Unique ID issued by UMIN | UMIN000043855 |
---|---|
Receipt number | R000050035 |
Scientific Title | Accuracy of diagnostic-support artificial intelligence interpretation to detect interstitial pneumonia in the medical examination |
Date of disclosure of the study information | 2021/04/09 |
Last modified on | 2024/08/28 17:56:01 |
Accuracy of diagnostic-support artificial intelligence interpretation to detect interstitial pneumonia in the medical examination
Accuracy of diagnostic-support artificial intelligence interpretation to detect interstitial pneumonia in the medical examination
Accuracy of diagnostic-support artificial intelligence interpretation to detect interstitial pneumonia in the medical examination
Accuracy of diagnostic-support artificial intelligence interpretation to detect interstitial pneumonia in the medical examination
Japan |
Interstitial lung disease
Medicine in general | Pneumology |
Others
NO
We previously developed the artificial intelligence engine to detect interstitial lung diseases on the chest X-ray image. Although we validated the detection ability of the AI engine with dataset of 200 chest X-ray images in the previous study, the number of images was not enough. Moreover, more than half of the images were derived from patients with interstitial lung diseases and this prevalence was completely different from the real world. By having the AI engine interpret the chest X-rays of the examinees who visit the medical examination center, we can evaluate the accuracy of the AI engine to detect interstitial lung diseases and estimate prevalence of interstitial lung diseases in the real world.
Efficacy
The sensitivity, specificity, positive predictive value and negative predictive value of the AI engine to detect interstitial lung diseases in the medical health check up examination.
The estimated prevalence of interstitial lung diseases in this cohort.
The comparison of the detection ability between the AI engine and human doctors.
Observational
50 | years-old | <= |
100 | years-old | >= |
Male and Female
Subjects aged >=50 years who visit Sapporo Fukujyuji Medical Health Center or Hokkaido Cancer Society Medical Health Center or Shin-yurigaoka General Hospital for the medical health check up from the approval day by the president to Dec 31, 2022.
Subjects who refuse written consent and do not undergo chest X-ray and/or blood examination. To maintain anonymization, elderly subjects aged >100 years, and those with very rares disease (i.e., <= 10 subjects in the database) will also be excluded.
3770
1st name | Hirofumi |
Middle name | |
Last name | Chiba |
Sapporo Medical University, School of Medicine
Department of Respiratory Medicine and Allergology
060-8543
S1-W16, Chuo-ku, Sapporo, Hokkaido
011-611-2111
hchiba@sapmed.ac.jp
1st name | Hirotaka |
Middle name | |
Last name | Nishikiori |
Sapporo Medical University, School of Medicine
Department of Respiratory Medicine and Allergology
060-8543
S1-W16, Chuo-ku, Sapporo, Hokkaido
011-611-2111
hnishiki@sapmed.ac.jp
Sapporo Medical University
Nippon Boehringer Ingelheim Co ., Ltd.
Profit organization
Sapporo Medical University
S1-W16, Chuo-ku, Sapporo, Hokkaido
011-611-2111
rinri@sapmed.ac.jp
NO
北海道結核予防会 札幌複十字総合健診センター(北海道)
北海道対がん協会札幌がん検診センター(北海道)
医療法人社団三成会新百合ヶ丘総合病院(神奈川県)
2021 | Year | 04 | Month | 09 | Day |
Unpublished
2805
No longer recruiting
2020 | Year | 10 | Month | 28 | Day |
2020 | Year | 10 | Month | 28 | Day |
2021 | Year | 06 | Month | 29 | Day |
2024 | Year | 01 | Month | 31 | Day |
2024 | Year | 01 | Month | 31 | Day |
2024 | Year | 04 | Month | 30 | Day |
2024 | Year | 06 | Month | 30 | Day |
We gather the data of the chest X-ray images and serum SP-D/KL-6 levels, gender, age, pre-existing illness, smoking history, etc. of examinees in the medical health examination. If examinees are indicated abnormalities on the chest X-ray images and/or high SP-D/KL-6 levels, they are recommended to go to the referral hospital for more detailed examination. We interpret the chest CT images taken at these hospitals. With the result of the CT interpretation as the correct answer, we make AI engine interpret the chest X-ray images at the first medical examination and evaluate the detectability of interstitial shadow of the AI engine. We estimate the actual prevalence of ILD from the number of images that AI engine judged to have interstitial lung shadows.
2021 | Year | 04 | Month | 06 | Day |
2024 | Year | 08 | Month | 28 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000050035