodule detection in clinical practice. In this study, the authors further verify that multiprojection chest radiography can greatly improve the performance of a computer-aided diagnostic (CAD) scheme. Our database consisted of 59 subjects, including 43 subjects with 45 nodules and 16 subjects without nodules. The 45 nodules included 7 real and 38 simulated ones. The authors developed a conventional CAD scheme and a new fusion CAD scheme to detect lung nodules. The conventional CAD scheme consisted of four steps for (1) identification of initial nodule candidates inside lungs, (2) nodule candidate segmentation based on dynamic programming, (3) extraction of 33 features from nodule candidates, and (4) false positive reduction using a piecewise linear classifier. The conventional CAD scheme processed each of the three projection images of a subject independently and discarded the correlation information between the three images. The fusion CAD scheme included the four steps in the conventional CAD scheme and two additional steps for (5) registration of all candidates in the three images of a subject, and (6) integration of correlation information between the registered candidates in the three images. The integration step retained all candidates detected at least twice in the three images of a subject and removed those detected only once in the three images as false positives. A leave-one-subject-out testing method was used for evaluation of the performance levels of the two CAD schemes. At the sensitivities of 70%, 65%, and 60%, our conventional CAD scheme reported 14.7, 11.3, and 8.6 false positives per image, respectively, whereas our fusion CAD scheme reported 3.9, 1.9, and 1.2 false positives per image, and 5.5, 2.8, and 1.7 false positives per patient, respectively. The low performance of the conventional CAD scheme may be attributed to the high noise level in chest radiography, and the small size and low contrast of most nodules. This study indicated that the fusion of correlation information in multiprojection chest radiography can markedly improve the performance of CAD scheme for lung nodule detection.

Study of Aided Diagnosis of Hepatic Carcinoma Based on Artificial Neural Network Combined with Tumor Marker Group

Abstract To develop a computer-aided diagnostic scheme by using an artificial neural network (ANN) combined with tumor markers for diagnosis of hepatic carcinoma (HCC) as a clinical assistant method. 140 serum samples (50 malignant, 40 benign and 50 normal) were analyzed for α-fetoprotein (AFP), carbohydrate antigen 125 (CA125), carcinoembryonic antigen (CEA), sialic acid (SA) and calcium (Ca). The five tumor marker values were then used as ANN inputs data. The result of ANN was compared with that of discriminant analysis by receiver operating characteristic (ROC) curve (AUC) analysis. The diagnostic accuracy of ANN and discriminant analysis among all samples of the test group was 95.5% and 79.3%, respectively. Analysis of multiple tumor markers based on ANN may be a better choice than the traditional statistical methods for differentiating HCC from benign or normal.

Computer‐aided diagnostic (CAD) scheme by use of contralateral subtraction technique. Application to detection of acute cerebral infarctions in brain computed tomography (CT)

AbstractWe developed a computer‐aided diagnostic (CAD) scheme for detection of subtle image findings of acute cerebral infarction in brain computed tomography (CT) by using a contralateral subtraction technique. In our computerized scheme, the lateral inclination of the image was first corrected automatically by rotating and shifting. The contralateral subtraction image was then derived by subtraction of the reversed image from the original image. Initial candidates for acute cerebral infarctions were identified by multiple thresholding and image filtering. As the first step for removing false‐positive candidates, 14 image features were extracted in each of the initial candidates. Halfway candidates were detected by applying a rule‐based test with these image features. At the second step, 5 image features were extracted using the overlapping scale with halfway candidates in the slice of interest and the upper/lower slice images. Finally, acute cerebral infarction candidates were detected by applying a rule‐based test with 5 image features. The sensitivity of detection in 74 training cases was 97.4% with 3.7 false positives per image. The performance of the CAD scheme for 44 testing cases had an approximately equivalent result to the training cases. Our CAD scheme using the contralateral subtraction technique can reveal suspected image findings of acute cerebral infarctions in CT images. © 2011 Wiley Periodicals, Inc. Electron Comm Jpn, 94(2): 32–41, 2011; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ecj.10299

Comparative study of ROC regression techniques—Applications for the computer-aided diagnostic system in breast cancer detection

The receiver operating characteristic (ROC) curve is the most widely used measure for statistically evaluating the discriminatory capacity of continuous biomarkers. It is well known that, in certain circumstances, the markers’ discriminatory capacity can be affected by factors, and several ROC regression methodologies have been proposed to incorporate covariates in the ROC framework. An in-depth simulation study of different ROC regression models and their application in the emerging field of automatic detection of tumour masses is presented. In the simulation study different scenarios were considered and the models were compared to each other on the basis of the mean squared error criterion. The application of the reviewed ROC regression techniques in evaluating computer-aided diagnostic (CAD) schemes can become a major factor in the development of such systems in Radiology.

