Search

The emu is the second largest ratite; thus, their sera and egg yolks, obtained after immunization, could provide therapeutic and diagnostically important immunoglobulins with improved production efficiency. Reliable purification tools are required to establish a pipeline for supplying practical emu-derived antibodies, the majority of which belongs to the immunoglobulin Y (IgY) class. Therefore, we generated a monoclonal secondary antibody specific to emu IgY. Initially, we immunized an emu with bovine serum albumin multiply haptenized with 2,4-dinitrophenyl (DNP) groups. Polyclonal emu anti-DNP antibodies were partially purified using conventional precipitation method and used as antigen for immunizing a BALB/c mouse. Splenocytes were fused with myeloma cells and a hybridoma clone secreting a desirable secondary antibody (mAb#2-16) was established. The secondary antibody bound specifically to emu-derived IgY, distinguishing IgYs from chicken, duck, ostrich, quail, and turkey, as well as human IgGs. Affinity columns immobilizing the mAb#2-16 antibodies enabled purification of emu IgY fractions from sera and egg yolks via simple protocols, with which we succeeded in producing IgYs specific to the severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) spike protein with a practical binding ability. We expect that the presented purification method, and the secondary antibody produced in this study, will facilitate the utilization of emus as a novel source of therapeutic and diagnostic antibodies.

Chronic obstructive pulmonary disease (COPD) is predicted to become the third leading cause of death worldwide by 2030. A longitudinal study using CT scans of COPD is useful to assess the changes in structural abnormalities. In this study, we performed visualization and unsupervised clustering of emphysema progression using t-distributed stochastic neighbor embedding (t-SNE) analysis of longitudinal CT images, smoking history, and SNPs. The procedure of this analysis is as follows: (1) automatic segmentation of lung lobes using 3D U-Net, (2) quantitative image analysis of emphysema progression in lung lobes, and (3) visualization and unsupervised clustering of emphysema progression using t-SNE. Nine explanatory variables were used for the clustering: genotypes at two SNPs (rs13180 and rs3923564), smoking history (smoking years, number of cigarettes per day, pack-year), and LAV distribution (LAV size and density in upper lobes, LAV size, and density in lower lobes). The objective variable was emphysema progression which was defined as the annual change in low attenuation volume (LAV%/year) using linear regression. The nine-dimensional space was transformed to two-dimensional space by t-SNE, and divided into three clusters by Gaussian mixture model. This method was applied to 37 smokers with 68.2 pack-years and 97 past smokers with 51.1 pack-years. The results demonstrated that this method could be effective for quantitative assessment of emphysema progression by SNPs, smoking history, and imaging features.

Progression of thoracic aortic calcification (TAC) has been shown to be associated with hard cardiovascular events including stroke and all-cause mortality as well as coronary events. In this study, we propose an automated detection method of TACs of non-contrast CT images using mediastinal anatomical label map. This method consists of two steps: (1) the construction of a mediastinal anatomical label map, and (2) the detection of TACs using the intensity and the mediastinal anatomical label map. The proposed method was applied to two non-contrast CT image datasets: 24 cases of chronic thromboembolic pulmonary hypertension (CTEPH) and 100 non-CTEPH cases of low-dose CT screening. The method was compared with two-dimensional U-Nets and the Swin UNETR. The results showed that the method achieved significantly higher F1 score of 0.937 than other methods for the non-CTEPH case dataset (p-value < 0.05, pairwise Wilcoxon signed rank test with Bonferroni correction).

