• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Development of artificial-intelligence skin disease classifier and digital biopsy by using deep learning

Research Project

Project/Area Number 21K08339
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 53050:Dermatology-related
Research InstitutionEhime University (2022-2023)
University of Tsukuba (2021)

Principal Investigator

Fujisawa Yasuhiro  愛媛大学, 医学系研究科, 教授 (70550193)

Project Period (FY) 2021-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2023: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2022: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2021: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
Keywords人工知能 / 人口知能 / 深層学習 / 皮膚腫瘍 / 分類 / ディープラーニング
Outline of Research at the Start

現在開発中のAIシステムの社会実装に向けた精度向上と,現在のシステムをベースに皮膚病全般の診断も出来るAI診断システムの土台を構築する事を目的とする.また,データベースには皮膚病理画像もあり,臨床写真とのペアとしてAIを学習させ,臨床写真から病変の病理診断および腫瘍の深達度も判定できるAIデジタルバイオプシーシステムの研究も行う.

Outline of Final Research Achievements

After examining the factors influencing diagnosis in both correct and incorrect examples using GradCam, it was initially thought that a decrease in accuracy would occur when not focusing on the tumor center in the heatmap. However, it was interesting to note that there was little difference in the distribution of heatmaps between correct and incorrect examples.
Subsequently, when the model was trained to focus on the center of the images, it was observed that the accuracy decreased compared to using the entire image for training. This finding aligns with the earlier results obtained using GradCam, suggesting that not only the tumor center but also the surrounding information is utilized in tumor diagnosis.

Academic Significance and Societal Importance of the Research Achievements

GradCamの解析によるとヒートマップにおいて腫瘍中心部に注目していない場合に正答率が下がると考えていたが,興味深いことに正答例でも誤答例でもヒートマップの分布にあまり違いが見られなかった.逆の味方をすると,ヒートマップで腫瘍部分に注目していなくとも正答してしまっている画像もかなり含まれていることを示している.また,そこで画像の中央に着目するように設定して学習をさせると逆に全体を用いた場合と比べて正答率が低下することが分かった.皮膚腫瘍の判定において中央部の腫瘍部分だけでなくその周囲の情報も判定に用いられていると言うことになる.今後の機械学習におけるアノテーションの範囲にも検討が必要となる.

Report

(4 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • 2021 Research-status Report
  • Research Products

    (7 results)

All 2023 2022 2021

All Journal Article (3 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 2 results) Presentation (4 results) (of which Invited: 2 results)

  • [Journal Article] 【皮膚科領域でのビッグデータの活用法】皮膚腫瘍AIの開発と承認に向けての課題2023

    • Author(s)
      藤澤 康弘
    • Journal Title

      Derma.

      Volume: 331 Pages: 14-22

    • Related Report
      2023 Annual Research Report
  • [Journal Article] Deep Neural Network for Early Image Diagnosis of Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis2022

    • Author(s)
      Fujimoto Atsushi、Iwai Yuki、Ishikawa Takashi、Shinkuma Satoru、Shido Kosuke、Yamasaki Kenshi、Fujisawa Yasuhiro、Fujimoto Manabu、Muramatsu Shogo、Abe Riichiro
    • Journal Title

      The Journal of Allergy and Clinical Immunology: In Practice

      Volume: 10 Issue: 1 Pages: 277-283

    • DOI

      10.1016/j.jaip.2021.09.014

    • Related Report
      2022 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Classification of large scale image database of various skin diseases using deep learning.2021

    • Author(s)
      M Tanaka, ASaito, K Shido, Y Fujisawa, K Yamasaki, M Fujimoto, K Murao, Y Ninomiya, S Satoh, A Shimizu
    • Journal Title

      Int J CARS

      Volume: 16 Issue: 11 Pages: 1875-1887

    • DOI

      10.1007/s11548-021-02440-y

    • Related Report
      2021 Research-status Report
    • Peer Reviewed
  • [Presentation] AIで皮膚病は診断できるのか2023

    • Author(s)
      藤澤康弘
    • Organizer
      日本美容皮膚科学会
    • Related Report
      2023 Annual Research Report
    • Invited
  • [Presentation] 皮膚科医が主体となって開発するAI医療機器の面白さと難しさ2023

    • Author(s)
      藤澤康弘
    • Organizer
      日本皮膚科学会総会
    • Related Report
      2023 Annual Research Report
    • Invited
  • [Presentation] 皮膚科領域におけるAI開発について2022

    • Author(s)
      藤澤康弘
    • Organizer
      愛媛県医師会
    • Related Report
      2022 Research-status Report
  • [Presentation] 人工知能(AI)と医療への応用について2022

    • Author(s)
      藤澤康弘
    • Organizer
      皮膚アレルギー学会
    • Related Report
      2022 Research-status Report

URL: 

Published: 2021-04-28   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi