研究者詳細 - 中澤　篤志

その他リンク： https://dblp.uni-trier.de/db/conf/icip/icip2019.html#YamamotoN19

優しい介護インタラクションの計算的・脳科学的解明査読

中澤篤志

ヒューマンインタフェースシンポジウム 2018 ( UBI-59 ) 2018年9月

Development of a skill evaluation system for the camera assistant using an infant-sized laparoscopic box trainer

Tetsuya Ishimaru, Kyoichi Deie, Tomoya Sakai, Hideyuki Satoh, Atsushi Nakazawa, Kanako Harada, Shinya Takazawa, Jun Fujishiro, Naohiko Sugita, Mamoru Mitsuishi, Tadashi Iwanaka

Journal of Laparoendoscopic and Advanced Surgical Techniques 28 ( 7 ) 906 - 911 2018年7月

掲載種別：研究論文（学術雑誌）

Aims: Our aims were to develop a training system for camera assistants (CA), and evaluate participants' performance as CA. Methods: A questionnaire on essential requirements to be a good CA was administered to experts in pediatric endoscopic surgery. An infant-sized box trainer with several markers and lines inside was developed. Participants performed marker capturing and line-tracing tasks using a 5-mm 30° scope. A postexperimental questionnaire on the developed system was administered. The task completion time was measured. Results: The 5-point evaluation scale was used for each item in the questionnaire survey of experts. The abilities to maintain a horizontal line (mean score: 4.5) and to center the target in a specified rectangle on the monitor (4.5) as well as having a full understanding of the operative procedure (4.3) were ranked as highly important. Fifty-two participants, including 5 surgical residents, were enrolled in the evaluation experiment. The completion time of capturing the markers was significantly longer in the resident group than in the nonresident group (244 versus 124 seconds, P = .04), but that of tracing the lines was not significantly different between the groups. The postexperimental questionnaire showed that the participants felt that the line-tracing tasks (3.7) were more difficult than marker-capturing tasks (2.9). Conclusions: Being proficient in manipulating a camera and having adequate knowledge of operative procedures are essential requirements to be a good CA. The ability was different between the resident and nonresident groups even in a simple task such as marker capturing.

DOI： 10.1089/lap.2017.0406

PubMed

Towards robust needle segmentation and tracking in pediatric endoscopic surgery.

Computational Tender-Care Science: Computational and Cognitive Neuro-scientific Approaches for Understanding the Tender Care 査読

Atsushi Nakazawa, Kyoto University, Japan, Ryo Kurazume, Kyushu University, Japan, Miwako Honda, Tokyo Medical Center, Sato Wataru, Kyoto University, Japan, Shogo Ishikawa, Shizuoka University, Japan, Sakiko Yoshikawa

Workshop on Elderly Care Robotics – Technology and Ethics 2018 ( UBI-59 ) 2018年5月

Computational Tender-Care Science: Computational and Cognitive Neuroscientific Approaches for Understanding the Tender Care 査読

Nakazawa A, Kurazume R, Honda M, Sato W, Ishikawa S, Yoshikawa S, Ito M

IUI Workshop on Symbiotic Interaction and Harmonius Collaboration for Wisdom Computing 1 ( UBI-59 ) 1 - 9 2018年3月

深層学習を用いた可視光画像からの瞳孔検出と注視点推定への応用

北角一哲, 中澤篤志, 西田豊明

電子情報通信学会技術研究報告 117 ( 392 ) 93 - 99 2018年1月

Yujun Chen, Murilo M. Marinho, Yusuke Kurose, Atsushi Nakazawa, Kyoichi Deie, Kanako Harada, Mamoru Mitsuishi

Medical Imaging 2018: Image-Guided Procedures, Robotic Interventions, and Modeling 105762 2018年

Point of Gaze Estimation Using Corneal Surface Reflection and Omnidirectional Camera Image. 査読

Taishi Ogawa, Atsushi Nakazawa, Toyoaki Nishida

IEICE Trans. Inf. Syst. 101-D ( 5 ) 1278 - 1287 2018年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Institute of Electronics, Information and Communication, Engineers, IEICE

We present a human point of gaze estimation system using corneal surface reflection and omnidirectional image taken by spherical panorama cameras, which becomes popular recent years. Our system enables to find where a user is looking at only from an eye image in a 360° surrounding scene image, thus, does not need gaze mapping from partial scene images to a whole scene image that are necessary in conventional eye gaze tracking system. We first generate multiple perspective scene images from an omnidirectional (equirectangular) image and perform registration between the corneal reflection and perspective images using a corneal reflection-scene image registration technique. We then compute the point of gaze using a corneal imaging technique leveraged by a 3D eye model, and project the point to an omnidirectional image. The 3D eye pose is estimate by using the particle-filter-based tracking algorithm. In experiments, we evaluated the accuracy of the 3D eye pose estimation, robustness of registration and accuracy of PoG estimations using two indoor and five outdoor scenes, and found that gaze mapping error was 5.546 [deg] on average.