A micro-CT system was developed using a 36M-pixel digital single-lens reflex camera as a cost-effective mode for large human lung specimen imaging. Scientific grade cameras used for biomedical x-ray imaging are much more expensive than consumer-grade cameras. During the past decade, advances in image sensor technology for consumer appliances have spurred the development of biomedical x-ray imaging systems using commercial digital single-lens reflex cameras fitted with high megapixel CMOS image sensors. This micro-CT system is highly specialized for visualizing whole secondary pulmonary lobules in a large human lung specimen. The secondary pulmonary lobule, a fundamental unit of the lung structure, reproduces the lung in miniature. The lung specimen is set in an acrylic cylindrical case of 36 mm diameter and 40 mm height. A field of view (FOV) of the micro-CT is 40.6 mm wide × 15.1 mm high with 3.07 μm pixel size using offset CT scanning for enlargement of the FOV. We constructed a 13,220 × 13,220 × 4912 voxel image with 3.07 μm isotropic voxel size for three-dimensional visualization of the whole secondary pulmonary lobule. Furthermore, synchrotron radiation has proved to be a powerful high-resolution imaging tool. This micro-CT system using a single-lens reflex camera and synchrotron radiation provides practical benefits of high-resolution and wide-field performance, but at low cost.

Chronic obstructive pulmonary disease (COPD) is predicted to become the third leading cause of death worldwide by 2030. Smoking is a well-known risk factor in the development of COPD. Association between COPD genes and smoking have been studied. This paper presents an association analysis of single nucleotide polymorphisms (SNPs) with a CT image-based phenotype of emphysema progression in heavy smokers. The emphysema progression was quantitatively represented by the annual increment of low attenuation volume (LAV) on CT scans for five years. 10 candidate SNPs were selected from 316 SNPs in 125 papers of genetic studies of COPD and lung cancer. The genotypes were determined by real-time polymerase chain reaction (PCR) using deoxyribonucleic acid (DNA) extracted from saliva samples. The association analysis was performed by Fisher's exact test and logistic regression analysis. This method was applied to a dataset with 144 participants (71 smokers, 61 past smokers, and 12 non-smokers). The results showed that the genotypes of rs3923564 and rs13180 SNPs were candidate SNPs associated with the CT image based-emphysema progression.

Enlargement of the pulmonary artery is a morphological abnormality of pulmonary hypertension patients. Diameters of the aorta and main pulmonary artery (MPA) are useful for predicting the presence of pulmonary hypertension. A major problem in the automatic segmentation of the aorta and MPA from non-contrast CT images is the invisible boundary caused by contact with blood vessels. In this study, we applied U-Net to the segmentation of the aorta and MPA from non-contrast CT images for normal and chronic thromboembolic pulmonary hypertension (CTEPH) cases and evaluated the robustness to the contacts between blood vessels. Our approach of the segmentation consists of three steps: (1) detection of trachea branch point, (2) cropping region of interest centered to the trachea branch point, and (3) segmentation of the aorta and MPA using U-Net. The segmentation performances were compared in seven methods: 2D U-Net, 2D U-Net with pre-trained VGG-16 encoder, 2D U-Net with pre-trained VGG-19 encoder, 2D Attention U-Net, 3D U-Net, an ensemble method of them, and our conventional method. The aorta and MPA segmentation methods using these U-Net achieved higher performance than a conventional method. Although the contact boundaries of blood vessels caused lower performance compared with the non-contact boundaries, the mean boundary distances were below about one pixel.

