Rie Tachibana

Classification of diffuse lung disease patterns on high-resolution computed tomography by a bag of words approach

Visual inspection of diffuse lung disease (DLD) patterns on high-resolution computed tomography (HRCT) is difficult because of their high complexity. We proposed a bag of words based method on the classification of these textural patters in order to improve the detection and diagnosis of DLD for radiologists. Six kinds of typical pulmonary patterns were considered in this work. They were consolidation, ground-glass opacity, honeycombing, emphysema, nodular and normal tissue. Because they were characterized by both CT values and shapes, we proposed a set of statistical measure based local features calculated from both CT values and the eigen-values of Hessian matrices. The proposed method could achieve the recognition rate of 95.85%, which was higher comparing with one global feature based method and two other CT values based bag of words methods.

Classification of diffuse lung diseases patterns by a sparse representation based method on HRCT images

This paper describes a computer-aided diagnosis (CAD) method to classify diffuse lung diseases (DLD) patterns on HRCT images. Due to the high variety and complexity of DLD patterns, the performance of conventional methods on recognizing DLD patterns featured by geometrical information is limited. In this paper, we introduced a sparse representation based method to classify normal tissues and five types of DLD patterns including consolidation, ground-glass opacity, honeycombing, emphysema and nodular. Both CT values and eigenvalues of Hessian matrices were adopted to calculate local features. The 2360 VOIs from 117 subjects were separated into two independent set. One set was used to optimize parameters, and the other set was adopted to evaluation. The proposed technique has a overall accuracy of 95.4%. Experimental results show that our method would be useful to classify DLD patterns on HRCT images.

Semantic characteristics prediction of pulmonary nodule using Artificial Neural Networks

Since it is difficult to choose which computer calculated features are effective to predict the malignancy of pulmonary nodules, in this study, we add a semantic-level of Artificial Neural Networks (ANNs) structure to improve intuition of features selection. The works of this study include two: 1) seeking the relationships between computer-calculated features and medical semantic concepts which could be understood by human; 2) providing an objective assessment method to predict the malignancy from semantic characteristics. We used 60 thoracic CT scans collected from the Lung Image Database Consortium (LIDC) database, in which the suspicious lesions had been delineated and annotated by 4 radiologists independently. Corresponding to the two works of this study, correlation analysis experiment and agreement experiment were performed separately.

Automatic detection of GGO regions on CT images in LIDC dataset based on statistical features

Detection of pulmonary nodules with ground glass opacity (GGO) is a difficult task in radiology. Follow up is often required in medical fields. But diagnosis based on CT images are dependent on ability and experience of radiologists. In addition to that, enormous number of images increase their burden. So, to improve the detection accuracy and to reduce the burden of doctors, a CAD (Computer Aided Diagnosis) system is expected. So, in this paper, we propose an automatic algorithm for GGO detection on CT images. At first, vessel areas are removed from original CT images by using 3D Line Filter and then candidate regions are detected by threshold processing. After that, we calculate statistical features of segmented candidate regions and use artificial neural network (ANN) to distinguish final candidate regions. We applied the proposed method to 31 CT image sets in the Lung Image Database Consortium (LIDC) which is supplied by National Center Institute (NCI). In this paper, we show the experimental results and give discussions.

Particle System Based Adaptive Sampling on Spherical Parameter Space to Improve the MDL Method for Construction of Statistical Shape Models

Minimum description length (MDL) based group-wise registration was a state-of-the-art method to determine the corresponding points of 3D shapes for the construction of statistical shape models (SSMs). However, it suffered from the problem that determined corresponding points did not uniformly spread on original shapes, since corresponding points were obtained by uniformly sampling the aligned shape on the parameterized space of unit sphere. We proposed a particle-system based method to obtain adaptive sampling positions on the unit sphere to resolve this problem. Here, a set of particles was placed on the unit sphere to construct a particle system whose energy was related to the distortions of parameterized meshes. By minimizing this energy, each particle was moved on the unit sphere. When the system became steady, particles were treated as vertices to build a spherical mesh, which was then relaxed to slightly adjust vertices to obtain optimal sampling-positions. We used 47 cases of (left and right) lungs and 50 cases of livers, (left and right) kidneys, and spleens for evaluations. Experiments showed that the proposed method was able to resolve the problem of the original MDL method, and the proposed method performed better in the generalization and specificity tests.

