Shuichiro Komatsu1, Takahiro Oike2, Yuka Komatsu1, Yoshiki Kubota3, Makoto Sakai3, Toshiaki Matsui1, Endang Nuryadi4, Tiara Bunga Mayang Permata4, Hiro Sato1, Hidemasa Kawamura3, Masahiko Okamoto3, Takuya Kaminuma3, Kazutoshi Murata3, Naoko Okano1, Yuka Hirota1, Tatsuya Ohno3, Jun-Ichi Saitoh5, Atsushi Shibata6, Takashi Nakano7. 1. Department of Radiation Oncology, Gunma University Graduate School of Medicine, Gunma, Japan. 2. Department of Radiation Oncology, Gunma University Graduate School of Medicine, Gunma, Japan. Electronic address: oiketakahiro@gmail.com. 3. Gunma University Heavy Ion Medical Center, Gunma, Japan. 4. Department of Radiation Oncology, Gunma University Graduate School of Medicine, Gunma, Japan; Department of Radiotherapy, Dr. Cipto Mangunkusumo National General Hospital - Faculty of Medicine Universitas Indonesia, Jakarta, Indonesia. 5. Department of Radiation Oncology, University of Toyama Faculty of Medicine, Japan. 6. Gunma University Initiative for Advanced Research (GIAR), Gunma University, Gunma, Japan. 7. Department of Radiation Oncology, Gunma University Graduate School of Medicine, Gunma, Japan; Gunma University Heavy Ion Medical Center, Gunma, Japan.
Abstract
BACKGROUND AND PURPOSE: Integrated analysis of existing radiosensitivity data obtained by the gold-standard clonogenic assay has the potential to improve our understanding of cancer cell radioresistance. However, extraction of radiosensitivity data from the literature is highly labor-intensive. To aid in this task, using deep convolutional neural networks (CNNs) and other computer technologies, we developed an analysis pipeline that extracts radiosensitivity data derived from clonogenic assays from the literature. MATERIALS AND METHODS: Three classifiers (C1-3) were developed to identify publications containing radiosensitivity data derived from clonogenic assays. C1 uses Faster Regions CNN with Inception Resnet v2 (fRCNN-IRv2), VGG-16, and Optical Character Recognition (OCR) to identify publications that contain semi-logarithmic graphs showing radiosensitivity data derived from clonogenic assays. C2 uses fRCNN-IRv2 and OCR to identify publications that contain bar graphs showing radiosensitivity data derived from clonogenic assays. C3 is a program that identifies publications containing keywords related to radiosensitivity data derived from clonogenic assays. A program (iSF2) was developed using Mask RCNN and OCR to extract surviving fraction after 2-Gy irradiation (SF2) as assessed by clonogenic assays, presented in semi-logarithmic graphs. The efficacy of C1-3 and iSF2 was tested using seven datasets (1805 and 222 publications in total, respectively). RESULTS: C1-3 yielded sensitivity of 91.2% ± 3.4% and specificity of 90.7% ± 3.6%. iSF2 returned SF2 values that were within 2.9% ± 2.6% of the SF2 values determined by radiation oncologists. CONCLUSION: Our analysis pipeline is potentially useful to acquire radiosensitivity data derived from clonogenic assays from the literature.
BACKGROUND AND PURPOSE: Integrated analysis of existing radiosensitivity data obtained by the gold-standard clonogenic assay has the potential to improve our understanding of cancer cell radioresistance. However, extraction of radiosensitivity data from the literature is highly labor-intensive. To aid in this task, using deep convolutional neural networks (CNNs) and other computer technologies, we developed an analysis pipeline that extracts radiosensitivity data derived from clonogenic assays from the literature. MATERIALS AND METHODS: Three classifiers (C1-3) were developed to identify publications containing radiosensitivity data derived from clonogenic assays. C1 uses Faster Regions CNN with Inception Resnet v2 (fRCNN-IRv2), VGG-16, and Optical Character Recognition (OCR) to identify publications that contain semi-logarithmic graphs showing radiosensitivity data derived from clonogenic assays. C2 uses fRCNN-IRv2 and OCR to identify publications that contain bar graphs showing radiosensitivity data derived from clonogenic assays. C3 is a program that identifies publications containing keywords related to radiosensitivity data derived from clonogenic assays. A program (iSF2) was developed using Mask RCNN and OCR to extract surviving fraction after 2-Gy irradiation (SF2) as assessed by clonogenic assays, presented in semi-logarithmic graphs. The efficacy of C1-3 and iSF2 was tested using seven datasets (1805 and 222 publications in total, respectively). RESULTS:C1-3 yielded sensitivity of 91.2% ± 3.4% and specificity of 90.7% ± 3.6%. iSF2 returned SF2 values that were within 2.9% ± 2.6% of the SF2 values determined by radiation oncologists. CONCLUSION: Our analysis pipeline is potentially useful to acquire radiosensitivity data derived from clonogenic assays from the literature.