Kenichi Goto,1,2 Emiko Ogawa,1,3 Kaoruko Shimizu,4 Hironi Makita,4 Hidenobu Suzuki,5 Yoshiki Kawata,5 Noboru Niki,5 Masaharu Nishimura,4 Yasutaka Nakano1 1Division of Respiratory Medicine, Department of Medicine, Shiga University of Medical Science, Otsu, Shiga, 2Department of Pulmonary Medicine, Takatsuki Red Cross Hospital, Takatsuki, Osaka, 3Health Administration Center, Shiga University of Medical Science, Otsu, Shiga, 4First Department of Medicine, Hokkaido University Hospital, Sapporo, Hokkaido, 5Institute of Technology and Science, Tokushima University, Tokushima, Tokushima, Japan Background: Osteoporosis is a well-known comorbidity in COPD. It is associated with poor health status and prognosis. Although the exact pathomechanisms are unclear, osteoporosis is suggested to be either a comorbidity due to shared risk factors with COPD or a systematic effect of COPD with a cause–effect relationship. This study aimed to evaluate whether progression of osteoporosis is synchronized with that of COPD. Materials and methods: Data from 103 patients with COPD included in the Hokkaido COPD cohort study were analyzed. Computed tomography (CT) attenuation values of thoracic vertebrae 4, 7, and 10 were measured using custom software, and the average value (average bone density; ABD4,7,10) was calculated. The percentage of low attenuation volume (LAV%) for each patient was also calculated for evaluation of emphysematous lesions. Annual change in thoracic vertebral CT attenuation, which is strongly correlated with dual-energy X-ray absorptiometry-measured bone mineral density, was compared with that in FEV1.0 or emphysematous lesions. Results: In the first CT data set, ABD4,7,10 was significantly correlated with age (ρ=–0.331; p=0.0006), body mass index (BMI;ρ=0.246; p=0.0136), St George’s Respiratory Questionnaire (SGRQ) activity score (ρ=–0.248; p=0.0115), eosinophil count (ρ=0.229; p=0.0198), and LAV% (ρ=–0.372; p=0.0001). However, ABD4,7,10 was not associated with FEV1.0. After adjustment for age, BMI, SGRQ activity score, and eosinophil count, no significant relationship was found between ABD4,7,10 and LAV%. Annual change in ABD4,7,10 was not associated with annual change in LAV% or FEV1.0. Conclusion: Progression of osteoporosis and that of COPD are not directly related or synchronized with each other. Keywords: osteoporosis, quantitative CT analysis, FEV1.0, low attenuation volume, automatic analysis, systemic effect

Chronic thromboembolic pulmonary hypertension (CTEPH) is characterized by obstruction of the pulmonary vasculature by residual organized thrombi. A morphological abnormality inside mediastinum of CTEPH patient is enlargement of pulmonary artery. This paper presents an automated assessment of aortic and main pulmonary arterial diameters for predicting CTEPH in low-dose CT lung screening. The distinctive feature of our method is to segment aorta and main pulmonary artery using both of prior probability and vascular direction which were estimated from mediastinal vascular region using principal curvatures of four-dimensional hyper surface. The method was applied to two datasets, 64 low-dose CT scans of lung cancer screening and 19 normal-dose CT scans of CTEPH patients through the training phase with 121 low-dose CT scans. This paper demonstrates effectiveness of our method for predicting CTEPH in low-dose CT screening.

Rationale and Objectives The aims of this study were to assess the influence of slice thickness on the ability of radiologists to detect or not detect nodules and to agree or disagree on the diagnosis and also to investigate the potential dependence of these relations on the sizes, average computed tomographic (CT) values, and locations of the nodules. Materials and Methods Six radiologists performed qualitative diagnostic readings of multislice CT images with a slice thickness of 2 or 10 mm obtained from 360 subjects. The nodules were diagnosed as nodules for further examination (NFEs), inactive nodules for no further examination (INNFEs), or no abnormality. The results of the diagnoses were cross-tabulated and quantitatively analyzed using the average CT values, sizes, and locations of the nodules with reference to the 2-mm slices. Multivariate logistic regression analyses were used to estimate the significant associations of these parameters with the ability of radiologists to detect or not detect nodules and to agree or disagree on the diagnosis. Results Totals of 130 NFEs and 403 INNFEs for 2-mm slice thickness and 142 NFEs and 338 INNFEs for 10-mm slice thickness were diagnosed. Nodule classifications were as follows: the same diagnosis on both slice thickness images (67.6%), different diagnoses on two slice thickness images (21%), missed on 10-mm slice thickness images (10.6%), and misinterpreted on 10-mm slice thickness images (0.7%). Regarding detection and nondetection, NFE diagnoses were influenced by size (odds ratio [OR], 132.50; 95% confidence interval [CI], 4.77–4711) and the average CT value (OR, 27.20; 95% CI, 3.21–645.3), and INNFE diagnoses were influenced by size (OR, 16.10; 95% CI, 6.18–55.19) and the average CT value (OR, 7.67; 95% CI, 2.19–30.91). Regarding diagnostic agreement and disagreement, the NFE diagnoses were influenced by size (OR, 3.60; 95% CI, 1.29–11.04), nodule distance from the lung border (OR, 2.85; 95% CI, 1.27–6.65), and nodule location in the right upper lobe (OR, 0.07; 95% CI, 0.003–0.477), while the INNFE diagnoses were influenced by the average CT value (OR, 11.84; 95% CI, 3.33–55.86), size (OR, 0.42; 95% CI, 0.25–0.70), and nodule distance from the lung border (OR, 0.41; 95% CI, 0.25–0.66). Conclusions The influence of slice thickness on the ability of radiologists to detect or not detect nodules and to agree or disagree on the diagnosis was quantitatively evaluated. Detection and nondetection of NFEs and INNFEs are influenced by size and average CT value. Agreement and disagreement on NFE and INNFE diagnoses are influenced not only by size and average CT value but also, importantly, by the locations of nodules.

