Details of the Researcher

PHOTO

Eichi Takaya
Section
Tohoku University Hospital
Job title
Research Associate
Degree

  • 修士（工学）（電気通信大学）

e-Rad No.
40968327

Papers 48

  1. PlaNet-S: an Automatic Semantic Segmentation Model for Placenta Using U-Net and SegNeXt. International-journal Peer-reviewed

    Isso Saito, Shinnosuke Yamamoto, Eichi Takaya, Ayaka Harigai, Tomomi Sato, Tomoya Kobayashi, Kei Takase, Takuya Ueda

    Journal of imaging informatics in medicine　2025/05/27

    DOI： 10.1007/s10278-025-01549-9 　

    More details Close

    [Purpose] To develop a fully automated semantic placenta segmentation model that integrates the U-Net and SegNeXt architectures through ensemble learning. [Methods] A total of 218 pregnant women with suspected placental anomalies who underwent magnetic resonance imaging (MRI) were enrolled, yielding 1090 annotated images for developing a deep learning model for placental segmentation. The images were standardized and divided into training and test sets. The performance of PlaNet-S, which integrates U-Net and SegNeXt within an ensemble framework, was assessed using Intersection over Union (IoU) and counting connected components (CCC) against the U-Net model. [Results] PlaNet-S had significantly higher IoU (0.73 +/- 0.13) than that of U-Net (0.78 +/- 0.010) (p<0.01). The CCC for PlaNet-S was significantly higher than that for U-Net (p<0.01), matching the ground truth in 86.0\% and 56.7\% of the cases, respectively. [Conclusion]PlaNet-S performed better than the traditional U-Net in placental segmentation tasks. This model addresses the challenges of time-consuming physician-assisted manual segmentation and offers the potential for diverse applications in placental imaging analyses.

  2. Outcomes and cost-effectiveness of intermediate care units for patients discharged from the intensive care unit: a nationwide retrospective observational study. International-journal

    Saori Ikumi, Kunio Tarasawa, Takuya Shiga, Takahiro Imaizumi, Yu Kaiho, Yudai Iwasaki, Shizuha Yabuki, Yukito Wagatsuma, Eichi Takaya, Kiyohide Fushimi, Yukiko Ito, Kenji Fujimori, Masanori Yamauchi

    Critical care (London, England)　29　(1)　157-157　2025/04/23

    DOI： 10.1186/s13054-025-05393-9 　

    More details Close

    BACKGROUND: The clinical and economic impacts of intermediate care units (IMCUs) on intensive care unit (ICU)-discharged patients remain unclear due to inconsistent outcomes in previous studies. Under Japan's National Health Insurance Scheme, ICUs are categorized by staffing intensity (high or low). Using a nationwide inpatient database in Japan, we evaluated the clinical outcomes and cost-effectiveness of IMCUs for ICU-discharged patients. METHODS: This retrospective observational study used a Japanese administrative database to identify patients admitted to the high-intensity ICU in hospitals with IMCUs between April 2020 and March 2023. Patients were categorized into the IMCU (IMCU group) and general ward (non-IMCU) groups. Propensity scores were estimated using a logistic regression model incorporating 14 variables, including patient demographics, and treatments received during ICU stay. One-to-one propensity score matching balanced baseline characteristics of each group. Clinical outcomes were compared between both groups, including in-hospital mortality, ICU readmission, length of ICU stay, length of hospital stay, and total medical costs. Surgical status and surgical area (e.g., cardiovascular) were considered in subgroup analyses. Data analyses were conducted using the chi-square test for categorical variables and t-test for continuous variables. RESULTS: Overall, 162,243 eligible patients were categorized into the IMCU (n = 21,548) and non-IMCU (n = 140,695) groups. Propensity score matching generated 18,220 pairs. The IMCU group had lower in-hospital mortality and ICU readmission rates than the non-IMCU group. However, total costs were higher in the IMCU group. Subgroup analyses revealed the IMCU group had significantly lower mortality and lower total costs than the non-IMCU group in the cardiovascular [open thoracotomy] surgery subgroup. CONCLUSIONS: Discharge to an IMCU is associated with lower in-hospital mortality and ICU readmission rates compared to general ward discharge. High-risk subgroups, such as cardiovascular surgery patients, experienced cost-effective benefits from IMCU care. These findings highlight an association between IMCU admission and improved patient outcomes, suggesting a potential role in optimizing resource use in intensive care. Given the likelihood of selection bias in admission allocation, these findings should be interpretation with caution.