DOI： 10.1587/transinf.2017MVP0020

その他リンク： https://dblp.uni-trier.de/db/journals/ieicet/ieicet101d.html#OgawaNN18

PRMU応用研究におけるオープンアイデア～ PRMU第二期グランドチャレンジ～招待

中澤篤志, 山崎俊彦, 松下康之, 安倍満, 舩冨卓哉, 木村昭悟, 内田誠一, 前田英作

電子情報通信学会技術研究報告 117 ( 362 ) 41 - 43 2017年12月

深層学習を用いた自己撮影画像の撮影位置検索

江川佳輝, 小川太士, 中澤篤志

電子情報通信学会技術研究報告 117 ( 392 ) 333 - 337 2017年12月

DEEP eye contact detector: Robust eye contact bid detection using convolutional neural network 査読

Mitsuzumi Y, Nakazawa A, Nishida T

British Machine Vision Conference (BMVC2017) 2017年9月

Noninvasive Corneal Image-Based Gaze Measurement System 査読国際共著

Eunji Chong, Christian Nitschke, Atsushi Nakazawa, Agata Rozga, James M Rehg

arXiv preprint arXiv:1708.00908 1 1 - 8 2017年8月

Point of gaze estimation using corneal surface reflection and omnidirectional camera image 査読

Taishi Ogawa, Atsushi Nakazawa, Toyoaki Nishida

Proceedings of the 15th IAPR International Conference on Machine Vision Applications, MVA 2017 227 - 230 2017年7月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：Institute of Electrical and Electronics Engineers Inc.

We present a human point of gaze estimation system using corneal surface reflection and omni-directional image taken by a fish eye. Only capturing an eye image, our system entables to find where a user is looking in 360P surrounding scene image. We first generates multiple perspective scene images from an equi-rectangular image and perform registration between corneal reflection and perspective images. We then compute the point of gaze using a 3D eye model and project the point to an omni-directional image. We evaluated the robustness of registration and accuracy of PoG estimations using two indoor and five outdoor scenes, and found that gaze mapping error was 5.526[deg] on average. This result shows the potential to the marketing and outdoor training system.

DOI： 10.23919/MVA.2017.7986842

Corneal Imaging

Christian Nitschke, Atsushi Nakazawa

The Wiley Handbook of Human Computer Interaction Set 1 445 - 514 2017年6月

記述言語：英語掲載種別：論文集(書籍)内論文出版者・発行元：Wiley Blackwell

The cornea of the human eye acts as a mirror that reflects light from a person's environment. These corneal reflections can be extracted from an image of the eye by modeling the eye-camera geometry as a catadioptric imaging system. As a result, one obtains the visual information of the environment and the relation to the observer (view, gaze), which allows for applications in a number of fields, most notably eye tracking. The recovered illumination map can be further applied to various computational tasks. This chapter provides a comprehensive introduction and overview on corneal imaging and applications. It includes (1) an explanation of anatomic structures and geometric modeling of the eye, (2) eye tracking and pose estimation from eye images, (3) corneal reflection modeling and image processing, (4) applications, and (5) a discussion of implications and promising future directions. While mainly targeting readers in eye image processing and eye tracking, this chapter is intended for a diverse audience with different backgrounds, levels of experience, and theoretical and practical interest.

DOI： 10.1002/9781118976005.ch21

医療コミュニケーションにおける人工知能の可能性医師の「五感」をも定量化できる時代へ招待

中澤篤志

総合診療 27 ( 5 ) 621 - 623 2017年5月

Eye gaze tracking using corneal imaging and active illumination devices. 査読

Atsushi Nakazawa, Hiroaki Kato, Christian Nitschke, Toyoaki Nishida

Adv. Robotics 31 ( 8 ) 413 - 427 2017年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：TAYLOR & FRANCIS LTD

This paper shows a novel eye gaze tracking (EGT) technique using the corneal imaging technique. Compared to the existing pupil center and pupil reflection techniques, our approach directly finds PoG in the reflected scene image at the human corneal surface. As a result, it does not suffer from the parallax issue and does not require per-setup system calibrations. To achieve this system, we develop following techniques: First, we use the idea of the gaze-reflection point (GRP), where light from the PoG in the scene reflects at the corneal surface into an eye image. Second, illuminating the whole scene or particular objects using coded structured light enables robust and accurate matching at the GRP to obtain the PoG in a scene image. For this purpose, we show two implementations: a special high-power IR LED-array projector and active LED markers. Experimental evaluation shows that the proposed scheme achieves considerable accuracy and successfully supports depth-varying environments as well as practical applications including observation in a conversation scene. We believe the proposed EGT technique has considerably large potential to solve major issues of current EGT systems, expands the application fields of the EGT and increases the usability of interactive systems.