Rationale and Objectives. An example-based assisting approach that supports decision making in classifying pulmonary nodules in 3-dimensional (3D) thoracic computed tomography images has been developed. Materials and Methods. The example-based assisting approach retrieves and displays nodules that exhibit morphologic and internal profiles consistent to the nodule in question. It uses a 3D computed tomography image database containing 143 pulmonary nodules for which diagnosis is known. The central module makes possible analysis of the query nodule image and extraction of the features of interest: shape, surrounding structure, and internal structure of the nodules. The principal axes and the compactness characterize the nodule shape. The surrounding and internal structures are represented by the distribution pattern of computed tomography density value and 3D curvature indexes. The nodule representation is then used for computing a similarity measure such as a correlation coefficient and a malignant likelihood of the query nodule. The malignant likelihood is estimated by the difference between the representation patterns of the query case and the retrieved lesions. The Mahalanobis distance was adopted as the difference measure. The approach performance was assessed through leave-one-out method by the false-positive rate. Results. Given a query nodule image, the proposed method retrieved benign and malignant images similar to the query case and provided its malignant likelihood. The number of cases that obtained enough number of the retrieved cases for estimating the malignant likelihood was 107 cases (malignant, 70; benign, 37) in our database. Sensitivity was 91.4% (64 of 70 malignant nodules), specificity was 51.4% (19 of 37 benign nodules), and accuracy values were 77.6% (83 of 107 nodules). Conclusion. Preliminary assessment of this approach showed that an example-based assisting approach is an effective tool for making the diagnostic decision in the classification of pulmonary nodules using the nodule image database.

240,000 participants have a screening for diagnosis of pneumoconiosis every year in Japan. Radiograph is used for staging of severity in pneumoconiosis worldwide. This paper presents a method for quantitative assessment of severity in pneumoconiosis using both size and frequency of lung nodules that detected by thin-section CT images. This method consists of three steps. First, thoracic organs (body, ribs, spine, trachea, bronchi, lungs, heart, and pulmonary blood vessels) are segmented. Second, lung nodules that have radius over 1.5mm are detected. These steps used functions of our developed computer aided detection system of chest CT images. Third, severity in pneumoconiosis is quantified using size and frequency of lung nodules. This method was applied to nine pneumoconiosis patients. The initial results showed that proposed method can assess severity in pneumoconiosis quantitatively. This paper demonstrates effectiveness of our method in diagnosis and prognosis of pneumoconiosis in CT screening.