  3. Predictive Potential of Contrast-Enhanced MRI-Based Delta-Radiomics for Chemoradiation Responsiveness in Muscle-Invasive Bladder Cancer. International-journal

    Kohei Isemoto, Yuma Waseda, Motohiro Fujiwara, Koichiro Kimura, Daisuke Hirahara, Tatsunori Saho, Eichi Takaya, Yuki Arita, Thomas C Kwee, Shohei Fukuda, Hajime Tanaka, Soichiro Yoshida, Yasuhisa Fujii

    Diagnostics (Basel, Switzerland)　15　(7)　2025/03/21

    DOI： 10.3390/diagnostics15070801 　

    More details Close

    Background/Objectives: Delta-radiomics involves analyzing feature variations at different acquisition time-points. This study aimed to assess the utility of delta-radiomics feature analysis applied to contrast-enhanced (CE) and non-contrast-enhanced (NE) T1-weighted images (WI) in predicting the therapeutic response to chemoradiotherapy (CRT) in patients diagnosed with muscle-invasive bladder cancer (MIBC). Methods: Forty-three patients with non-metastatic MIBC (cT2-4N0M0) who underwent partial or radical cystectomy after induction CRT were, retrospectively, reviewed. Pathological complete response (pCR) to CRT was defined as the absence of residual viable tumor cells in the cystectomy specimen. Identical volumes of interest corresponding to the index bladder cancer lesions on CE- and NE-T1WI on pre-therapeutic 1.5-T MRI were collaboratively delineated by one radiologist and one urologist. Texture analysis was performed using "LIFEx" software. The subtraction of radiological features between CE- and NE-T1WI yielded 112 delta-radiomics features, which were utilized in multiple machine-learning algorithms to construct optimal predictive models for CRT responsiveness. Additionally, the predictive performance of the radiomics model constructed using CE-T1WI alone was assessed. Results: Twenty-one patients (49%) achieved pCR. The best-performing delta-radiomics model, employing the "Extreme Gradient Boosting" algorithm, yielded an area under the receiver operating characteristic curve (AUC) of 0.85 (95% confidence interval [CI]: 0.75-0.95), utilizing four signal intensity-based delta-radiomics features. This outperformed the best model derived from CE-T1WI alone (AUC: 0.63, 95% CI: 0.50-0.75), which incorporated two morphological features and one signal intensity-based radiomics feature. Conclusions: Delta-radiomics analysis applied to pre-therapeutic CE- and NE-MRI demonstrated promising predictive ability for CRT responsiveness prior to treatment initiation.

  4. Breast cancer classification based on breast tissue structures using the Jigsaw puzzle task in self-supervised learning

    Keisuke Sugawara, Eichi Takaya, Ryusei Inamori, Yuma Konaka, Jumpei Sato, Yuta Shiratori, Fumihito Hario, Tomoya Kobayashi, Takuya Ueda, Yoshikazu Okamoto

    Radiological Physics and Technology　2025/01/06

    Publisher: Springer Science and Business Media LLC

    DOI： 10.1007/s12194-024-00874-y 　

    ISSN： 1865-0333

    eISSN： 1865-0341

    More details Close

    Abstract Self-supervised learning (SSL) has gained attention in the medical field as a deep learning approach utilizing unlabeled data. The Jigsaw puzzle task in SSL enables models to learn both features of images and the positional relationships within images. In breast cancer diagnosis, radiologists evaluate not only lesion-specific features but also the surrounding breast structures. However, deep learning models that adopt a diagnostic approach similar to human radiologists are still limited. This study aims to evaluate the effectiveness of the Jigsaw puzzle task in characterizing breast tissue structures for breast cancer classification on mammographic images. Using the Chinese Mammography Database (CMMD), we compared four pre-training pipelines: (1) IN-Jig, pre-trained with both the ImageNet classification task and the Jigsaw puzzle task, (2) Scratch-Jig, pre-trained only with the Jigsaw puzzle task, (3) IN, pre-trained only with the ImageNet classification task, and (4) Scratch, that is trained from random initialization without any pre-training tasks. All pipelines were fine-tuned using binary classification to distinguish between the presence or absence of breast cancer. Performance was evaluated based on the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. Additionally, detailed analysis was conducted for performance across different radiological findings, breast density, and regions of interest were visualized using gradient-weighted class activation mapping (Grad-CAM). The AUC for the four models were 0.925, 0.921, 0.918, 0.909, respectively. Our results suggest the Jigsaw puzzle task is an effective pre-training method for breast cancer classification, with the potential to enhance diagnostic accuracy with limited data.