SU‐FF‐I‐03: Computer‐Aided Diagnostic Scheme for Detection of Hepatocellular Carcinoma in Contrast‐Enhanced Hepatic CT: Preliminary Results

Purpose: Hepatocellular carcinoma (HCC) is one of the most common primary malignant tumors of the liver. We developed a computer‐aided diagnostic (CAD) scheme for detection of HCC in arterial‐phase hepatic CT in order to assist radiologists in their HCC detection. Materials and Methods: Our database consists of isotropic arterial‐phase CT scans of 14 patients acquired with a multi‐detector‐row CT system with a 64 channel detector scanner, which includes 16 radiologist‐determined HCCs that were confirmed pathologically. Lesion sizes ranged 6–50 mm, with a mean of 22.6 mm. We developed a CAD scheme consisting of 1) liver segmentation, using fast‐marching rough‐segmentation followed by geodesic‐active‐contour fine‐segmentation coupled with a level‐set algorithm, 2) detection of HCC candidates from the segmented liver, based on a watershed segmentation algorithm, 3) calculation of both 2D and 3D morphologic‐, intensity‐ and texture‐based features of the detected candidates, and 4) classification of HCC candidates by means of linear discriminant analysis (LDA) of the calculated features, with a stepwise feature selection method based on the Wilks. lambda and the F value. The performance of the CAD scheme was evaluated by free‐response receiver‐operating‐characteristic analysis. Results: With the candidate selection segment of the detection scheme, we achieved 100% sensitivity with 15.1 false positives per patient. Of the 254 calculated features, the stepwise LDA selected 45 “useful” features for candidate characterization. Using these 45 features, we decreased the false positive rate to 6.1 false positives per patient, with no loss in sensitivity. Conclusion: In contrast‐enhanced arterial‐phase hepatic CT, our CAD scheme achieved a 100% sensitivity in detection of HCCs with 6.1 false positives per patient. This detection scheme could assist radiologists in detecting HCCs; thus, it would potentially improve radiologists' detection sensitivity for HCCs.

Computer-aided Diagnosis of Focal Liver Lesions by Use of Physicians' Subjective Classification of Echogenic Patterns in Baseline and Contrast-enhanced Ultrasonography

To develop a computer-aided diagnostic (CAD) scheme for classifying focal liver lesions (FLLs) by use of physicians' subjective classification of echogenic patterns of FLLs on baseline and contrast-enhanced ultrasonography (US). A total of 137 hepatic lesions in 137 patients were evaluated with B-mode and NC100100 (Sonazoid)-enhanced pulse-inversion US; lesions included 74 hepatocellular carcinomas (HCCs) (23: well-differentiated, 36: moderately differentiated, 15: poorly differentiated HCCs), 33 liver metastases, and 30 liver hemangiomas. Three physicians evaluated single images at B-mode and arterial phases with a cine mode. Physicians were asked to classify each lesion into one of eight B-mode and one of eight enhancement patterns, but did not make a diagnosis. To classify five types of FLLs, we employed a decision tree model with four decision nodes and four artificial neural networks (ANNs). The results of the physicians' pattern classifications were used successively for four different ANNs in making decisions at each of the decision nodes in the decision tree model. The classification accuracies for the 137 FLLs were 84.8% for metastasis, 93.3% for hemangioma, and 98.6% for all HCCs. In addition, the classification accuracies for histological differentiation types of HCCs were 65.2% for well-differentiated HCC, 41.7% for moderately differentiated HCC, and 80.0% for poorly differentiated HCC. This CAD scheme has the potential to improve the diagnostic accuracy of liver lesions. However, the accuracy in the histologic differential diagnosis of HCC based on baseline and contrast-enhanced US is still limited.