A sparse representation based method to classify pulmonary patterns of diffuse lung diseases.

We applied and optimized the sparse representation (SR) approaches in the computer-aided diagnosis (CAD) to classify normal tissues and five kinds of diffuse lung disease (DLD) patterns: consolidation, ground-glass opacity, honeycombing, emphysema, and nodule. By using the K-SVD which is based on the singular value decomposition (SVD) and orthogonal matching pursuit (OMP), it can achieve a satisfied recognition rate, but too much time was spent in the experiment. To reduce the runtime of the method, the K-Means algorithm was substituted for the K-SVD, and the OMP was simplified by searching the desired atoms at one time (OMP1). We proposed three SR based methods for evaluation: SR1 (K-SVD+OMP), SR2 (K-Means+OMP), and SR3 (K-Means+OMP1). 1161 volumes of interest (VOIs) were used to optimize the parameters and train each method, and 1049 VOIs were adopted to evaluate the performances of the methods. The SR based methods were powerful to recognize the DLD patterns (SR1: 96.1%, SR2: 95.6%, SR3: 96.4%) and significantly better than the baseline methods. Furthermore, when the K-Means and OMP1 were applied, the runtime of the SR based methods can be reduced by 98.2% and 55.2%, respectively. Therefore, we thought that the method using the K-Means and OMP1 (SR3) was efficient for the CAD of the DLDs.

Medical image processing and computer-aided detection/diagnosis (CAD)

Computer-aided detection/diagnosis (CAD) is emerging as an innovative interdisciplinary technology for medical service. The traditional concept of automated computer diagnosis is encountered with a significant barrier because computerized medical systems cannot fully replace human doctors with the comparable level of performance. By contrast, CAD is becoming widely adopted in clinical work because it offers complementary computing power to enhance doctors competence for medical examination. 4 state-of-the-art CAD technologies were presented in the special session of medical image processing and CAD at ICCH 2012 as reported in this short paper. Those technologies will be briefly introduced here to show the current trend of development of CAD and to demonstrate how CAD helps in medical care.

Development of image viewer for analyzing of temporal subtraction from chest CT images

Recently, CT (Computed Tomography) scanner is used for detecting the abnormalities because CT scanner gradually becomes high resolution and high speed. However, with the improvement of the resolution of CT images, the number of CT images becomes huge. Therefore, radiologists have to analyze huge number of images and they sometimes misdiagnoses are happened. Hence, to deal with this problem the CAD (Computer Aided Diagnosis) system is developed. One of the CAD systems called temporal subtraction technique is useful to detect abnormalities in medical field. There is no viewer system which displays abnormal region using temporal subtraction technique. In this paper, we develop a novel temporal subtraction technique to help the radiologists to reduce diagnostic ti me and improve the diagnostic accuracy.

Automatic identification of lung candidate nodules on chest CT images based on temporal subtraction images

Lung cancer is the most common cause of death from cancer worldwide. Therefore, for the purpose of early detection of cancer, mass screening and thorough examination have been carried out. Lung cancer is detected easily by using chest MDCT (Multi Detector-row Computed Tomography) images. However, radiologists are apprehended burden by many chest MDCT images which are required interpretation of radiograms. So the CAD (Computer Aided Diagnosis) systems that could relieve radiologists stress and diagnose accuracy could be improved are expected. One of the CAD systems, temporal subtraction technique that emphasized time-dependent change is reported. This technique is used for diagnosis assistance of detected candidate nodules from CT images. In this paper, the candidate nodules under 20[mm] are extracted from temporal subtraction images. We highlighted the candidate nodules based on features analysis of images. We applied proposed method to 31 cases of chest MDCT images in which the number of nodules was more than one. We got a result of TPR:96.9[%], FPR:6.45[/case].