  5. Multimodal deep-learning model using pre-treatment endoscopic images and clinical information to predict efficacy of neoadjuvant chemotherapy in esophageal squamous cell carcinoma

    Takuma Miura, Takumi Yashima, Eichi Takaya, Yusuke Taniyama, Chiaki Sato, Hiroshi Okamoto, Yohei Ozawa, Hirotaka Ishida, Michiaki Unno, Takuya Ueda, Takashi Kamei

    Esophagus　2025

    DOI： 10.1007/s10388-025-01106-x 　

    ISSN： 1612-9059

    eISSN： 1612-9067

  6. Monitoring prostate cancer after low-dose-rate hemigland brachytherapy with delta-radiomics of diffusion-weighted magnetic resonance imaging

    Kotaro Shimada, Motohiro Fujiwara, Daisuke Hirahara, Eichi Takaya, Soichiro Yoshida, Yasuhisa Fujii

    International Journal of Urology　31　(12)　1443-1445　2024/12

    DOI： 10.1111/iju.15581 　

    ISSN： 0919-8172

    eISSN： 1442-2042

  7. Impact of Downsampling Size and Interpretation Methods on Diagnostic Accuracy in Deep Learning Model for Breast Cancer Using Digital Breast Tomosynthesis Images.

    Ryusei Inamori, Tomofumi Kaneno, Ken Oba, Eichi Takaya, Daisuke Hirahara, Tomoya Kobayashi, Kurara Kawaguchi, Maki Adachi, Daiki Shimokawa, Kengo Takahashi, Hiroko Tsunoda, Takuya Ueda

    The Tohoku journal of experimental medicine　2024/07/25

    DOI： 10.1620/tjem.2024.J071 　

  8. Deep learning model to predict Ki-67 expression of breast cancer using digital breast tomosynthesis. Peer-reviewed

    Ken Oba, Maki Adachi, Tomoya Kobayashi, Eichi Takaya, Daiki Shimokawa, Toshinori Fukuda, Kengo Takahashi, Kazuyo Yagishita, Takuya Ueda, Hiroko Tsunoda

    Breast cancer (Tokyo, Japan)　2024/03/07

    DOI： 10.1007/s12282-024-01549-7 　

    More details Close

    BACKGROUND: Developing a deep learning (DL) model for digital breast tomosynthesis (DBT) images to predict Ki-67 expression. METHODS: The institutional review board approved this retrospective study and waived the requirement for informed consent from the patients. Initially, 499 patients (mean age: 50.5 years, range: 29-90 years) referred to our hospital for breast cancer were participated, 126 patients with pathologically confirmed breast cancer were selected and their Ki-67 expression measured. The Xception architecture was used in the DL model to predict Ki-67 expression levels. The high Ki-67 vs low Ki-67 expression diagnostic performance of our DL model was assessed by accuracy, sensitivity, specificity, areas under the receiver operating characteristic curve (AUC), and by using sub-datasets divided by the radiological characteristics of breast cancer. RESULTS: The average accuracy, sensitivity, specificity, and AUC were 0.912, 0.629, 0.985, and 0.883, respectively. The AUC of the four subgroups separated by radiological findings for the mass, calcification, distortion, and focal asymmetric density sub-datasets were 0.890, 0.750, 0.870, and 0.660, respectively. CONCLUSIONS: Our results suggest the potential application of our DL model to predict the expression of Ki-67 using DBT, which may be useful for preoperatively determining the treatment strategy for breast cancer.

