担当区分：筆頭著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Respirology  

Background and objective: Idiopathic pulmonary fibrosis (IPF) has poor prognosis, and the multidisciplinary diagnostic agreement is low. Moreover, surgical lung biopsies pose comorbidity risks. Therefore, using data from non-invasive tests usually employed to assess interstitial lung diseases (ILDs), we aimed to develop an automated algorithm combining deep learning and machine learning that would be capable of detecting and differentiating IPF from other ILDs. Methods: We retrospectively analysed consecutive patients presenting with ILD between April 2007 and July 2017. Deep learning was used for semantic image segmentation of HRCT based on the corresponding labelled images. A diagnostic algorithm was then trained using the semantic results and non-invasive findings. Diagnostic accuracy was assessed using five-fold cross-validation. Results: In total, 646,800 HRCT images and the corresponding labelled images were acquired from 1068 patients with ILD, of whom 42.7% had IPF. The average segmentation accuracy was 96.1%. The machine learning algorithm had an average diagnostic accuracy of 83.6%, with high sensitivity, specificity and kappa coefficient values (80.7%, 85.8% and 0.665, respectively). Using Cox hazard analysis, IPF diagnosed using this algorithm was a significant prognostic factor (hazard ratio, 2.593; 95% CI, 2.069–3.250; p < 0.001). Diagnostic accuracy was good even in patients with usual interstitial pneumonia patterns on HRCT and those with surgical lung biopsies. Conclusion: Using data from non-invasive examinations, the combined deep learning and machine learning algorithm accurately, easily and quickly diagnosed IPF in a population with various ILDs.

DOI： 10.1111/resp.14310

Web of Science

Scopus

PubMed

担当区分：筆頭著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：European Respiratory Journal  

DOI： 10.1183/13993003.01346-2020

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Respirology  

Background and objective: The development of clinically progressive fibrosis complicates a wide array of interstitial lung diseases (ILDs). However, there are limited data regarding its prevalence and prognosis. Methods: We analysed consecutive patients seen for initial evaluation of a fibrosing form of ILD (FILD). Patients were evaluated for evidence of progressive fibrosis over the first 24 months of follow-up. We defined a progressive phenotype as the presence of at least one of the following: a relative decline in forced vital capacity (FVC) of ≥10%; a relative decline in FVC of ≥5%–<10% with a relative decline in diffusing capacity of the lung for carbon monoxide of ≥15%, increased fibrosis on HRCT or progressive symptoms. Results: Eight hundred and forty-four patients (397 with idiopathic pulmonary fibrosis [IPF] and 447 non-IPF FILD) made up the final analysis cohort. Three hundred and fifty-five patients (42.1%) met the progressive phenotype criteria (59.4% of IPF patients and 26.6% of non-IPF FILD patients, p <0.01). In both IPF and non-IPF FILD, transplantation-free survival differed between patients with a progressive phenotype and those without (p <0.01). Multivariable analysis showed that a progressive phenotype was an independent predictor of transplantation-free survival (hazard ratio [HR]: 3.36, 95% CI: 2.68–4.23, p <0.01). Transplantation-free survival did not differ between non-IPF FILD with a progressive phenotype and IPF (HR: 1.12, 95% CI: 0.85–1.48, p = 0.42). Conclusion: Over one-fourth of non-IPF FILD patients develop a progressive phenotype compared to approximately 60% of IPF patients. The survival of non-IPF FILD patients with a progressive phenotype is similar to IPF.

