Yuichi Motai, Ph.D.

Associate Professor, Department of Electrical and Computer Engineering

  • Engineering West Hall, Room 212, Richmond VA UNITED STATES

Associate Professor of the Department of Electrical and Computer Engineering of Virginia Commonwealth University.



Yuichi Motai, Ph.D. is currently an associate professor of Electrical and Computer Engineering at the Virginia Commonwealth University, Richmond, VA, USA after having moved from the University of Vermont. He had a visiting appointment in Radiology Department at Harvard Medical School, MA, USA. He was born in Japan, studied in both Japan and the USA, and completed his Ph.D. with the Robot Vision Laboratory in the School of Electrical and Computer Engineering, Purdue University, West Lafayette, Indiana in 2002. His first laboratory work was in Biomedical Instrumentation Laboratory at Keio University in 1990-1991, where he wrote his bachelor thesis on a psycho-physiological experiment, color matching between computer display and actual object. He wrote his master thesis on 3D object reconstruction by depth from focus using a single camera at Image Informatics laboratory at Kyoto University in 1991-1993. He then spent 4 years as a tenured research scientist in an industrial laboratory called Intelligent Systems Lab in the image processing division where he built an infrared image sensing systems for detecting human behavior. In 1997, he started working for his Ph.D. in the School of Electrical and Computer Engineering at Purdue University. His thesis was on a 3D robot vision system for acquiring an object model with human-computer interaction framework. He spent more than 3 years on this project funded by Ford Motor Company. He has been selected for more than 10 prestigious fellowships and awards, and has been invited to more than 50 research seminars from the top universities and research companies. He has granted over 10 Ph.D as an adviser, and has published 4 research books, more than 100 journals and conferences papers as a corresponding author. He was ONR SFRP Visiting Summer Senior Faculty, Naval Research Lab, Naval Surface Warfare Center, Dahlgren, VA, AFOSR-AFRL/RI Summer Faculty Fellowship Program (SFFP) at Air Force Research Lab, Rome AFB, NY, and AFRL-ASEE SFFP at the Air Force Research Lab in Hanscom AFB, MA. He was a winner of NSF CAREER Award in the Division of Electrical, Communications, and Cyber System at the Directorate for Engineering, and awarded 10+ peer-reviewed grants from both internal and external sources as the PI. His research interests are in the broad area of sensory intelligence, especially of medical imaging, computer vision, and robotics.

Industry Expertise

Electrical Engineering
Industrial Automation
Computer Software
Health and Wellness
Information Technology and Services
Writing and Editing

Areas of Expertise

System Safety
Intelligent Systems with Adaptive Tracking
Online Classification Methodologies
Medical Imaging
Pattern Recognition
Computer Vision
Biomedical Applications
Sensory Intelligence
Data Analysis
Data Science
Software and Development


Purdue University


Electrical and Computer Engineering


Kyoto University


Applied Systems Science


Keio University


Instrumentation Engineering



  • Virginia Commonwealth University
  • Sigma Xi (Scientific Research Society)
  • IEEE (Institute of Electrical and Electronics Engineers) Senior member
  • ACM (Association for Computing Machinery) Lifetime member
  • ASEE (American Society for Engineering Education)

Media Appearances

Editor's Note

IEEE Intelligent transportation systems magazine  print


Review of the Book Predicting Vehicle Trajectory [Book Review] Christos-Nikolaos E. Anagnostopoulos
IEEE Intelligent Transportation Systems Magazine
Year: 2017, Volume: 9, Issue: 3
Pages: 156 - 157, DOI: 10.1109/MITS.2017.2711255

View More

Research Grants

CAREER: Engineering Data-intensive Prediction and Classification for Medical Testbeds with Nonlinear, Distributed, and Interdisciplinary Approaches

NSF $448000


Research Objectives and Approaches

The objective of this research is to contribute to the interdisciplinary topic on STEM and basic medical science, specifically Patient-Centered Health Informatics Applications, with the newly proposed techniques that stand to benefit from the investigator?s expertise in engineering and from the collaboration with medical experts.

The approach is to study the medical data from several institutions comprehensively with the dynamics of all the datasets in their entirety such as non-linearity with kernel factors, and the network characteristics of whole multiple database distribution, rather than applying traditional techniques of prediction and classification to the very limited number of small medical testbeds.

Intellectual Merit

When the proposed adaptive tracking method is used on soft tissue tumors, radiosurgery systems maintain precise targeting of the tumor by predicting tumor motion using a motion tracking system. The successful development of the proposed dynamic classification method will substantially advance the clinical implementation of cancer screening, promote the early diagnosis of colon cancer, lead to an improved screening rate, and ultimately contribute toward reducing the mortality due to colon cancer.

Broader impacts

The proposed data-intensive solutions can save millions of cancer patients every year. The expected outcomes will be applied to medical problems and benefit society as a whole by enhancing the quality of all our lives, through unprecedented advances in the early diagnosis and treatment of cancer. The advancements in the developed framework will make use of and expand the Nation's cyber infrastructure and high performance computing capability.

View more

CRADA ARL JWS 15-18-01 Omni-Directional Infrared Imagery to Enhance Localization and Tracking Algorithms

U.S. ARMY Research Laboratory $30163


Robot Vision

NIH R01 CA159471-07 Subaward – Washington University

National Institutes of Health $25099


Medical Imaging


Signals and Systems I

Presents the concept of linear continuous-time and discrete-time signals and systems, their classification, and analysis and design using mathematical models. Topics to be covered: the concepts of linear systems and classification of these systems, continuous-time linear systems and differential and difference equations, convolution, frequency domain analysis of systems, Fourier series and Fourier transforms and their application, and continuous-time to discrete-time conversion, and Laplace transformation and Transfer function representation.