  9. Intensive care unit mortality and cost-effectiveness associated with intensivist staffing: a Japanese nationwide observational study Peer-reviewed

    Saori Ikumi, Takuya Shiga, Takuya Ueda, Eichi Takaya, Yudai Iwasaki, Yu Kaiho, Kunio Tarasawa, Kiyohide Fushimi, Yukiko Ito, Kenji Fujimori, Masanori Yamauchi

    Journal of Intensive Care　11　(1)　2023/12

    DOI： 10.1186/s40560-023-00708-w 　

    eISSN： 2052-0492

  10. Deep learning model for predicting the presence of stromal invasion of breast cancer on digital breast tomosynthesis Peer-reviewed

    Daiki Shimokawa, Kengo Takahashi, Ken Oba, Eichi Takaya, Takuma Usuzaki, Mizuki Kadowaki, Kurara Kawaguchi, Maki Adachi, Tomofumi Kaneno, Toshinori Fukuda, Kazuyo Yagishita, Hiroko Tsunoda, Takuya Ueda

    Radiological Physics and Technology　16　(3)　406-413　2023/09

    DOI： 10.1007/s12194-023-00731-4 　

    ISSN： 1865-0333

    eISSN： 1865-0341

  11. Machine learning approach to stratify complex heterogeneity of chronic heart failure: A report from the CHART-2 study

    Kenji Nakano, Kotaro Nochioka, Satoshi Yasuda, Daito Tamori, Takashi Shiroto, Yudai Sato, Eichi Takaya, Satoshi Miyata, Eiryo Kawakami, Tetsuo Ishikawa, Takuya Ueda, Hiroaki Shimokawa

    ESC Heart Failure　10　(3)　1597-1604　2023/06

    DOI： 10.1002/ehf2.14288 　

    eISSN： 2055-5822

  12. Deep learning model for breast cancer diagnosis based on bilateral asymmetrical detection (BilAD) in digital breast tomosynthesis images

    Daiki Shimokawa, Kengo Takahashi, Daiya Kurosawa, Eichi Takaya, Ken Oba, Kazuyo Yagishita, Toshinori Fukuda, Hiroko Tsunoda, Takuya Ueda

    Radiological Physics and Technology　16　(1)　20-27　2023/03

    DOI： 10.1007/s12194-022-00686-y 　

    ISSN： 1865-0333

    eISSN： 1865-0341

  13. Radiomics model of diffusion-weighted whole-body imaging with background signal suppression (DWIBS) for predicting axillary lymph node status in breast cancer

    Takafumi Haraguchi, Yasuyuki Kobayashi, Daisuke Hirahara, Tatsuaki Kobayashi, Eichi Takaya, Mariko Takishita Nagai, Hayato Tomita, Jun Okamoto, Yoshihide Kanemaki, Koichiro Tsugawa

    Journal of X-ray science and technology　31　(3)　627-640　2023

    DOI： 10.3233/XST-230009 　

    eISSN： 1095-9114

  14. Development and validation of machine learning prediction model for post-rehabilitation functional outcome after intracerebral hemorrhage

    Shinya Sonobe, Tetsuo Ishikawa, Kuniyasu Niizuma, Eiryo Kawakami, Takuya Ueda, Eichi Takaya, Carlos Makoto Miyauchi, Junya Iwazaki, Ryuzaburo Kochi, Toshiki Endo, Arun Shastry, Vijayananda Jagannatha, Ajay Seth, Atsuhiro Nakagawa, Masahiro Yoshida, Teiji Tominaga

    Interdisciplinary Neurosurgery: Advanced Techniques and Case Management　29　2022/09

    DOI： 10.1016/j.inat.2022.101560 　

    eISSN： 2214-7519

  15. Deep learning approach of diffusion-weighted imaging as an outcome predictor in laryngeal and hypopharyngeal cancer patients with radiotherapy-related curative treatment: a preliminary study

    Hayato Tomita, Tatsuaki Kobayashi, Eichi Takaya, Sono Mishiro, Daisuke Hirahara, Atsuko Fujikawa, Yoshiko Kurihara, Hidefumi Mimura, Yasuyuki Kobayashi

    European Radiology　32　(8)　5353-5361　2022/08

    DOI： 10.1007/s00330-022-08630-9 　

    ISSN： 0938-7994

    eISSN： 1432-1084

  16. Effect of the Pixel Interpolation Method for Downsampling Medical Images on Deep Learning Accuracy Peer-reviewed

    Daisuke Hirahara, Eichi Takaya, Mizuki Kadowaki, Yasuyuki Kobayashi, Takuya Ueda

    Journal of Computer and Communications　09　(11)　150-156　2021/11

    Publisher: Scientific Research Publishing, Inc.