The aims of this study were to assess the influence of slice thickness on the ability of radiologists to detect or not detect nodules and to agree or disagree on the diagnosis and also to investigate the potential dependence of these relations on the sizes, average computed tomographic (CT) values, and locations of the nodules.Six radiologists performed qualitative diagnostic readings of multislice CT images with a slice thickness of 2 or 10 mm obtained from 360 subjects. The nodules were diagnosed as nodules for further examination (NFEs), inactive nodules for no further examination (INNFEs), or no abnormality. The results of the diagnoses were cross-tabulated and quantitatively analyzed using the average CT values, sizes, and locations of the nodules with reference to the 2-mm slices. Multivariate logistic regression analyses were used to estimate the significant associations of these parameters with the ability of radiologists to detect or not detect nodules and to agree or disagree on the diagnosis.Totals of 130 NFEs and 403 INNFEs for 2-mm slice thickness and 142 NFEs and 338 INNFEs for 10-mm slice thickness were diagnosed. Nodule classifications were as follows: the same diagnosis on both slice thickness images (67.6%), different diagnoses on two slice thickness images (21%), missed on 10-mm slice thickness images (10.6%), and misinterpreted on 10-mm slice thickness images (0.7%). Regarding detection and nondetection, NFE diagnoses were influenced by size (odds ratio [OR], 132.50; 95% confidence interval [CI], 4.77-4711) and the average CT value (OR, 27.20; 95% CI, 3.21-645.3), and INNFE diagnoses were influenced by size (OR, 16.10; 95% CI, 6.18-55.19) and the average CT value (OR, 7.67; 95% CI, 2.19-30.91). Regarding diagnostic agreement and disagreement, the NFE diagnoses were influenced by size (OR, 3.60; 95% CI, 1.29-11.04), nodule distance from the lung border (OR, 2.85; 95% CI, 1.27-6.65), and nodule location in the right upper lobe (OR, 0.07; 95% CI, 0.003-0.477), while the INNFE diagnoses were influenced by the average CT value (OR, 11.84; 95% CI, 3.33-55.86), size (OR, 0.42; 95% CI, 0.25-0.70), and nodule distance from the lung border (OR, 0.41; 95% CI, 0.25-0.66).The influence of slice thickness on the ability of radiologists to detect or not detect nodules and to agree or disagree on the diagnosis was quantitatively evaluated. Detection and nondetection of NFEs and INNFEs are influenced by size and average CT value. Agreement and disagreement on NFE and INNFE diagnoses are influenced not only by size and average CT value but also, importantly, by the locations of nodules.

The purpose of this study is to develop an image-guided decision support system that assists decision-making in clinical differential diagnosis of pulmonary nodules. This approach retrieves and displays nodules that exhibit morphological and internal profiles consistent to the nodule in question. It uses a three-dimensional (3-D) CT image database of pulmonary nodules for which diagnosis is known. In order to build the system, there are following issues that should be solved, (1) to categorize the nodule database with respect to morphological and internal features, (2) to quickly search nodule images similar to an indeterminate nodule from a large database, and (3) to reveal malignancy likelihood computed by using similar nodule images. Especially, the first problem influences the design of other issues. The successful categorization of nodule pattern might lead physicians to find important cues that characterize benign and malignant nodules. This paper focuses on an approach to categorize the nodule database with respect to nodule shape and CT density patterns inside nodule.

An example-based assisting approach that supports decision making in classifying pulmonary nodules in 3-dimensional (3D) thoracic computed tomography images has been developed.The example-based assisting approach retrieves and displays nodules that exhibit morphologic and internal profiles consistent to the nodule in question. It uses a 3D computed tomography image database containing 143 pulmonary nodules for which diagnosis is known. The central module makes possible analysis of the query nodule image and extraction of the features of interest: shape, surrounding structure, and internal structure of the nodules. The principal axes and the compactness characterize the nodule shape. The surrounding and internal structures are represented by the distribution pattern of computed tomography density value and 3D curvature indexes. The nodule representation is then used for computing a similarity measure such as a correlation coefficient and a malignant likelihood of the query nodule. The malignant likelihood is estimated by the difference between the representation patterns of the query case and the retrieved lesions. The Mahalanobis distance was adopted as the difference measure. The approach performance was assessed through leave-one-out method by the false-positive rate.Given a query nodule image, the proposed method retrieved benign and malignant images similar to the query case and provided its malignant likelihood. The number of cases that obtained enough number of the retrieved cases for estimating the malignant likelihood was 107 cases (malignant, 70; benign, 37) in our database. Sensitivity was 91.4% (64 of 70 malignant nodules), specificity was 51.4% (19 of 37 benign nodules), and accuracy values were 77.6% (83 of 107 nodules).Preliminary assessment of this approach showed that an example-based assisting approach is an effective tool for making the diagnostic decision in the classification of pulmonary nodules using the nodule image database.

