Osamu Honda
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Osamu Honda has a total of 76 co-authors. Here are the top co-authors by number of shared publications:
- Noriyuki Tomiyama
Department of Radiology, Osaka University Graduate School of Medicine, Suita, Japan
10 shared publications - Hiromitsu Sumikawa
Department of Radiology, National Hospital Organization Kinki-Chuo Chest Medical Center, Osaka, Japan
7 shared publications Mitsuhiro Koyama
6 shared publicationsSeiki Hamada
6 shared publicationsMitsuko Tsubamoto
5 shared publicationsNoriyuki Tomiyama
Department of Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita-City, Osaka, 565-0871, Japan
5 shared publications
Latest Publications
- 4 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 24 (7) , 2041-2052
Pulmonary Textures Classification via a Multi-Scale Attention Network
Precise classification of pulmonary textures is crucial to develop a computer aided diagnosis (CAD) system of diffuse lung diseases (DLDs). Although deep learning techniques have been applied to this task, the classification performance is not satisfied for clinical requirements due to that commonly-used deep networks built by stacking convolutional blocks are not able to learn discriminative feature representation to distinguish complex pulmonary textures. For addressing this problem, we design a multi-scale attention network (MSAN) architecture comprised by several stacked residual attention modules followed by a multi-scale fusion module. Our deep network can not only exploit powerful information on different scales but also automatically select optimal features for more discriminative feature representation. Besides, we develop visualization techniques to make the proposed deep model transparent for humans. The proposed method is evaluated by using a large dataset. Experimental results show that our method has achieved the average classification accuracy of 94.78 and the average f-value of 0.9475 in the classification of 7 categories of pulmonary textures. Besides, visualization results intuitively explain the working behavior of the deep network. The proposed method has achieved the state-of-the-art performance to classify pulmonary textures on high resolution CT images.
- 4 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 24 (7) , 2041-2052
Pulmonary Textures Classification via a Multi-Scale Attention Network
Precise classification of pulmonary textures is crucial to develop a computer aided diagnosis (CAD) system of diffuse lung diseases (DLDs). Although deep learning techniques have been applied to this task, the classification performance is not satisfied for clinical requirements due to that commonly-used deep networks built by stacking convolutional blocks are not able to learn discriminative feature representation to distinguish complex pulmonary textures. For addressing this problem, we design a multi-scale attention network (MSAN) architecture comprised by several stacked residual attention modules followed by a multi-scale fusion module. Our deep network can not only exploit powerful information on different scales but also automatically select optimal features for more discriminative feature representation. Besides, we develop visualization techniques to make the proposed deep model transparent for humans. The proposed method is evaluated by using a large dataset. Experimental results show that our method has achieved the average classification accuracy of 94.78 and the average f-value of 0.9475 in the classification of 7 categories of pulmonary textures. Besides, visualization results intuitively explain the working behavior of the deep network. The proposed method has achieved the state-of-the-art performance to classify pulmonary textures on high resolution CT images.