    DOI： 10.4236/jcc.2021.911010 　

    ISSN： 2327-5219

    eISSN： 2327-5227

  17. Sequential semi-supervised segmentation for serial electron microscopy image with small number of labels

    Eichi Takaya, Yusuke Takeichi, Mamiko Ozaki, Satoshi Kurihara

    Journal of Neuroscience Methods　351　2021/03/01

    DOI： 10.1016/j.jneumeth.2021.109066 　

    ISSN： 0165-0270

    eISSN： 1872-678X

  18. Deep learning for the preoperative diagnosis of metastatic cervical lymph nodes on contrast-enhanced computed tomography in patients with oral squamous cell carcinoma

    Hayato Tomita, Tsuneo Yamashiro, Joichi Heianna, Toshiyuki Nakasone, Tatsuaki Kobayashi, Sono Mishiro, Daisuke Hirahara, Eichi Takaya, Hidefumi Mimura, Sadayuki Murayama, Yasuyuki Kobayashi

    Cancers　13　(4)　1-11　2021/02/02

    DOI： 10.3390/cancers13040600 　

    eISSN： 2072-6694

  19. Object Detection Model Utilizing Deep Learning to Identify Retained Surgical Gauze in the Body on Postoperative Radiography: Phantom Study

    田沼隆夫, 小林達明, 高屋英知, 鈴木大吾, 井上美香, 吉川達生, 小林泰之

    日本放射線技術学会雑誌　77　(8)　2021

    ISSN： 0369-4305

  20. Proposal of multi-dimensional data dimension reduction method using principle of synthetic wave

    蛭田興明, 高屋英知, 栗原聡

    人工知能学会全国大会論文集(Web)　35th　2021

    ISSN： 2758-7347

  21. Proposition of Multimodal Dialogue System Reflecting Human Emotions

    覚井悠生, 蛭田興明, 須賀聖, 川野陽慈, 高屋英知, 加藤慶彦, 栗原聡

    人工知能学会全国大会論文集(Web)　35th　2021

    ISSN： 2758-7347

  22. Proposal of an Interactive Action Selection Network

    小山宗三, 高屋英知, 加藤慶彦, 覚井悠生, 栗原聡

    情報処理学会研究報告(Web)　2020　(ICS-198)　2020

  23. Proposal of Multi Agent Planning in Dynamic Environment

    吉田直人, 高屋英知, 加藤慶彦, 栗原聡

    情報処理学会研究報告(Web)　2020　(ICS-198)　2020

  24. 矛盾塊自動生成手法の検討

    川野陽慈, 高屋英知, 栗原聡

    情報処理学会研究報告(Web)　2020　(ICS-198)　2020

  25. Proposal of Multi-agent Based Dynamic Traffic Signal Control System

    大野啓介, 神崎陽平, 高屋英知, 栗原聡

    情報処理学会研究報告(Web)　2020　(ICS-198)　2020

  26. Proposition of CNN based feature extraction of multi-channel data through data-colorization techniques

    蛭田興明, 高屋英知, 伊藤千輝, 荒牧大樹, 稲垣隆雄, 山岸典生, 栗原聡

    電子情報通信学会技術研究報告　119　(469(AI2019 54-64)(Web))　2020

    ISSN： 0913-5685

  27. Movie Script Story Development Structure Analysis and Plot Generation for Automatic Scenario Generation

    川野陽慈, 宇都宮悠輝, 高屋英知, 山野辺一記, 栗原聡

    人工知能学会全国大会論文集(Web)　34th　2020

    ISSN： 2758-7347

  28. Tactile tile detection integrated with ground detection using an RGB-depth sensor

    Yutaro Yamanaka, Eichi Takaya, Satoshi Kurihara

    ICAART 2020 - Proceedings of the 12th International Conference on Agents and Artificial Intelligence　2　750-757　2020

