• 研究課題をさがす
  • 研究者をさがす
  • KAKENの使い方
  1. 課題ページに戻る

2021 年度 実施状況報告書

Toward New-Generation AI-Based CAD System: Development of Interpretable Deep Learning-Based CAD System for Breast Cancer Diagnosis Using Mammogram

研究課題

研究課題/領域番号 20K08012
研究機関仙台高等専門学校

研究代表者

張 暁勇  仙台高等専門学校, 総合工学科, 准教授 (90722752)

研究分担者 費 仙鳳  東北文化学園大学, 工学部, 准教授 (20620470)
研究期間 (年度) 2020-04-01 – 2023-03-31
キーワードMammograpy / Deep Learning / Explainable AI / Computer-Aided Diagnosis
研究実績の概要

The purpose of this research is to develop an interpretable deep learning (DL)-based computer-aided diagnosis (CAD) system for breast cancer diagnosis in mammogram. According to the research plan, the research achievements in the FY2020 are summarized as follows.

(1) Tow datasets, including mammograms, radiologist annotations, and biopsy-proven diagnosis results, have been collected at Tohoku university hospital and Miyagi cancer society.
(2) Preliminary experiments have been conducted to utilize several state-of-the-art end-to-end DL models, such as faster R-CNN, SSD, to detect/localize the suspicious lesions in mammograms.
(3) Four papers have been published in international conferences.

現在までの達成度 (区分)
現在までの達成度 (区分)

2: おおむね順調に進展している

理由

In the FY2021, the research has been conducted and progressed smoothly according to the research plan.
Biopsy diagnosis result and radiologist diagnosis annotation on the datasets have been collected with cooperation of radiologists in Tohoku university hospital. These data will be used to train, test, and evaluate the DL models in the following research work. In addition, based on our current results, two papers have been submitted to prime international journals, and four papers have been published in international conferences.

今後の研究の推進方策

According to the research plan, the main research in FY2022 will be focused on the following three tasks. (1) Diagnosis accuracy of DL models will be evaluated quantitatively using collected datasets. (2) Several state-of-the-art visualization techniques will be implemented for generating explainable diagnosis result, and assessed qualitatively in comparison of radiologist diagnosis. (3) A conclusive paper will be submitted to prime international journal.

次年度使用額が生じた理由

Due to the COVID-19 pandemic, the travel expense used for attending the domestic and international conferences was not used during the FY2021. In addition, a GPU shortage came about in the late 2020, purchasing a GPU used for deep learning was pending in the FY2021.
We plan to use these part of expense for journal publication fee and international conference registration fee in the next fiscal year

  • 研究成果

    (5件)

すべて 2021

すべて 雑誌論文 (1件) (うち査読あり 1件) 学会発表 (4件) (うち国際学会 3件)

  • [雑誌論文] Improved Tumor Image Estimation in X-ray Fluoroscopic Images by Augmenting 4DCT Data for Radiotherapy2021

    • 著者名/発表者名
      Takumi Shinohara, Kei, Ichiji, Xiaoyong Zhang, Norihiro Sugita, Noriyasu Homma,
    • 雑誌名

      Journal of Advanced Computational Intelligence and Intelligent Informatics

      巻: - ページ: -

    • 査読あり
  • [学会発表] An Interpretable DL-Based Method for Diagnosis of H. Pylori Infection Using Gastric X-ray Images2021

    • 著者名/発表者名
      Reima Ishii, Xiaoyong Zhang, Noriyasu Homma
    • 学会等名
      IEEE 3rd Global Conference on Life Sciences and Technologies
    • 国際学会
  • [学会発表] Deep Learning-Based Interpretable Computer Aided Diagnosis of Drowning for Forensic Radiology2021

    • 著者名/発表者名
      Yuwen Zeng, Xiaoyong Zhang, et al.
    • 学会等名
      60th Annual Conference of SICE
    • 国際学会
  • [学会発表] Deep CNN-Based Computer-Aided Diagnosis for Drowning Detection using Post-mortem Lungs CT Images2021

    • 著者名/発表者名
      Amber H. Qureshi, Xiaoyong Zhang, et al.
    • 学会等名
      2021 IEEE International Conference on Bioinformatics and Biomedicine
    • 国際学会
  • [学会発表] 4次元CTデータの内挿・外挿によるX線透視像中の腫瘍像推定モデルの性能向上の試み2021

    • 著者名/発表者名
      篠原匠,市地慶, 本間経康, 張曉勇, 杉田典大, 吉澤誠
    • 学会等名
      インテリジェント・システム・シンポジウム

URL: 

公開日: 2022-12-28  

サービス概要 検索マニュアル よくある質問 お知らせ 利用規程 科研費による研究の帰属

Powered by NII kakenhi