Open AccessA homological approach to a mathematical definition of pulmonary fibrosis and emphysema on computed tomography
Author(s) -
Naoya Tanabe,
Shizuo Kaji,
Susumu Satô,
Tomoo Yokoyama,
Tsuyoshi Oguma,
Kiminobu Tanizawa,
Tomohiro Handa,
Takashi Sakajo,
Toyohiro Hirai
Publication year - 2021
Publication title -
journal of applied physiology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.253
H-Index - 229
eISSN - 8750-7587
pISSN - 1522-1601
DOI - 10.1152/japplphysiol.00150.2021
Subject(s) - interpretability , computed tomography , pulmonary emphysema , artificial intelligence , segmentation , feature (linguistics) , deep learning , tomography , medicine , radiology , computer science , pattern recognition (psychology) , mathematics , lung , linguistics , philosophy
This study proposes a homological approach to mathematically define a three-dimensional texture feature of emphysema and fibrosis on chest computed tomography using persistent homology. The proposed definition enabled accurate segmentation with comparable quality to deep learning while offering higher interpretability than deep learning-based methods.