Computer‐aided diagnosis for the classification of focal liver lesions by use of contrast‐enhanced ultrasonography

The authors developed a computer-aided diagnostic (CAD) scheme for classifying focal liver lesions (FLLs) as liver metastasis, hemangioma, and three histologic differentiation types of hepatocellular carcinoma (HCC), by use of microflow imaging (MFI) of contrast-enhanced ultrasonography. One hundred and three FLLs obtained from 97 cases used in this study consisted of 26 metastases (15 hyper- and 11 hypovascularity types), 16 hemangiomas (five hyper- and 11 hypovascularity types) and 61 HCCs: 24 well differentiated (w-HCC), 28 moderately differentiated (m-HCC), and nine poorly differentiated (p-HCC). Pathologies of all cases were determined based on biopsy or surgical specimens. Locations and contours of FLLs on contrast-enhanced images were determined manually by an experienced physician. MFI was obtained with contrast-enhanced low-mechanical-index (MI) pulse subtraction imaging at a fixed plane which included a distinctive cross section of the FLL. In MFI, the inflow high signals in the plane, which were due to the vascular patterns and the contrast agent, were accumulated following flash scanning with a high-MI ultrasound exposure. In the initial step of our computerized scheme, a series of the MFI images was extracted from the original cine clip (AVI format). We applied a smoothing filter and time-sequential running average techniques in order to reduce signal noise on the single MFI image and cyclic noise on the sequential MFI images, respectively. A kidney, vessels, and a liver parenchyma region were segmented automatically by use of the last image of a series of MFI images. The authors estimated time-intensity curves for an FLL by use of a series of the temporally averaged MFI images in order to determine temporal features such as estimated replenishment times at early and delayed phases, flow rates, and peak times. In addition, they extracted morphologic and gray-level image features which were determined based on the physicians' knowledge of the diagnosis of the FLL, such as the size of lesion, vascular patterns, and the presence of hypoechoic regions. They employed a cascade of six independent artificial neural networks (ANNs) by use of extracted temporal and image features for classifying five types of liver diseases. A total of 16 temporal and image features, which were selected from 43 initially extracted features, were used for six different ANNs for making decisions at each decision in the cascade. The ANNs were trained and tested with a leave-one-lesion-out test method. The classification accuracies for the 103 FLLs were 88.5% for metastasis, 93.8% for hemangioma, and 86.9% for all HCCs. In addition, the classification accuracies for histologic differentiation types of HCCs were 79.2% for w-HCC, 50.0% for m-HCC, and 77.8% for p-HCC. The CAD scheme for classifying FLLs by use of the MFI on contrast-enhanced ultrasonography has the potential to improve the diagnostic accuracy in the histologic diagnosis of HCCs and the other liver diseases.

Computer-aided diagnosis in chest radiography for detection of childhood pneumonia

Objectives This article presents a novel approach based on computer-aided diagnostic (CAD) scheme and wavelet transforms to aid pneumonia diagnosis in children, using chest radiograph images. The prototype system, named Pneumo-CAD, was designed to classify images into presence (PP) or absence of pneumonia (PA). Materials and methods The knowledge database for the Pneumo-CAD comprised chest images confirmed as PP or PA by two radiologists trained to interpret chest radiographs according to the WHO guidelines for the diagnosis of pneumonia in children. The performance of the Pneumo-CAD was evaluated by a subset of images randomly selected from the knowledge database. The retrieval of similar images was made by feature extraction using wavelets transform coefficients of the image. The energy of the wavelet coefficients was used to compose the feature vector in order to support the computational classification of images as PP or PA. Methodology I worked with a rank-weighted 15-nearest-neighbour scheme, while methodology II employed a distance-dependent weighting for image classification. The performance of the prototype system was assessed by the ROC curve. Results Overall, the Pneumo-CAD using the Haar wavelet presented the best accuracy in discriminating PP from PA for both, methodology I (AUC = 0.97) and methodology II (AUC = 0.94), reaching sensitivity of 100% and specificity of 80% and 90%, respectively. Conclusion Pneumo-CAD could represent a complementary tool to screen children with clinical suspicion of pneumonia, and so to contribute to gather information on the burden of-pneumonia estimates in order to help guide health policies toward preventive interventions.