Dynamic and Multivariable Systems

This course covers the use of state space methods to model analog and digital linear and nonlinear systems. Emphasis is placed on the student gaining mathematical modeling experience, performing sensitivity and stability analysis and designing compensators to meet systems specifications. Topics treated will include a review of root locus and frequency design methods, linear algebraic equations, state variable equations, state space design and digital control systems (principles and case studies). The students will use complex dynamic systems for analysis and design. The laboratory will consist of modeling and control demonstrations and experiments single-input/single-output and multivariable systems, analysis and simulation using matlab control toolbox and other control software.

Automatic Control

This course covers the design and analysis of linear feedback systems. Emphasis is placed upon the student gaining mathematical modeling experience and performing sensitivity and stability analysis. The use of compensators to meet systems design specifications will be treated. Topics include: an overview and brief history of feedback control, dynamic models, dynamic response, basic properties of feedback, root-locus, frequency response and state space design methods. The laboratory will consist of modeling and control demonstrations and experiments single-input/single-output and multivariable systems, analysis and simulation using matlab/simulink and other control system analysis/design/implementation software.

Show All +

Selected Articles

Heterogeneous data analysis: Online learning for medical-image-based diagnosis

Pattern Recognition


Heterogeneous Data Analysis (HDA) is proposed to address a learning problem of medical image databases of Computed Tomographic Colonography (CTC). The databases are generated from clinical CTC images using a Computer-aided Detection (CAD) system, the goal of which is to aid radiologists' interpretation of CTC images by providing highly accurate, machine-based detection of colonic polyps. We aim to achieve a high detection accuracy in CAD in a clinically realistic context, in which additional CTC cases of new patients are added regularly to an existing database. In this context, the CAD performance can be improved by exploiting the heterogeneity information that is brought into the database through the addition of diverse and disparate patient populations. In the HDA, several quantitative criteria of data compatibility are proposed for efficient management of these online images. After an initial supervised offline learning phase, the proposed online learning method decides whether the online data are heterogeneous or homogeneous. Our previously developed Principal Composite Kernel Feature Analysis (PC-KFA) is applied to the online data, managed with HDA, for iterative construction of a linear subspace of a high-dimensional feature space by maximizing the variance of the non-linearly transformed samples. The experimental results showed that significant improvements in the data compatibility were obtained when the online PC-KFA was used, based on an accuracy measure for long-term sequential online datasets. The computational time is reduced by more than 93% in online training compared with that of offline training.

View more

Intra- and Inter-Fractional Variation Prediction of Lung Tumors Using Fuzzy Deep Learning

IEEE Journal of Translational Engineering in Health and Medicine


Tumor movements should be accurately predicted to improve delivery accuracy and reduce unnecessary radiation exposure to healthy tissue during radiotherapy. The tumor movements pertaining to respiration are divided into intra-fractional variation occurring in a single treatment session and inter-fractional variation arising between different sessions. Most studies of patients' respiration movements deal with intra-fractional variation. Previous studies on inter-fractional variation are hardly mathematized and cannot predict movements well due to inconstant variation. Moreover, the computation time of the prediction should be reduced. To overcome these limitations, we propose a new predictor for intra- and inter-fractional data variation, called intra- and inter-fraction fuzzy deep learning (IIFDL), where FDL, equipped with breathing clustering, predicts the movement accurately and decreases the computation time. Through the experimental results, we validated that the IIFDL improved root-mean-square error (RMSE) by 29.98% and prediction overshoot by 70.93%, compared with existing methods. The results also showed that the IIFDL enhanced the average RMSE and overshoot by 59.73% and 83.27%, respectively. In addition, the average computation time of IIFDL was 1.54 ms for both intra- and inter-fractional variation, which was much smaller than the existing methods. Therefore, the proposed IIFDL might achieve real-time estimation as well as better tracking techniques in radiotherapy.

View more

Trajectory Estimations Using Smartphones

IEEE Transactions on Industrial Electronics


This paper investigates whether the smartphones' built-in sensors can accurately predict future trajectories for a possible implementation in a vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) system. If smartphones could be used, vehicles without the V2V/V2I technology could use them to tap into the V2V/V2I infrastructure and help to populate the gap of vehicles off the V2V/V2I grid. To evaluate this, we set up a dead-reckoning system that uses Kalman filters to predict the future trajectory of a vehicle, information that could be used in a V2V/V2I system to warn drivers if the trajectories of vehicles will intersect at the same time. Then, we use a vehicle with accelerometer, GPS, and speedometer sensors mounted on it and evaluate its accuracy in predicting the future trajectory. Afterward, we place a smartphone securely on the vehicle's dashboard, and we use its internal accelerometer and GPS to feed the same dead reckoning and Kalman filter setup to predict the future trajectory of the vehicle. We end by comparing both results and evaluating if a smartphone can achieve similar accuracy in predicting the future trajectory of a vehicle. Our results show that some smartphones could be used to predict a future position, but the use of their accelerometer sensors introduces some measurements that can be incorrectly interpreted as spatial changes.

View more

Show All +