  29. Effects of data count and image scaling on Deep Learning training

    Daisuke Hirahara, Eichi Takaya, Taro Takahara, Takuya Ueda

    PeerJ Computer Science　6　1-13　2020

    DOI： 10.7717/peerj-cs.312 　

    eISSN： 2376-5992

  30. 物語生成のためのシナリオライター個性抽出に関する試み

    川野陽慈, 宇都宮悠輝, 高屋英知, 山野辺一紀, 立原祐樹, 栗原聡

    言語処理学会年次大会発表論文集(Web)　25th　2019

    ISSN： 2188-4420

  31. Proposition of Multimodal Time Series Data Analysis Framework by CNN based on Multi-Channel Image Conversion

    蛭田興明, 梁木俊冴, 高屋英知, 伊藤千輝, 荒牧大樹, 稲垣隆雄, 山岸典生, 栗原聡

    電子情報通信学会技術研究報告　118　(492(AI2018 53-59)(Web))　2019

    ISSN： 0913-5685

  32. 映像のシーンに対応した視聴者の感情・印象抽出

    山口想, 梁木俊冴, 高屋英知, 高野雅典, 森下壮一郎, 山本悠二, 福田鉄也, 福田一郎, 栗原聡

    情報処理学会研究報告(Web)　2019　(ICS-194)　2019

  33. Proposition of Multimodal Time Series Data Analysis Framework by CNN based on Multi-Channel Image Conversion

    蛭田興明, 梁木俊冴, 高屋英知, 伊藤千輝, 荒牧大樹, 稲垣隆雄, 山岸典生, 栗原聡

    人工知能学会全国大会論文集(Web)　33rd　2019

    ISSN： 2758-7347

  34. Proposal of latent sensory estimation method for BGM recommendation using VR and EEG measurement

    川野陽慈, 佐藤季久恵, 高屋英知, 須賀聖, 山内和樹, 栗原聡

    人工知能学会全国大会論文集(Web)　33rd　2019

    ISSN： 2758-7347

  35. Proposition of Pseudo-labeling for Segmentation in Stacks of Electron Microscopy Images

    高屋英知, 竹市裕介, 尾崎まみこ, 栗原聡

    人工知能学会全国大会論文集(Web)　33rd　2019

    ISSN： 2758-7347

  36. Semi-Supervised learning for electron microscopy image segmentation

    Eichi Takaya, Yusuke Takeichi, Mamiko Ozaki, Satoshi Kurihara

    33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Innovative Applications of Artificial Intelligence Conference, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2019　10047-10048　2019

  37. Putative Neural Network Within an Olfactory Sensory Unit for Nestmate and Non-nestmate Discrimination in the Japanese Carpenter Ant: The Ultra-structures and Mathematical Simulation

    Yusuke Takeichi, Tatsuya Uebi, Naoyuki Miyazaki, Kazuyoshi Murata, Kouji Yasuyama, Kanako Inoue, Toshinobu Suzaki, Hideo Kubo, Naoko Kajimura, Jo Takano, Toshiaki Omori, Ryoichi Yoshimura, Yasuhisa Endo, Masaru K. Hojo, Eichi Takaya, Satoshi Kurihara, Kenta Tatsuta, Koichi Ozaki, Mamiko Ozaki

    Frontiers in Cellular Neuroscience　12　2018/09/19

    DOI： 10.3389/fncel.2018.00310 　

    ISSN： 1662-5102

  38. Proposition of VR-MR hybrid system for sharing living-in-room

    Naoki Murata, Satoshi Ueno, Satoshi Suga, Yoji Kiyota, Eichi Takaya, Satoshi Kurihara

    RETech 2018 - Proceedings of the 2018 ACM Workshop on Multimedia for Real Estate Tech, Co-located with ICMR 2018　22-26　2018/06/08

    DOI： 10.1145/3210499.3210527 　

  39. Localization of innexins in the antennae of the Japanese carpenter ant, Camponotus japonicus and its putative involvement in the chemosensory mechanism for nestmate-nonnestmate discrimination Peer-reviewed

    Tatsuya Uebi, Yusuke Takeichi, Kouji Yasuyama, Naoyuki Miyazaki, Kazuyoshi Murata, Satoshi Kurihara, Eichi Takaya, Hideo Kubo, Toshiaki Omori, Mamiko Ozaki