We are developing computerized feature extraction and classification methods to analyze malignant and benign pulmonary nodules in three-dimensional (3-D) thoracic CT images. Internal structure features were derived from CT density and 3-D curvatures to characterize the inhomogeneous of CT density distribution inside the nodule. In the classification step, we combined an unsupervised k-means clustering (KMC) procedure and a supervised linear discriminate (LD) classifier. The KMC procedure classified the sample nodules into two classes by using the mean CT density values for two different regions such as a core region and a complement of the core region in 3-D nodule image. The LD classifier was designed for each class by using internal structure features. The stepwise procedure was used to select the best feature subset from multi-dimensional feature spaces. The discriminant scores output from the classifier were analyzed by receiver operating characteristic (ROC) method and the classification accuracy was quantified by the area, Az, under the ROC curve. We analyzed a data set of pulmonary nodules in this study. The results of this study indicate the potential of combining the KMC procedure and the LD classifier for computer-aided classification of pulmonary nodules.

The objective of this study is to develop a computer-aided diagnosis (CAD) system to support comparative reading of sequential helical CT images for lung cancer screening without using film displays. The placement of pulmonary shadows between sequential helical CT images sometimes differs due to the changes in lung size and shape caused by inspiration. The proposed algorithm consists of two sections; identification of region of interest and the comparison of sequential CT images. We validated the effectiveness of the algorithm by its application to images from 60 subjects. The algorithm could compare the slice images correctly with respect to a physician's point of view. The experimental results indicate that the proposed algorithm is useful in increasing the efficiency of the mass screening process.

This paper presents a method to analyze volume evolutions of pulmonary nodules for discrimination between malignant and benign nodules. Our method consists of four steps; (1) The 3-D rigid registration of the two successive 3-D thoracic CT images, (2) the 3-D affine registration of the two successive region-of-interest (ROI) images, (3) non-rigid registration between local volumetric ROIs, and (4) analysis of the local displacement field between successive temporal images In preliminary study, the method was applied to the successive 3-D thoracic images of two pulmonary lesions including a metastasis malignant case and a inflammatory benign to quantify the evolving process in the pulmonary nodules and surrounding structure. The time intervals between successive 3-D thoracic images for the benign and malignant cases were 150 and 30 days, respectively. From the display of the displacement fields and the contrasted image by the vector field operator based on the Jacobian, it was observed that the benign case reduced in the volume and the surrounding structure was involved into the nodule in the evolution process. It was also observed that the malignant case expanded in the volume. These experimental results indicate that our method is a promising tool to quantify how the lesions evolve their volume and surrounding structures.

This paper focuses on an approach for characterizing the internal structure which is one of important clues for differentiating between malignant and benign nodules in three-dimensional (3-D) thoracic images. In this approach, each voxel was described in terms of shape index derived from curvatures on the voxel. The voxels inside the nodule were aggregated via shape histogram to quantify how much shape category was present in the nodule. Topological features were introduced to characterize the morphology of the cluster constructed from a set of voxels with the same shape category. The properties such as curvedness and CT density were also built into the representation. In the classification step, a hybrid unsupervised/supervised structure was performed to improve the classifier performance. It combined the k-means clustering procedure and the linear discriminate (LD) classifier. The performance of the hybrid classifier was compared to that of LD classifier alone. Receiver operating characteristics (ROC) analysis was used to evaluate the accuracy of the classifiers. We also compared the performance of the hybrid classifier with those of physicians. The classification performance reached the performance of physicians. Our results demonstrate the feasibility of the hybrid classifier based on the topological and histogram features to assist physicians in making diagnostic decisions.

This paper presents a scheme to analyze surrounding structures of pulmonary nodules by using differential geometry based vector fields. In this scheme the differential characteristics such as the principal curvatures and directions are computed from the differential values of the isointensity surfaces. Each voxel in the nodule surrounding is described in terms of shape index and curvedness derived from the principal curvatures. Two vector fields are formed from the directions of the maximum principal curvatures of nodule surrounding and gradient vectors of nodule surface, respectively. The gradient vector field is computed by diffusing the gradient vector on the nodule surface. The regions corresponding to the cylindrical or conic figures which are similar to vessel and plural images are segmented by the shape index and curvedness values. Then, the relationship between the segmented regions and the nodule is evaluated by the inner product of the direction of the maximum principal curvature and the gradient vector. We demonstrate the feasibility of the scheme to classify benign and malignant nodules.

