Shintaro Suzuki, Xiaoyong Zhang, N. Homma, K. Ichiji, N. Sugita, Y. Kawasumi, T. Ishibashi, M. Yoshizawa
{"title":"Mass detection using deep convolutional neural network for mammographic computer-aided diagnosis","authors":"Shintaro Suzuki, Xiaoyong Zhang, N. Homma, K. Ichiji, N. Sugita, Y. Kawasumi, T. Ishibashi, M. Yoshizawa","doi":"10.1109/SICE.2016.7749265","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749265","url":null,"abstract":"In recent years, a deep convolutional neural network (DCNN) has attracted great attention due to its outstanding performance in recognition of natural images. However, the DCNN performance for medical image recognition is still uncertain because collecting a large amount of training data is difficult. To solve the problem of the DCNN, we adopt a transfer learning strategy, and demonstrate feasibilities of the DCNN and of the transfer learning strategy for mass detection in mammographic images. We adopt a DCNN architecture that consists of 8 layers with weight, including 5 convolutional layers, and 3 fully-connected layers in this study. We first train the DCNN using about 1.2 million natural images for classification of 1,000 classes. Then, we modify the last fully-connected layer of the DCNN and subsequently train the DCNN using 1,656 regions of interest in mammographic image for two classes classification: mass and normal. The detection test is conducted on 198 mammographic images including 99 mass images and 99 normal images. The experimental results showed that the sensitivity of the mass detection was 89.9 % and the false positive was 19.2 %. These results demonstrated that the DCNN trained by transfer learning strategy has a potential to be a key system for mammographic mass detection computer-aided diagnosis (CAD). In addition, to the best of our knowledge, our study is the first demonstration of the DCNN for mammographic CAD application.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133504224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Development of gait assist method for parkinson's disease patients with FOG in walking","authors":"Akira Uehara, H. Kawamoto, Y. Sankai","doi":"10.1109/SICE.2016.7749257","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749257","url":null,"abstract":"Parkinson's disease (PD) is progressive degenerative neurological disorder by dopamine depletion. Many PD patients have Freezing of Gait (FOG). During FOG, they have difficulty to start to take a step in several seconds to over ten seconds. FOG leads to fall, and lower Quality of life due to get injured and psychological influence by anxious and fear of falling. For symptomatic relief of FOG in daily life, it is necessary to predict automatically FOG and physical intervention by wearable system without environment arrangement and human assistance. The purpose of this study is to develop a wearable gait assist system that predicts FOG episode and performs physical assistance, and to confirm the feasibility of the system through an experiment. Subjects of the system have enough muscle to walk alone in this study. For this reason, the system is single leg mechanism of exoskeleton type and a power unit is installed in a knee joint. Prediction method for FOG episode is to detect reduction of gait cycle (GC) based on peak-to-peak of toe floor reaction force. For the experiment with an able-bodied participant, the participant performed normal gait and walking with short steps. As a result, the developed system detected all of GC reduction and performed the assistance, and participant's knee joint was flexed. Therefore, we confirmed the feasibility of the developed system with the prediction method through the experiment.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"152 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129347846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Training system using ground reaction force of the affected leg","authors":"Ryoichiro Shiraishi, H. Kawamoto, Y. Sankai","doi":"10.1109/SICE.2016.7749206","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749206","url":null,"abstract":"Most patients with a paralyzed leg have difficulty in putting the body weight on the heel when they stand up, as well as diminished motor function of the affected part. This paper proposes a new training system for increasing the heel load of the affected leg in several postures. The developed system integrates Ground Reaction Force (GRF) sensors and a graphical monitor with a motion support device. The GRF sensors measure the load of the heel and toe on both legs and the users can recognize their motor information by looking at the monitor. The users can also hold their standing position and half-rising posture by using the motion support device. A basic experiment confirmed that the GRF sensors have measurement accuracy within 5 % error and measures the GRF of the toe and the heel part separately. We carried out a feasibility experiment in two patients to confirm that the users were able to conduct the training by using the system, and the users increased the GRF of the heel part. All the participants recognized their movement of the affected leg by looking at the monitor, and increased the GRF of the heel part when the visual motor information was provided. Moreover, the orthosis decreased the number of times that spasticity occurred and the results suggest that the training is feasible, and the system is effective in helping patients with the affected leg to conduct the training for increasing the GRF of the heel part.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122056526","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yuta Suzuki, Naoya Toyozumi, Junji Takahashi, Lopez Guillaume, H. Hosaka, K. Itao
{"title":"Wearable individual adapting cooling system using smartphone and heart beat sensor","authors":"Yuta Suzuki, Naoya Toyozumi, Junji Takahashi, Lopez Guillaume, H. Hosaka, K. Itao","doi":"10.1109/SICE.2016.7749266","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749266","url":null,"abstract":"In Japan, comfortable lifestyle and environment realized by abundant electric power broke down after March 11 2011. This has been accompanied by augmentation of heatstroke or hypothermia risks, as well as deterioration of labor productivity. An energy-saving method has been proposed that consist in direct neck cooling or warming to avoid decline of both task productivity and comfort. However, current neck cooling system is stand-alone type, so user cannot perceive running state and control state. In addition, temperature control algorithm using discomfort index do not consider individual sensation differences. To wear the system comfortably in daily life, this study proposes enabling system's running state visualization, monitoring, and control by connecting it to user's smartphone. A smartphone application to monitor system's running state has been developed, and an auto-control method to estimate human comfort sensation in real time using small heart-rate sensor has been proposed and validated.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129167948","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Real-time simultaneously updating a linearized state-space model and pole placement gain","authors":"Pyone Ei Ei Shwe, S. Yamamoto","doi":"10.1109/SICE.2016.7749267","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749267","url":null,"abstract":"Data-driven pole placement proposed by Yamamoto et al. (2015) is extended to treat online update of a linearized model and the pole placement gain for a nonlinear system. The data-driven pole placement method can simultaneously derive the pole placement state feedback gain and the state-space model directly from a pair of state and input measurements of a controlled discrete-time system. In this study, we investigate the capability of the real-time data-driven pole placement method by numerical simulations.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122431729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xiaoyong Zhang, N. Homma, K. Ichiji, N. Sugita, Y. Takai, M. Yoshizawa