    CHEMICAL SENSES　43　(5)　E142-E142　2018/06

    ISSN： 0379-864X

    eISSN： 1464-3553

  40. 神経細胞セグメンテーションにおける深層畳み込みアーキテクチャの適用方法に関する検討

    高屋英知, 竹市裕介, 尾崎まみこ, 栗原聡

    情報処理学会研究報告(Web)　2018　(ICS-190)　2018

  41. ソーシャルメディアにおけるアクティブソーシャルセンシングを用いた楽曲視聴状況の抽出

    佐藤圭, 池田圭佑, 坂井栞, 川野陽慈, 高屋英知, 栗原聡, 山内和樹, 大矢隼士

    電子情報通信学会技術研究報告　117　(468(AI2017 43-51))　2018

    ISSN： 0913-5685

  42. A study on traffic flow prediction method using probe information in transportation system

    大野啓介, 高屋英知, 松本洋, 森田哲郎, 栗原聡

    人工知能学会全国大会論文集(CD-ROM)　32nd　2018

    ISSN： 1347-9881

  43. BGM Recommendation System Considering the Atmosphere

    佐藤季久恵, 坂井栞, 高屋英知, 山内和樹, 大矢隼士, 栗原聡

    人工知能学会全国大会論文集(CD-ROM)　32nd　2018

    ISSN： 1347-9881

  44. Segmentation in Stacks of Electron Microscopy Images Using Deep Learning

    高屋英知, 竹市裕介, 尾崎まみこ, 栗原聡

    人工知能学会全国大会論文集(CD-ROM)　32nd　2018

    ISSN： 1347-9881

  45. Proposal of traffic signal control system by multi agent using Deep Q-Network

    神崎陽平, 佐藤季久恵, 高屋英知, 小川亮, 芦原佑太, 芦原佑太, 栗原聡

    人工知能学会全国大会論文集(CD-ROM)　32nd　2018

    ISSN： 1347-9881

  46. Automatic plot generation framework for scenario creation

    Yoji Kawano, Eichi Takaya, Kazuki Yamanobe, Satoshi Kurihara

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)　11318 LNCS　453-461　2018

    DOI： 10.1007/978-3-030-04028-4_52 　

    ISSN： 0302-9743

    eISSN： 1611-3349

  47. Development of Traffic Control System by use of Deep Q-Network

    佐藤季久恵, 高屋英知, 小川亮, 芦原佑太, 芦原佑太, 栗原聡

    人工知能学会全国大会論文集(CD-ROM)　31st　2017

    ISSN： 1347-9881

  48. Autonomy and Multimodality for Constructing Artificial General Intelligence

    栗原聡, 栗原聡, 高屋英知, 高橋良暢, 芦原祐太

    人工知能学会全国大会論文集(CD-ROM)　31st　2017

    ISSN： 1347-9881

Show all ︎Show first 5

Misc. 14

  1. In-context learning for medical image segmentation

    Eichi Takaya, Shinnosuke Yamamoto

    2024/12/17

    More details Close

    Annotation of medical images, such as MRI and CT scans, is crucial for evaluating treatment efficacy and planning radiotherapy. However, the extensive workload of medical professionals limits their ability to annotate large image datasets, posing a bottleneck for AI applications in medical imaging. To address this, we propose In-context Cascade Segmentation (ICS), a novel method that minimizes annotation requirements while achieving high segmentation accuracy for sequential medical images. ICS builds on the UniverSeg framework, which performs few-shot segmentation using support images without additional training. By iteratively adding the inference results of each slice to the support set, ICS propagates information forward and backward through the sequence, ensuring inter-slice consistency. We evaluate the proposed method on the HVSMR dataset, which includes segmentation tasks for eight cardiac regions. Experimental results demonstrate that ICS significantly improves segmentation performance in complex anatomical regions, particularly in maintaining boundary consistency across slices, compared to baseline methods. The study also highlights the impact of the number and position of initial support slices on segmentation accuracy. ICS offers a promising solution for reducing annotation burdens while delivering robust segmentation results, paving the way for its broader adoption in clinical and research applications.

