Michio Inoue
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Michio Inoue has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Ichizo Nishino
Department of Neuromuscular Research National Institute of Neuroscience, National Center of Neurology and Psychiatry (NCNP) Tokyo Japan
2 shared publications - Jantima Tanboon
Department of Pathology, Faculty of Medicine, Siriraj Hospital, Mahidol University, Bangkok, Thailand
2 shared publications - Yoshihiko Saito
Department of Neuromuscular Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Tokyo, Japan
2 shared publications - Akinori Uruha
Department of Neurology, Tokyo Metropolitan Neurological Hospital, Japan
1 shared publication - Atsushi Takano
IBM Japan Ltd.
1 shared publication - Fumihiko Matsuda
Center for Genomic Medicine, Kyoto University Graduate School of Medicine , Kyoto
1 shared publication
Latest Publications
- Open Access
- 15 June 2023
- Published by The Journal of Rheumatology in The Journal of Rheumatology
- Vol. 50 (9) , 1159-1164
Association Between HLA Alleles and Autoantibodies in Dermatomyositis Defined by Sarcoplasmic Expression of Myxovirus Resistance Protein A
Objective: The diagnosis in the studies analyzing HLA of dermatomyositis (DM) was based on a combined clinical category of polymyositis/DM. This retrospective study investigated the associations of HLA with 5 DM-specific autoantibodies in Japanese patients diagnosed by muscle pathology.Methods: We diagnosed Japanese patients with DM based on sarcoplasmic expression of myxovirus resistance protein A. These patients underwent investigation for 5 DM-specific autoantibodies and HLA genotyping.Results: Of 175 patients (83 males and 92 females; range 1-86 yrs; mean 46 yrs), 173 (98.9%) had 1 of the 5 autoantibodies. Seven alleles—A*02:07, B*46:01, DRB1*04:07, DRB1*07:01, DRB1*08:03, DQB1*06:01, andDPB1*02:02—were more frequently detected in the patients with DM than healthy controls, but these associations were not significant after multiple testing correction. Stratifying by DM-specific autoantibodies, we found the associations of 6 already known and 7 new alleles—B*48:01, B*52:01, C*12:02, DRB1*04:05, DRB1*15:02, DPB1*05:01, andDPB1*09:01—with subsets of DM. Moreover, significant associations of 5 alleles with antinucleosome remodeling deacetylase complex (Mi-2) remained after multiple testing correction. In particular, theDRB1*04:07(odds ratio [OR 28.9]; correctedP= 2.7 × 10−6) andDQB1*06:01(OR 4.0; correctedP= 1.6 × 10−4) alleles were significantly more prevalent in patients with anti-Mi-2 antibody than in controls.Conclusion: This study demonstrates DM-specific autoantibodies defined immunogenetic subsets of DM.
- 15 June 2023
- Published by The Journal of Rheumatology in The Journal of Rheumatology
- Vol. 50 (9) , 1159-1164
Open AccessAssociation Between HLA Alleles and Autoantibodies in Dermatomyositis Defined by Sarcoplasmic Expression of Myxovirus Resistance Protein A
Objective: The diagnosis in the studies analyzing HLA of dermatomyositis (DM) was based on a combined clinical category of polymyositis/DM. This retrospective study investigated the associations of HLA with 5 DM-specific autoantibodies in Japanese patients diagnosed by muscle pathology.Methods: We diagnosed Japanese patients with DM based on sarcoplasmic expression of myxovirus resistance protein A. These patients underwent investigation for 5 DM-specific autoantibodies and HLA genotyping.Results: Of 175 patients (83 males and 92 females; range 1-86 yrs; mean 46 yrs), 173 (98.9%) had 1 of the 5 autoantibodies. Seven alleles—A*02:07, B*46:01, DRB1*04:07, DRB1*07:01, DRB1*08:03, DQB1*06:01, andDPB1*02:02—were more frequently detected in the patients with DM than healthy controls, but these associations were not significant after multiple testing correction. Stratifying by DM-specific autoantibodies, we found the associations of 6 already known and 7 new alleles—B*48:01, B*52:01, C*12:02, DRB1*04:05, DRB1*15:02, DPB1*05:01, andDPB1*09:01—with subsets of DM. Moreover, significant associations of 5 alleles with antinucleosome remodeling deacetylase complex (Mi-2) remained after multiple testing correction. In particular, theDRB1*04:07(odds ratio [OR 28.9]; correctedP= 2.7 × 10−6) andDQB1*06:01(OR 4.0; correctedP= 1.6 × 10−4) alleles were significantly more prevalent in patients with anti-Mi-2 antibody than in controls.Conclusion: This study demonstrates DM-specific autoantibodies defined immunogenetic subsets of DM.
- Open Access
A deep convolutional neural network-based algorithm for muscle biopsy diagnosis outperforms human specialists
Histopathologic evaluation is essential for categorizing and studying neuromuscular disorders. However, experienced specialists and pathologists are limited, especially in underserved areas. Although new technologies, such as artificial intelligence, are expected to improve medical reach, their use in rare diseases is challenging because of the limited availability of training datasets. To address this knowledge gap, we developed an algorithm based on deep convolutional neural networks that used data from microscopic images of hematoxylin-and-eosin-stained pathology slides. Our algorithm differentiated idiopathic inflammatory myopathies (mostly treatable) from hereditary muscle diseases (mostly non-treatable) and achieved better sensitivity and specificity than real physicians' diagnoses. Furthermore, it successfully and accurately classified four subtypes of the abovementioned muscular conditions. These results suggest that our algorithm can be safely used in a clinical setting. We established the similarity between the algorithm's and physicians' predictions using visualization technology, and clarified the validity of the predictions.
Open AccessA deep convolutional neural network-based algorithm for muscle biopsy diagnosis outperforms human specialists
Histopathologic evaluation is essential for categorizing and studying neuromuscular disorders. However, experienced specialists and pathologists are limited, especially in underserved areas. Although new technologies, such as artificial intelligence, are expected to improve medical reach, their use in rare diseases is challenging because of the limited availability of training datasets. To address this knowledge gap, we developed an algorithm based on deep convolutional neural networks that used data from microscopic images of hematoxylin-and-eosin-stained pathology slides. Our algorithm differentiated idiopathic inflammatory myopathies (mostly treatable) from hereditary muscle diseases (mostly non-treatable) and achieved better sensitivity and specificity than real physicians' diagnoses. Furthermore, it successfully and accurately classified four subtypes of the abovementioned muscular conditions. These results suggest that our algorithm can be safely used in a clinical setting. We established the similarity between the algorithm's and physicians' predictions using visualization technology, and clarified the validity of the predictions.