- 3 April 2018
- Published by Elsevier in European Journal of Radiology
- Vol. 103, 71-75
Influence of gantry rotation time and scan mode on image quality in ultra-high-resolution CT system
- 3 April 2018
- Published by Elsevier in European Journal of Radiology
- Vol. 103, 71-75
Influence of gantry rotation time and scan mode on image quality in ultra-high-resolution CT system
- 1 June 2014
- Published by Elsevier in Academic Radiology
- Vol. 21 (6) , 695-703
Ultra-low-dose CT of the Lung
- 1 June 2014
- Published by Elsevier in Academic Radiology
- Vol. 21 (6) , 695-703
Ultra-low-dose CT of the Lung
- 1 January 2014
- Published by Elsevier in Computerized Medical Imaging and Graphics
- Vol. 38 (1) , 15-21
320-Row wide volume CT significantly reduces density heterogeneity observed in the descending aorta: Comparisons with 64-row helical CT
- 1 January 2014
- Published by Elsevier in Computerized Medical Imaging and Graphics
- Vol. 38 (1) , 15-21
320-Row wide volume CT significantly reduces density heterogeneity observed in the descending aorta: Comparisons with 64-row helical CT
- 1 January 2013
- Published by Wolters Kluwer Health in Journal of Computer Assisted Tomography
- Vol. 37 (5) , 707-711
The Effect of the Virtual Monochromatic Spectral Imaging for the Metallic Artifact and the Pulmonary Nodule Detection
This study aimed to evaluate whether dual-energy computed tomography can reduce metal artifacts and improve detection of pulmonary nodules. Twelve simulated nodules were randomly placed inside a chest phantom with a pacemaker. Then, dual-energy computed tomography was performed, and 5 virtual monochromatic images at 40, 50, 65, 100, and 140 keV were reconstructed with 5- and 0.625-mm slice thicknesses. Two independent observers assessed the metal artifact (3-point scale from 1, none, to 3, severe) and detection of the nodule (5-point scale from 1, definitely absent, to 5, definitely present). Statistical analysis was performed with a P value of less than 0.01 (0.05/5). With both slice thicknesses, the metallic artifact increased at 40 or 50 keV and decreased at 100 or 140 keV relative to that at 65 keV (P < 0.01). The nodule detection score was not significantly different between each kiloelectron volt level with the 0.625-mm slice thickness; however, the score was significantly worse at 40 keV compared to 65 keV (P < 0.01) with the 5-mm slice thickness. High monochromatic energy images can reduce metal artifacts without a change in nodule detection score. Low monochromatic energy images increase metal artifacts and worsen nodule detection in thick slices.
- 1 January 2013
- Published by Wolters Kluwer Health in Journal of Computer Assisted Tomography
- Vol. 37 (5) , 707-711
The Effect of the Virtual Monochromatic Spectral Imaging for the Metallic Artifact and the Pulmonary Nodule Detection
This study aimed to evaluate whether dual-energy computed tomography can reduce metal artifacts and improve detection of pulmonary nodules. Twelve simulated nodules were randomly placed inside a chest phantom with a pacemaker. Then, dual-energy computed tomography was performed, and 5 virtual monochromatic images at 40, 50, 65, 100, and 140 keV were reconstructed with 5- and 0.625-mm slice thicknesses. Two independent observers assessed the metal artifact (3-point scale from 1, none, to 3, severe) and detection of the nodule (5-point scale from 1, definitely absent, to 5, definitely present). Statistical analysis was performed with a P value of less than 0.01 (0.05/5). With both slice thicknesses, the metallic artifact increased at 40 or 50 keV and decreased at 100 or 140 keV relative to that at 65 keV (P < 0.01). The nodule detection score was not significantly different between each kiloelectron volt level with the 0.625-mm slice thickness; however, the score was significantly worse at 40 keV compared to 65 keV (P < 0.01) with the 5-mm slice thickness. High monochromatic energy images can reduce metal artifacts without a change in nodule detection score. Low monochromatic energy images increase metal artifacts and worsen nodule detection in thick slices.
- 1 October 2012
- Published by Elsevier in European Journal of Radiology
- Vol. 81 (10) , 2877-2886
Pulmonary nodules: Effect of adaptive statistical iterative reconstruction (ASIR) technique on performance of a computer-aided detection (CAD) system—Comparison of performance between different-dose CT scans
- 1 October 2012
- Published by Elsevier in European Journal of Radiology
- Vol. 81 (10) , 2877-2886
Pulmonary nodules: Effect of adaptive statistical iterative reconstruction (ASIR) technique on performance of a computer-aided detection (CAD) system—Comparison of performance between different-dose CT scans
- 1 April 2011
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 84 (1000) , 335-341
Image quality of multiplanar reconstruction of pulmonary CT scans using adaptive statistical iterative reconstruction
Objective: We investigated the image quality of multiplanar reconstruction (MPR) using adaptive statistical iterative reconstruction (ASIR). Methods: Inflated and fixed lungs were scanned with a garnet detector CT in high-resolution mode (HR mode) or non-high-resolution (HR) mode, and MPR images were then reconstructed. Observers compared 15 MPR images of ASIR (40%) and ASIR (80%) with those of ASIR (0%), and assessed image quality using a visual five-point scale (1, definitely inferior; 5, definitely superior), with particular emphasis on normal pulmonary structures, artefacts, noise and overall image quality. Results: The mean overall image quality scores in HR mode were 3.67 with ASIR (40%) and 4.97 with ASIR (80%). Those in non-HR mode were 3.27 with ASIR (40%) and 3.90 with ASIR (80%). The mean artefact scores in HR mode were 3.13 with ASIR (40%) and 3.63 with ASIR (80%), but those in non-HR mode were 2.87 with ASIR (40%) and 2.53 with ASIR (80%). The mean scores of the other parameters were greater than 3, whereas those in HR mode were higher than those in non-HR mode. There were significant differences between ASIR (40%) and ASIR (80%) in overall image quality (p<0.01). Contrast medium in the injection syringe was scanned to analyse image quality; ASIR did not suppress the severe artefacts of contrast medium. Conclusion: In general, MPR image quality with ASIR (80%) was superior to that with ASIR (40%). However, there was an increased incidence of artefacts by ASIR when CT images were obtained in non-HR mode.