  2. Exploring annotation cost reduction in breast cancer segmentation using weak labeling with active learning

    針尾文仁, 高屋英知, 高屋英知, 小林智哉, 小仲悠真, 佐藤潤平, 菅原圭亮, 植田琢也, 岡本嘉一

    日本診療放射線技師会誌　71　(10)　2024

    ISSN： 2187-2538

  3. Evaluation of breast cancer diagnosis AI in mammography using multimodal learning method with images and text

    佐藤潤平, 高屋英知, 高屋英知, 小林智哉, 植田琢也, 岡本嘉一

    日本診療放射線技師会誌　71　(10)　2024

    ISSN： 2187-2538

  4. Effect of X-ray image with Gaussian noise on AI model training

    菅原圭亮, 高屋英知, 高屋英知, 小林智哉, 植田琢也, 岡本嘉一

    日本診療放射線技師会誌　71　(10)　2024

    ISSN： 2187-2538

  5. Cost-effectiveness analysis for intensive care unit management fees 1 and 2.

    井汲沙織, 志賀卓弥, 岩崎夢大, 海法悠, 高屋英知, たら澤邦男, 伊藤由希子, 藤森研司, 山内正憲

    日本集中治療医学会学術集会(Web)　50th　2023

  6. 乳癌の深層学習Segmentationにおける人工生成Fractal画像を用いた転移学習の有効性の検討

    八島拓海, 新城勇弥, 高屋英知, 小林智哉, 大庭建, 八木下和代, 角田博子, 植田琢也, 植田琢也

    日本乳癌画像研究会プログラム・抄録集　32nd (Web)　2023

  7. 医学的見地を考慮した病変分類に基づく,乳房トモシンセシスの乳癌画像診断AIモデルの検討

    安達眞紀, 川口くらら, 金野智史, 大庭建, 高屋英知, 八木下和代, 角田博子, 植田琢也, 植田琢也

    日本乳癌画像研究会プログラム・抄録集　31st　2022

  8. 大規模自然画像データの転移学習(Big Transfer)による乳房トモシンセシス乳癌AI診断の精度向上の試み

    高橋健吾, 下川大輝, 大庭建, 高屋英知, 川口くらら, 立石朱紗美, 安達眞紀, 金野智史, 八木下和代, 福田俊憲, 角田博子, 植田琢也, 植田琢也

    日本乳癌学会学術総会プログラム・抄録集　30th (CD-ROM)　2022

  9. 医療・ヘルスケア領域におけるAI-AIの基礎から活用まで 人工知能の基礎知識 I ざっくり学ぶ深層学習の仕組み

    高屋英知

    Rad Fan　19　(7)　2021

    ISSN： 1348-3498

  10. AIを用いた定量的かつ動的な転倒転落リスクの検出

    園部真也, 園部真也, 門脇水樹, 門脇水樹, 石川哲朗, 石川哲朗, 高屋英知, 高屋英知, 菅原寛子, 中川敦寛, 中川敦寛, 植田琢也, 植田琢也, 川上英良, 川上英良, 田畑雅央, 冨永悌二

    医療の質・安全学会誌　16　(Supplement)　2021

    ISSN： 1881-3658

  11. 腫瘤形成性乳癌の乳房トモシンセシス画像のAI解析によるセンチネルリンパ節転移リスクの予測

    下川大輝, 高橋健吾, 渋谷貴一, 高屋英知, 平原大助, 臼崎琢磨, 大庭建, 福田俊憲, 八木下和代, 角田博子, 植田琢也

    日本乳癌学会学術総会プログラム・抄録集　29th (CD-ROM)　2021

  12. 人工知能基礎知識 II(深層学習)

    高屋英知, 高屋英知

    Japanese Journal of Diagnostic Imaging　39　2021

    ISSN： 2187-266X

  13. 電子顕微鏡連続切片画像に対する半自動セグメンテーション手法の提案

    高屋英知

    天野工業技術研究所年次報告　2020　2021

  14. 新時代の乳腺画像診断 MRI:乳房MRIの現状と将来展望について

    印牧義英, 小林達明, 高屋英知, 平原大助, 小林泰之, 三村秀文, 津川浩一郎

    日本医学放射線学会秋季臨床大会抄録集　55th　2019

    ISSN： 0048-0428

Show all ︎Show first 5

Books and Other Publications 2

  1. サクッとわかる医療AI

    小林, 泰之, 平原, 大助, 高屋, 英知

    シービーアール　2022/06

    ISBN: 9784908083792

  2. Pythonによる医用画像処理入門

    上杉, 正人, 平原, 大助, 齋藤, 静司

    オーム社　2020/04

    ISBN: 9784274225468

Research Projects 1

  1. 医用画像データ作成エコシステムの開発

    高屋 英知

    Offer Organization: 日本学術振興会

    System: 科学研究費助成事業

    Category: 基盤研究(C)

    Institution: 東北大学

    2024/04/01 - 2027/03/31