Development of high resolution multispectral fundus imaging with wavefront compensation and coded imaging system

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K20401
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 62010:Life, health and medical informatics-related
• Research Institution: Wakayama University
• Principal Investigator: Saita Yusuke 和歌山大学, システム工学部, 助教 (30708756)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: シャックハルトマン波面センサー / 広ダイナミックレンジ / 高空間分解能 / 回折光学素子 / 深層学習 / ハイパースペクトルイメージング / 分光画像 / 圧縮センシング

Outline of Final Research Achievements

In fundus diagnosis, there is a problem that the image is distorted due to wavefront aberration caused by the cornea and crystalline lens in the eyeball, but it can be solved by introducing adaptive optics. However, high-precision imaging is difficult when aberrations that exceed the potential of the wavefront sensor. In addition, if fundus images can be obtained in multiple wavelength bands, useful information for various diagnoses can be obtained, but multiple measurements were required to obtain hyperspectral images.
In this research, we have achieved improvement of measurement accuracy, robustness to the influence of disturbance and measurement environment, and speeding up of analysis time by introducing a wavefront dividing lens and various image processing techniques to the wavefront sensor. In addition, by using a diffraction grating, it is possible to acquire hyperspectral images with a single measurement without sacrificing spatial resolution.

Free Research Field

情報フォトニクス

Academic Significance and Societal Importance of the Research Achievements

本研究は大きく分けて二つの課題の改善を目指したものである．まず，眼底検査の際に眼球内の屈折異常など結像系で観察が困難な場合に用いられる波面センサーの改善をおこなった．一般的に用いられるシャックハルトマン波面センサーに波面分割レンズアレイを導入し，解析アルゴリズムを改良することで，計測精度の改善，外乱や測定環境による影響への堅牢化，解析時間の高速化を達成した．また，複数の波長帯のイメージングを単一撮像でおこないながら空間分解能の改善を実証し，これは生体のように動的な物体に対しても適用可能である．これらを組み合わせたイメージング法は従来よりも高精度かつ広範な医療診断の可能性を期待できるものである．