DOI： 10.1080/01691864.2016.1277552

その他リンク： https://dblp.uni-trier.de/db/journals/ar/ar31.html#NakazawaKNN17

Toward autonomous collision avoidance for robotic neurosurgery in deep and narrow spaces in the brain

Hiroaki Ueda, Ryoya Suzuki, Atsushi Nakazawa, Yusuke Kurose, Murilo M. Marinho, Naoyuki Shono, Hirofumi Nakatomi, Nobuhito Saito, Eiju Watanabe, Akio Morita, Kanako Harada, Naohiko Sugita, Mamoru Mitsuishi

3RD CIRP CONFERENCE ON BIOMANUFACTURING 65 110 - 114 2017年

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：ELSEVIER SCIENCE BV

The present authors have been developing a master-slave neurosurgical robot and its intelligent control for tasks in the deep and narrow spaces of the brain. This paper proposes a robotic autonomous control method for avoiding possible collisions between the shaft of a surgical robotic instrument and the surrounding tissues. To this end, a new robotic simulator was developed and used to evaluate the proposed method. The results showed the proof of concept of the proposed autonomous collision avoidance, which has the potential to enhance the safety of robotic neurosurgery in deep and narrow spaces. (C) 2016 The Authors. Published by Elsevier B.V.

DOI： 10.1016/j.procir.2017.04.027

Registration of eye reflection and scene images using an aspherical eye model 査読

Atsushi Nakazawa, Christian Nitschke, Toyoaki Nishida

JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION 33 ( 11 ) 2264 - 2276 2016年11月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：OPTICAL SOC AMER

This paper introduces an image registration algorithm between an eye reflection and a scene image. Although there are currently a large number of image registration algorithms, this task remains difficult due to nonlinear distortions at the eye surface and large amounts of noise, such as iris texture, eyelids, eyelashes, and their shadows. To overcome this issue, we developed an image registration method combining an aspherical eye model that simulates nonlinear distortions considering eye geometry and a two-step iterative registration strategy that obtains dense correspondence of the feature points to achieve accurate image registrations for the entire image region. We obtained a database of eye reflection and scene images featuring four subjects in indoor and outdoor scenes and compared the registration performance with different asphericity conditions. Results showed that the proposed approach can perform accurate registration with an average accuracy of 1.05 deg by using the aspherical cornea model. This work is relevant for eye image analysis in general, enabling novel applications and scenarios. (C) 2016 Optical Society of America

DOI： 10.1364/JOSAA.33.002264

Feedback methods for collision avoidance using virtual fixtures for robotic neurosurgery in deep and narrow spaces

Atsushi Nakazawa, Kodai Nanri, Kanako Harada, Shinichi Tanaka, Hiroshi Nukariya, Yusuke Kurose, Naoyuki Shono, Hirohumi Nakatomi, Akio Morita, Eiju Watanabe, Naohiko Sugita, Mamoru Mitsuishi

Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics 2016-July 247 - 252 2016年7月

掲載種別：研究論文（国際会議プロシーディングス）

Robotic assistance enables a surgeon to perform dexterous and precise manipulations. However, conducting robot assisted neurosurgery within the deep and narrow spaces of the brain presents the risk of unexpected collisions between the shafts of robotic instruments and their surroundings out of the microscopic view. Thus, we propose the provision of feedback using a truncated cone shaped virtual fixture generated by marking the edges of the top and bottom plane of a workspace in the deep and narrow spaces within the brain with the slave manipulator. The experimental results show that the virtual fixture generation method could precisely model the workspace. We also implemented force feedback, visual feedback, and motion scaling feedback in the microsurgical robotic system in order to inform the surgeon of the risk of collision. Performance of each feedback method and their combinations was evaluated in two experiments. The experimental results showed that the combination of the force and the visual feedback methods were the most beneficial for avoiding collisions.

DOI： 10.1109/BIOROB.2016.7523632