DOI： 10.1111/resp.14245

Web of Science

Scopus

PubMed

記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Respiratory Investigation  

Background: The COPD Assessment Test (CAT) has been studied as a measure of health status in idiopathic pulmonary fibrosis (IPF) and interstitial lung disease associated with connective tissue disease. However, its prognostic value is unknown. The present study explored the association between CAT score and mortality in fibrotic interstitial lung disease (FILD), including IPF and other forms of ILD. Methods: We retrospectively analyzed 501 consecutive patients with FILD who underwent clinical assessment, including pulmonary function test and CAT. The association between CAT score and 3-year mortality was assessed using Cox proportional hazard analysis, Kaplan–Meier plots, and the log-rank test for trend. To handle missing data, the imputed method was used. Results: The patients’ median age was 68 years, and 320 were male (63.9%). Regarding CAT severity, 203 patients had a low impact level (score <10), 195 had a medium level (10–20), 80 had a high level (21–30), and 23 had a very high level (31–40). During the 3-year study period, 118 patients died. After adjusting for age, sex, forced vital capacity, diffusion capacity for carbon monoxide, IPF diagnosis, and usual interstitial pneumonia pattern on high-resolution computed tomography, the CAT score was significantly associated with 3-year mortality (hazard ratio in increments of 10 points: 1.458, 95% confidence interval 1.161–1.830; p < 0.001). In addition, patients with high and very high impact levels had twofold and threefold higher mortality risk than those with low levels, respectively. Conclusion: The CAT has prognostic value in FILD.

DOI： 10.1016/j.resinv.2021.07.007

Web of Science

Scopus

PubMed

記述言語：日本語   掲載種別：研究論文（研究会，シンポジウム資料等）   出版者・発行元：(一社)日本医療情報学会  

記述言語：日本語 著書種別：学術書

記述言語：日本語 著書種別：学術書

担当区分：筆頭著者,　責任著者   記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）   出版者・発行元：(株)日本医事新報社  

担当区分：筆頭著者,　最終著者,　責任著者   記述言語：日本語   掲載種別：記事・総説・解説・論説等（学術雑誌）   出版者・発行元：(有)科学評論社  

記述言語：日本語   掲載種別：記事・総説・解説・論説等（学術雑誌）   出版者・発行元：(有)科学評論社  

担当区分：筆頭著者,　最終著者,　責任著者   記述言語：日本語   掲載種別：記事・総説・解説・論説等（学術雑誌）  

CiNii Research

担当区分：筆頭著者,　最終著者,　責任著者   記述言語：日本語   掲載種別：記事・総説・解説・論説等（学術雑誌）   出版者・発行元：(株)総合医学社  

間質性肺炎(ILD)の正確な診断には、呼吸器内科医、放射線科医、病理医の集学的議論(MDD)、外科的肺生検やクライオバイオプシーが重要であるが、近年、新しいテクノロジーを用いた診療が広がりつつある。機械学習と深層学習を用いたILDの画像所見の抽出、ゲノム評価、診断と治療効果予測について概説した。また、間質性肺炎のAI(人工知能)創薬、遠隔診療(オンラインMDD診断)、在宅モニタリングと呼吸器リハビリテーションについて述べた。

開催年月日： 2022年12月

記述言語：日本語   会議種別：口頭発表（一般）  

開催年月日： 2022年11月

記述言語：英語   会議種別：口頭発表（一般）  

開催年月日： 2022年10月

記述言語：日本語   会議種別：ポスター発表  

開催年月日： 2022年10月

記述言語：日本語   会議種別：シンポジウム・ワークショップ パネル（指名）  

開催年月日： 2022年9月

記述言語：日本語   会議種別：シンポジウム・ワークショップ パネル（指名）  

出願人：国立大学法人東海国立大学機構

出願番号：特願2022-175512  出願日：2022年11月

出願番号：特願US20200372650  出願日：2020年5月

公開日：2020年11月

特許番号/登録番号：特許US11,361,443 B2  登録日：2022年6月 

出願人：国立大学法人東海国立大学機構名古屋大学

出願番号：特願2022-1388807  出願日：2022年8月

出願人：国立大学法人東海国立大学機構名古屋大学

出願番号：特願2022-1388808  出願日：2022年8月

出願番号：特願2022-043291  出願日：2022年3月