Probabilistic model-based time-series forecasting neural networks and related applications to biosignal forecasting

• Project/Area Number: 17K12752
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Soft computing
• Research Institution: Kyushu University
• Principal Investigator: Hayashi Hideaki 九州大学, システム情報科学研究院, 助教 (00790015)
• Project Period (FY): 2017-04-01 – 2021-03-31
• Project Status: Completed (Fiscal Year 2020)
• Budget Amount: ¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000); Fiscal Year 2020: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000); Fiscal Year 2019: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000); Fiscal Year 2018: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000); Fiscal Year 2017: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
• Keywords: ニューラルネットワーク / 深層学習 / 時系列予測 / 確率モデル / 生体信号 / 生体信号解析 / パターン認識 / 機械学習 / 時系列解析 / 時系列 / 時系列信号 / 予測

Outline of Final Research Achievements

We proposed mathematical models that can represent biological signals, which are electrical signals that can be measured from the human body, and applied them to the analysis of actual data. For example, we modeled myoelectric signals, which are electrical signals of muscles, and applied them to the analysis of muscle strength. We also proposed models for electrocardiogram and electroencephalogram and applied them to signal classification. Furthermore, we developed neural networks based on probabilistic models, and applied them to data classification and time series forecasting. In addition, we constructed large medical datasets such as biological signals of a pregnant woman called cardiotocography and endoscopic images, and applied them to fetal condition prediction and organ classification.

Academic Significance and Societal Importance of the Research Achievements

学術的意義として最大の点は，本研究においてニューラルネットワーク（NN）へのドメイン知識埋め込み法を提案している点である．提案法では，データの特性を確率モデルに基づき表現し，それをNNへ埋め込むことで解釈性や汎化性を向上させる．
社会的意義としては，生体信号の予測が実現できれば医療モニタリング応用に役立つ．在宅医療を受ける患者は約18万人いるとされる（2017年厚生労働省調べ）．そのような患者に対し，自宅でも計測が容易な血圧や指尖容積脈波などと提案法を組み合わせることにより容態変化を予測することができれば，異常が起こる前に医師に連絡することができスムーズな治療が期待できる．