{"title":"A real-time homography-based tracking method for tracking deformable tumor motion in fluoroscopy","authors":"Xiaoyong Zhang, N. Homma, K. Ichiji, N. Sugita, Y. Takai, M. Yoshizawa","doi":"10.1109/SICE.2016.7749183","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749183","url":null,"abstract":"In radiation therapy, respiration-induced tumor motion significantly limits the efficiency of the radiation delivery, and brings potential risk to healthy organs and tissues. In order to deliver a sufficient high-dose radiation in adaptive with the tumor motion, a kilo-voltage (kV) X-ray fluoroscopy imaging system has been used to monitor the tumor motion in real-time during the treatment. In this paper, we present a fast and robust tracking algorithm to track deformable lung tumor motion in the kV fluoroscopic image sequence. Given a kV fluoroscopy, the tumor motion is represented by a nonlinear homographic transformation of a pre-defined tumor template. The homographic transformation is then estimated by minimizing a sum-of-squared-difference (SSD) between the template image and the observed image. To improve the computational efficiency, an efficient second-order minimization method is employed to solve the problem of SSD minimization. The experimental results conducted on clinical kV fluoroscopies demonstrated that the proposed method is capable of tracking the tumor motion in real-time and its performance is superior to conventional tracking methods in terms of tracking accuracy and computational cost.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122494450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"On SDRE method using by coordinate transformation for an obstacle avoidance problem","authors":"T. Tsuzuki","doi":"10.1109/SICE.2016.7749250","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749250","url":null,"abstract":"As is well known, the state dependence Riccati equation (SDRE) method may compute a controller for a nonlinear optimal regulator problem. We have shown a modified SDRE method and the effectiveness for a simple obstacle avoidance problem. The modified method is similar to the conventional SDRE method after applying coordinate transformation. This paper studies the relationship between the modified SDRE method and the conventional SDRE method after applying coordinate transformation.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123704664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Imitative motion generation for smart device interlocked robot partner based on neuro-genetic approach","authors":"T. Obo, N. Kubota","doi":"10.1109/SICE.2016.7749258","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749258","url":null,"abstract":"This paper proposes a method of imitative motion generation for smart device interlocked robot partner. Imitation is an essential ability for robot partners to provide human like communication. In this study, we have developed a robot simulation for testing human-like motion patterns on a various types of body structure. Moreover, we apply a neuro-genetic approach to generate imitative motions for the robot partner. In the experiment, we show that the smart device interlocked robot partner can perform imitative motion corresponding to human motion measured by 3D image sensor, regardless of the bodily structure differences.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128107223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Model based development of an integrated control of front steering and torque vectoring differential gear system","authors":"Y. Hirano","doi":"10.1109/SICE.2016.7749235","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749235","url":null,"abstract":"For future low carbon mobility society, new-type electric vehicles are developed actively in recent period. Those new EVs have integrated power unit which take place of conventional engine, transmission and differential gear components. Additionally it is rather easy to integrate torque vectoring function to those power units using gear sets to control torque distribution between left wheel and right wheel. In this paper, model-based development of an integrated control of front steering angle and torque vectoring differential (TVD) gear system is described. A new control logic was developed using model matching control to let the vehicle yaw rate and vehicle slip angle follow the desired dynamics. Simulation results using a single track model of vehicle dynamics are shown to prove the efficacy of the proposed control. Though, full vehicle model considering all of vehicle dynamics and drive train motion using Modelica clarified the problem of this method in actual cases. Finally modified control was developed and confirmed by both single track model and full vehicle model.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126588542","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Energy management optimization in consideration of battery deterioration for commuter plug-in hybrid electric vehicle","authors":"Fuguo Xu, Xiaohong Jiao, M. Sasaki, Yuying Wang","doi":"10.1109/SICE.2016.7749262","DOIUrl":"https://doi.org/10.1109/SICE.2016.7749262","url":null,"abstract":"In view of battery deterioration and cost of fuel and electricity, this paper mainly focuses on how to distribute power between the engine and electric equipment for commuter plug-in hybrid electric vehicles(PHEVs) in order to improve the total economy in driving process. A battery lifetime model deriving from models of battery cycle degradation and calendar degradation is first given so that this lifetime model can be applied to energy management optimization directly. Then the Pareto energy optimization problem considering battery wear model is converted into a single-objective optimal control problem. Later a stochastic dynamic programming(SDP) approach is utilized to solve this optimal control problem. Simulation results show that the proposed strategy would effectively benefit vehicle economy by reducing total cost significantly.","PeriodicalId":167635,"journal":{"name":"2016 55th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128193647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}