- 1 April 2011
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 84 (1000) , 335-341
Image quality of multiplanar reconstruction of pulmonary CT scans using adaptive statistical iterative reconstruction
Objective: We investigated the image quality of multiplanar reconstruction (MPR) using adaptive statistical iterative reconstruction (ASIR). Methods: Inflated and fixed lungs were scanned with a garnet detector CT in high-resolution mode (HR mode) or non-high-resolution (HR) mode, and MPR images were then reconstructed. Observers compared 15 MPR images of ASIR (40%) and ASIR (80%) with those of ASIR (0%), and assessed image quality using a visual five-point scale (1, definitely inferior; 5, definitely superior), with particular emphasis on normal pulmonary structures, artefacts, noise and overall image quality. Results: The mean overall image quality scores in HR mode were 3.67 with ASIR (40%) and 4.97 with ASIR (80%). Those in non-HR mode were 3.27 with ASIR (40%) and 3.90 with ASIR (80%). The mean artefact scores in HR mode were 3.13 with ASIR (40%) and 3.63 with ASIR (80%), but those in non-HR mode were 2.87 with ASIR (40%) and 2.53 with ASIR (80%). The mean scores of the other parameters were greater than 3, whereas those in HR mode were higher than those in non-HR mode. There were significant differences between ASIR (40%) and ASIR (80%) in overall image quality (p<0.01). Contrast medium in the injection syringe was scanned to analyse image quality; ASIR did not suppress the severe artefacts of contrast medium. Conclusion: In general, MPR image quality with ASIR (80%) was superior to that with ASIR (40%). However, there was an increased incidence of artefacts by ASIR when CT images were obtained in non-HR mode.
- 31 December 2010
- Published by Elsevier in Lung Cancer
- Vol. 70 (3) , 286-294
Automated assessment of malignant degree of small peripheral adenocarcinomas using volumetric CT data: Correlation with pathologic prognostic factors
- 31 December 2010
- Published by Elsevier in Lung Cancer
- Vol. 70 (3) , 286-294
Automated assessment of malignant degree of small peripheral adenocarcinomas using volumetric CT data: Correlation with pathologic prognostic factors
- 31 October 2010
- Published by Elsevier in Academic Radiology
- Vol. 17 (10) , 1259-1266
Adaptive Statistical Iterative Reconstruction Technique for Pulmonary CT: Image Quality of the Cadaveric Lung on Standard- and Reduced-Dose CT
- 31 October 2010
- Published by Elsevier in Academic Radiology
- Vol. 17 (10) , 1259-1266
Adaptive Statistical Iterative Reconstruction Technique for Pulmonary CT: Image Quality of the Cadaveric Lung on Standard- and Reduced-Dose CT
- 1 November 2009
- Published by Elsevier in Lung Cancer
- Vol. 66 (2) , 211-217
Doubling time of lung cancer determined using three-dimensional volumetric software: Comparison of squamous cell carcinoma and adenocarcinoma
- 1 November 2009
- Published by Elsevier in Lung Cancer
- Vol. 66 (2) , 211-217
Doubling time of lung cancer determined using three-dimensional volumetric software: Comparison of squamous cell carcinoma and adenocarcinoma