Research Products

• [Int'l Joint Research] University of Cambridge(英国)
• [Int'l Joint Research] University of Cambridge(英国)
• [Journal Article] Markerless Measurement and Evaluation of General Movements in Infants Scientific Reports (2020)
  • Author(s): Toshio Tsuji, Shota Nakashima, Hideaki Hayashi, Zu Soh, Akira Furui, Taro Shibanoki, Keisuke Shima, Koji Shimatani
  • Journal Title: Scientific Reports Volume: 10 Issue: 1 Pages: 1422-1422
  • DOI: 10.1038/s41598-020-57580-z
  • Peer Reviewed / Open Access
• [Journal Article] Combining Noise-to-Image and Image-to-Image GANs: Brain MR Image Augmentation for Tumor Detection (2019)
  • Author(s): Changhee Han, Leonardo Rundo, Ryosuke Araki, Yudai Nagano, Yujiro Furukawa, Giancarlo Mauri, Hideki Nakayama, Hideaki Hayashi
  • Journal Title: IEEE Access Volume: 7 Pages: 156966-156977
  • DOI: 10.1109/access.2019.2947606
  • Peer Reviewed / Open Access / Int'l Joint Research
• [Journal Article] Biosignal Generation and Latent Variable Analysis with Recurrent Generative Adversarial Networks (2019)
  • Author(s): Shota Harada, Hideaki Hayashi, Seiichi Uchida
  • Journal Title: IEEE Access Volume: 7 Pages: 144292-144302
  • DOI: 10.1109/access.2019.2934928
  • Peer Reviewed / Open Access
• [Journal Article] GlyphGAN: Style-Consistent Font Generation Based on Generative Adversarial Networks (2019)
  • Author(s): Hideaki Hayashi, Kohtaro Abe, Seiichi Uchida
  • Journal Title: Knowledge-Based Systems Volume: 186 Pages: 104927-104927
  • DOI: 10.1016/j.knosys.2019.104927
  • Peer Reviewed
• [Journal Article] A Scale Mixture-based Stochastic Model of Surface EMG Signals with Variable Variances (2019)
  • Author(s): Akira Furui, Hideaki Hayashi, and Toshio Tsuji
  • Journal Title: IEEE Transactions on Biomedical Engineering Volume: 印刷中 Issue: 10 Pages: 2780-2788
  • DOI: 10.1109/tbme.2019.2895683
  • Peer Reviewed
• [Presentation] Layer-Wise Interpretation of Deep Neural Networks Using Identity Initialization (2021)
  • Author(s): Shohei Kubota, Hideaki Hayashi, Tomohiro Hayase, Seiichi Uchida
  • Organizer: International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2021)
  • Int'l Joint Research
• [Presentation] A Discriminative Gaussian Mixture Model with Sparsity (2021)
  • Author(s): Hideaki Hayashi and Seiichi Uchida
  • Organizer: International Conference on Learning Representations (ICLR 2021)
  • Int'l Joint Research
• [Presentation] Regularized Pooling (2020)
  • Author(s): Takato Otsuzuki, Hideaki Hayashi, Yuchen Zheng, Seiichi Uchida
  • Organizer: International Conference on Artificial Neural Networks (ICANN 2020)
  • Int'l Joint Research
• [Presentation] Handwriting Prediction Considering Inter-Class Bifurcation Structures (2020)
  • Author(s): Masaki Yamagata, Hideaki Hayashi, and Seiichi Uchida
  • Organizer: International Conference on Frontiers of Handwriting Recognition (ICFHR 2020)
  • Int'l Joint Research
• [Presentation] 深層パーセプトロンの単位初期化に基づく中間層の貢献度と尤度の解析 (2020)
  • Author(s): 久保田祥平，早志英朗，早瀬友裕，内田誠一
  • Organizer: 電子情報通信学会技術研究報告
• [Presentation] 正則化プーリング (2020)
  • Author(s): 緒續隆人，早志英朗，Zheng Yuchen，内田誠一
  • Organizer: 電子情報通信学会技術研究報告
• [Presentation] クラスの存在を利用した時系列予測とその手書きパターンへの応用 (2020)
  • Author(s): 山縣将貴，早志英朗，内田誠一
  • Organizer: 電子情報通信学会技術研究報告
• [Presentation] 識別・生成のハイブリッドモデルと弱教師あり学習への応用 (2020)
  • Author(s): 早志英朗，内田誠一
  • Organizer: 画像の認識・理解シンポジウム (MIRU)
• [Presentation] 単位初期化による深層パーセプトロン学習：ヤコビ行列を用いた誤差逆伝播に関する考察 (2020)
  • Author(s): 久保田祥平，早志英朗，早瀬友裕，内田誠一
  • Organizer: 画像の認識・理解シンポジウム (MIRU)
• [Presentation] Class-Guided Handwriting Prediction with Uncertainty (2020)
  • Author(s): Masaki Yamagata, Hideaki Hayashi, Seiichi Uchida
  • Organizer: 画像の認識・理解シンポジウム (MIRU)
