TY - JOUR
T1 - A study of evaluating specific tissue oxygen saturation values of gastrointestinal tumors by removing adherent substances in oxygen saturation imaging
AU - Nishihara, Keiichiro
AU - Hori, Keisuke
AU - Saito, Takaaki
AU - Omori, Toshihiko
AU - Sunakawa, Hironori
AU - Minamide, Tatsunori
AU - Suyama, Masayuki
AU - Yamamoto, Yoichi
AU - Yoda, Yusuke
AU - Shinmura, Kensuke
AU - Ikematsu, Hiroaki
AU - Yano, Tomonori
N1 - Funding Information:
The National Cancer Center Research and Development Fund provided support for this study in the form of a grant awarded to TY (29-A-10). FUJIFILM Corporation also provided support for this study in the form of a research grant awarded to TY, oxygen saturation imaging for the clinical study, and salaries for TS and TO. The specific roles of these authors are articulated in the ?author contributions? section. FUJIFILM played a role in the image analysis and oxygen saturation endoscopy system, but no other role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Publisher Copyright:
© 2021 Nishihara et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
PY - 2021/1
Y1 - 2021/1
N2 - Objectives Oxygen saturation (OS) imaging is a new method of endoscopic imaging that has clinical applications in oncology which can directly measure tissue oxygen saturation (Sto2) of the surface of gastrointestinal tract without any additional drugs or devices. This imaging technology is expected to contribute to research into cancer biology which leads to clinical benefit such as prediction to efficacy of chemotherapy or radiotherapy. However, adherent substances on tumors such as blood and white coating, pose a challenge for accurate measurements of the StO2 values in tumors. The aim of this study was to develop algorithms for discriminating between the tumors and their adherent substances, and to investigate whether it is possible to evaluate the tumor specific StO2 values excluding adherent substances during OS imaging. Methods We plotted areas of tumors and their adherent substances using white-light images of 50 upper digestive tumors: blood (68 plots); reddish tumor (83 plots); white coating (89 plots); and whitish tumor (79 plots). Scatter diagrams and discriminating algorithms using spectrum signal intensity values were constructed and verified using validation datasets. StO2 values were compared between the tumors and tumor adherent substances using OS images of gastrointestinal tumors. Results The discriminating algorithms and their accuracy rates (AR) were as follows: blood vs. reddish tumor: Y> - 4.90X+7.13 (AR: 95.9%) and white coating vs. whitish tumor: Y< -0.52X +0.17 (AR: 96.0%). The StO2 values (median, [range]) were as follows: blood, 79.3% [37.8%–100.0%]; reddish tumor, 74.5% [62.0%–86.9%]; white coating, 73.8% [42.1%–100.0%]; and whitish tumor, 65.7% [53.0%–76.3%]. Conclusions OS imaging is strongly influenced by adherent substances for evaluating the specific StO2 value of tumors; therefore, it is important to eliminate the information of adherent substances for clinical application of OS imaging.
AB - Objectives Oxygen saturation (OS) imaging is a new method of endoscopic imaging that has clinical applications in oncology which can directly measure tissue oxygen saturation (Sto2) of the surface of gastrointestinal tract without any additional drugs or devices. This imaging technology is expected to contribute to research into cancer biology which leads to clinical benefit such as prediction to efficacy of chemotherapy or radiotherapy. However, adherent substances on tumors such as blood and white coating, pose a challenge for accurate measurements of the StO2 values in tumors. The aim of this study was to develop algorithms for discriminating between the tumors and their adherent substances, and to investigate whether it is possible to evaluate the tumor specific StO2 values excluding adherent substances during OS imaging. Methods We plotted areas of tumors and their adherent substances using white-light images of 50 upper digestive tumors: blood (68 plots); reddish tumor (83 plots); white coating (89 plots); and whitish tumor (79 plots). Scatter diagrams and discriminating algorithms using spectrum signal intensity values were constructed and verified using validation datasets. StO2 values were compared between the tumors and tumor adherent substances using OS images of gastrointestinal tumors. Results The discriminating algorithms and their accuracy rates (AR) were as follows: blood vs. reddish tumor: Y> - 4.90X+7.13 (AR: 95.9%) and white coating vs. whitish tumor: Y< -0.52X +0.17 (AR: 96.0%). The StO2 values (median, [range]) were as follows: blood, 79.3% [37.8%–100.0%]; reddish tumor, 74.5% [62.0%–86.9%]; white coating, 73.8% [42.1%–100.0%]; and whitish tumor, 65.7% [53.0%–76.3%]. Conclusions OS imaging is strongly influenced by adherent substances for evaluating the specific StO2 value of tumors; therefore, it is important to eliminate the information of adherent substances for clinical application of OS imaging.
UR - http://www.scopus.com/inward/record.url?scp=85099439610&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85099439610&partnerID=8YFLogxK
U2 - 10.1371/journal.pone.0243165
DO - 10.1371/journal.pone.0243165
M3 - Article
C2 - 33411775
AN - SCOPUS:85099439610
SN - 1932-6203
VL - 16
JO - PLoS One
JF - PLoS One
IS - 1 January
M1 - e0243165
ER -