To diagnose the lung cancer as to determine if it has malignant or benign nature, it is important to understand the spatial relationship among the abnormal nodule and other pulmonary organs. But the lung field has very complicated structure, so it is difficult to understand the connectivity of the pulmonary organs using thin-section CT images. This method consists of two parts. The first is the classification of the pulmonary structure based on the anatomical information. The second is the quantitative analysis that is then applicable to differential diagnosis, such as differentiation of malignant or benign abnormal tissue.

This paper examines the problem of obtaining a representation of the three-dimensional(3-D) pulmonary nodule images, which is a key problem in discriminating benign and malignant nodules for differential diagnosis of the lung cancer using thin-section CT images. A curvature based approach is developed with the aim of characterizing internal intensity structures of benign and malignant nodules. This approach makes use of curvature indexes to represent locally each voxel in a three-dimensional (3-D) pulmonary nodule image. From the distribution of curvature indexes and CT value over the 3-D pulmonary nodule image a set of histogram features is computed for global characterization of benign and malignant nodules. Linear discriminant analysis is used for classification and leave-one-out method is used to evaluate the classification accuracy. Compared with the performance of experienced physicians the potential usefulness of the curvature based features in the computer-aided differential diagnosis is demonstrated by using receiver operating characteristic (ROC) curves as the performance measure.

In this paper, we present a computer assisted automatic diagnostic system for lung cancer that detects nodule candidates at an early stage from helical CT images of the thorax. Our diagnostic system consists of analytic and diagnostic procedures. In the analysis procedure, we extract the lung and the blood vessel regions using the fuzzy clustering algorithm, then we analyze the features of these regions. In the diagnosis procedure, we define diagnostic rules utilizing the extracted features which support the determination of the candidate nodule locations.

Lung cancer is known as one of the most difficult cancers to cure. In order to improve the recovery rate for lung cancer, detection and treatment at an early stage of growth is necessary. For the purpose, chest CT images obtained by helical CT scanner have drawn interest in the detection of suspicious regions. This paper present a method to improve the detection accuracy of the system that we have developed to detect candidates of lung cancer based on helical CT images. The basic improvement of the system is to enhance the difference of intensity surfaces between suspicious and normal regions by using surface curvatures such as the Gaussian and the mean curvatures. Experiments to show its feasibility of improving the detection accuracy are demonstrated by applying the method to the chest CT images.

Automated distinction of medical images is an important preprocessing in Computer-Aided Diagnosis (CAD) systems. The CAD systems have been developed using medical image sets with specific scan conditions and body parts. However, varied examinations are performed in medical sites. The specification of the examination is contained into DICOM textual meta information. Most DICOM textual meta information can be considered reliable, however the body part information cannot always be considered reliable. In this paper, we describe an automated distinction of DICOM images as a preprocessing for lung cancer CAD system. Our approach uses DICOM textual meta information and low cost image processing. Firstly, the textual meta information such as scan conditions of DICOM image is distinguished. Secondly, the DICOM image is set to distinguish the body parts which are identified by image processing. The identification of body parts is based on anatomical structure which is represented by features of three regions, body tissue, bone, and air. The method is effective to the practical use of lung cancer CAD system in medical sites.

We have developed an anonymization system for DICOM images. It requires consent from the patient to use the DICOM images for research or education. However, providing the DICOM image to the other facilities is not safe because it contains a lot of personal data. Our system is a server that provides anonymization service of DICOM images for users in the facility. The distinctive features of the system are, input interface, flexible anonymization policy, and automatic body part identification. In the first feature, we can use the anonymization service on the existing DICOM workstations. In the second feature, we can select a best policy fitting for the Protection of personal data that is ruled by each medical facility. In the third feature, we can identify the body parts that are included in the input image set, even if the set lacks the body part tag in DICOM header. We installed the system for the first time to a hospital in December 2005. Currently, the system is working in other four facilities. In this paper we describe the system and how it works.