• [Presentation] 識別と生成のハイブリッドニューラルネットワーク (2020)
  • Author(s): 早志英朗，内田誠一
  • Organizer: パターン認識・メディア理解研究会
• [Presentation] Efficient Soft-Constrained Clustering for Group-Based Labeling (2019)
  • Author(s): Ryoma Bise, Kentaro Abe, Hideaki Hayashi, Kiyohito Tanaka, and Seiichi Uchida
  • Organizer: The 22nd International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2019)
  • Int'l Joint Research
• [Presentation] Modality Conversion of Handwritten Patterns by Cross Variational Autoencoders (2019)
  • Author(s): Taichi Sumi, Brian Kenji Iwana, Hideaki Hayashi, and Seiichi Uchida
  • Organizer: The 15th International Conference on Document Analysis and Recognition (ICDAR 2019)
  • Int'l Joint Research
• [Presentation] Page Segmentation using a Convolutional Neural Network with Trainable Co-occurrence Features (2019)
  • Author(s): Joonho Lee, Hideaki Hayashi, Wataru Ohyama, and Seiichi Uchida
  • Organizer: The 15th International Conference on Document Analysis and Recognition (ICDAR 2019)
  • Int'l Joint Research
• [Presentation] Endoscopic Image Clustering with Temporal Ordering Information Based on Dynamic Programming (2019)
  • Author(s): Shota Harada, Hideaki Hayashi, Ryoma Bise, Kiyohito Tanaka, Qier Meng, and Seiichi Uchida
  • Organizer: The 41st International Engineering in Medicine and Biology Conference (EMBC2019)
  • Int'l Joint Research
• [Presentation] Cardiotocogramの識別に基づく胎児の状態推定 (2019)
  • Author(s): 原田翔太, 早志英朗, 古賀俊介, 重見大介, 柴田綾子, 吉田昌義, 蓮尾泰之, 内田誠一
  • Organizer: 医用画像研究会
• [Presentation] 混合正規分布に基づくニューラルネットワークのスパースベイズ学習 (2018)
  • Author(s): 早志英朗，内田誠一
  • Organizer: パターン認識・メディア理解研究会
• [Presentation] A Trainable Multiplication Layer for Auto-correlation and Co-occurrence Extraction (2018)
  • Author(s): Hideaki Hayashi and Seiichi Uchida
  • Organizer: 14th Asian Conference on Computer Vision
  • Int'l Joint Research
• [Presentation] A Multiplication Layer for Sequence Data (2018)
  • Author(s): Joonho Lee, Hideaki Hayashi, Seiichi Uchida
  • Organizer: 電気・情報関係学会九州支部連合大会
• [Presentation] A Trainable Multiplication Layer (2018)
  • Author(s): Hideaki Hayashi and Seiichi Uchida
  • Organizer: 画像の認識・理解シンポジウム
• [Presentation] A Trainable Multiplication Layer and its Applications (2018)
  • Author(s): Hideaki Hayashi and Seiichi Uchida
  • Organizer: 14th Joint Workshop on Machine Perception and Robotics
  • Int'l Joint Research
• [Presentation] Biosignal Data Augmentation Based on Generative Adversarial Networks (2018)
  • Author(s): Shota Harada, Hideaki Hayashi, Seiichi Uchida
  • Organizer: The 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC'18)
  • Int'l Joint Research
• [Presentation] A Probabilistic Model-based Neural Network (2017)
  • Author(s): Hideaki Hayashi
  • Organizer: The 13th Joint Workshop on Machine Perception and Robotics (MPR 2017)
  • Int'l Joint Research
• [Presentation] Generative Adversarial Networksに基づく生体信号生成 (2017)
  • Author(s): 原田翔太，早志英朗，内田誠一
  • Organizer: 電気・情報関係学会九州支部連合大会
• [Presentation] Johnson分布に基づくニューラルネットワーク (2017)
  • Author(s): 早志英朗，内田誠一，辻敏夫
  • Organizer: パターン認識・メディア理解研究会
• [Presentation] A Time-series Discriminant Component Network (2017)
  • Author(s): Hideaki Hayashi and Toshio Tsuji
  • Organizer: 第20回画像の認識・理解シンポジウム (MIRU2017)
• [Book] Infinite Brain MR Images: PGGAN-based Data Augmentation for Tumor Detection. In Neural Approaches to Dynamics of Signal Exchanges, (Eds. by A. Esposito, M. Faundez-Zanuy, F.C. Morabito, and E. Pasero) (2020)
  • Author(s): Changhee Han, Leonardo Rundo, Ryosuke Araki, Yujiro Furukawa, Giancarlo Mauri, Hideki Nakayama, and Hideaki Hayashi
  • Total Pages: 521
  • Publisher: Springer
  • ISBN: 9789811389504