About Me - Ph.D. Computer Science

jack bobak mortazavi
I am an Assistant Professor in Computer Science & Engineering at Texas A&M University, and a member of the Center for Remote Health Technologies & Systems.

I currently have several openings with my group for highly motivated and driven students in all areas of our group's research including: clinical outcomes research, machine learning, embedded systems design, and internet of things research, to name a few! If you are interested in joining the team, please read through

this page

Prior to joining Texas A&M, I was a postdoctoral associate under the supervision of Prof. Harlan Krumholz at the Center for Outcomes Research and Evaluation (CORE) and Prof. Sahand Negahban of the Department of Statistics. I worked as a graduate student (and earned my Ph.D.) in Computer Science under the supervision of Prof. Majid Sarrafzadeh at the University of California Los Angeles (UCLA) Wireless Health Institute (2014). I earned a B.A. in Applied Mathematics and a B.S. in Electrical Engineering and Computer Science from the Univeristy of California Berkeley (2007).

My research interests include end-to-end research on medical embedded systems and the application of data mining and machine learning algorithms necessary to make personalized, preventative medical treatments possible through advanced health analytics . My background is in embedded systems design, where I studied sensor fusion, reconfigurable architectures and systems, hardware accelerators, and gpu computing. During my Ph.D. I applied data mining and machine learning techniques to these systems to develop a personalized, exercise-level activity-recognition video game with wearable sensors. I am now primarily concerned with the ability to use supervised and unsupervised techniques to learn more about medical prediction and risk-stratification in order to better develop personalized medical systems, prediction models, comparative effectiveness techniques, and combine wearable sensors and other necessary data to make a clinical impact at the system level, provider level, and patient level.

My office:

328A HRBB
College Station, TX 06510
Office: (979) 458-2642

Back to top

Research


Currently, my interdisciplinary work involves bridging of computer science with the medical research realm, particularly in Patient-Centered Outcomes Research. For example, I have worked on projects ranging from Personalized Activity Recognition for exergaming, exercise repetition, rehabilitation and nutrition monitoring (while at UCLA) to Remote Monitoring and Clinical prediction in clinical settings using Clinical Trial Data, Clinical Registry Data, and Clinical Electronic Health Record Data (while at Yale). These projects investigate system design, user adherence and cheating, risk prediction, comparative effectiveness, and a deeper understanding and evaluation of clinical questions and challenges at a patient-centered outcomes focus.

To detailed description of research projects Back to top

Publications

Journals

[j13] Mortazavi, B., Downing, N., Bucholz, E., Dharmarajan, K., Manhapra, A., Li, S., Negahban, S., Krumholz, H. (2015) Analysis of Machine Learning Techniques for Heart Failure Readmissions. Circulation: Cardiovascular Quality and Outcomes (Accepted - In Press) [link] [pdf]

[j12] Hunter, D., Torkelson, J., Bodnar, J., Mortazavi, B., Laurent, T., Deason, J., Thephavongsa, K., Zhong, J. (2015) The Rickettsia Endosymbiont of Ixodes Pacificus Contains All Genes of De Novo Folate Biosynthesis. PLOS One (Accepted - In Press) [link] [pdf]

[j11] Mortazavi, B., Pourhomayoun, M., Lee, S.I., Nyamathi, S., Wu, B., Sarrafzadeh, M. (2015) User-Optimized Activity Recognition for Exergaming. Elsevier Journal of Pervasive and Mobile Computing (Accepted - In Press) [link] [pdf]

[j10] Lee, S.I., Park, E., Huang, A., Mortazavi, B., Garst, J.H., Espinal, M., Siero, T., Pollack, S., Afridi, M., Daneshvar, M., Ghias, S., Lu, D.C., Sarrafzadeh, M. (2015) Objectively Quantifying Walking Ability in Degenerative Spinal Disorder Patients using Sensor Equipped Smart Shoes. Medical Engineering & Physics (Med Eng Phys) (Under Review - Second Revision) [link] [pdf]

[j9] Lee, S.I., Li, C., Hoffman, H., Lu, D., Getachew, R., Mortazavi, B., Garst, J., Espinal, M, Razaghy, Ghalehsari, N., Paak, B., Chavam, A., Afridi, M., Ostowari, A., Ghasemzadeh, H., Lu, D., Sarrafzadeh, M. (2014) Quantitative Assessment of Hand Motor Function in Cervical Spinal Disorder Patients Using Target Tracking Tests. Journal of Rehabilitation Research and Development (J-RRD) (Accepted - In Press) [link] [pdf]

[j8] Kalantarian, H., Alshuarafa, N., Sideris, C., Le, T., Mortazavi, B., Sarrafzadeh, M. (2015) A Comparison of Piezoelectric-Based Inertial Sensing and Audio-Based Detection of Swallows. Elsivier Journal of Pervasive and Mobile Computing (Under Review) [link] [pdf]

[j7] Mortazavi, B., Nemati, E., VanderWall, K., Flores-Rodriguez, H., Cai, J., Lucier, J., Naeim, A., Sarrafzadeh, M. (2015) Can Smartwatches Replace Smartphones for Posture Tracking? Sensors vol. 15, no. 10, pp. 26783-26800 [link] [pdf]

[j6] Woodbridge, J., Mortazavi, B., Bui, A.A.T., Sarrafzadeh, M. (2015) Improving Biomedical Signal Search Results in Big Data Case-Based Reasoning Environments. Elsevier Journal of Pervasive and Mobile Computing (Accepted - In Press) [link] [pdf]

[j5] Lee, S.I., Mortazavi, B., Hoffman, H., Lu, D., Paak, B., Garst, J., Razaghy, M., Lu, D., Sarrafzadeh, M. (2014) A Prediction Model for Functional Outcomes in Spinal Cord Injured Patients Using Gaussian Process Regression. IEEE Journal of Biomedical and Health Informatics (J-BHI) (Accepted - In Press) [link] [pdf]

[j4] Mortazavi, B., Pourhomayoun, M., Ghasemzadeh, H., Jafari, R., Roberts, C.K., Sarrafzadeh, M. (2014) Context-Aware Data Processing to Enhance Quality Measurements in Wireless Health Systems: An Application to MET Calculation of Exergaming Actions. IEEE Internet of Things Journal (J-IOT) vol. 2, no. 1, pp. 84-93 [link] [pdf]

[j3] Alshurafa, N., Xu, W., Liu, J., Huang, M.C., Mortazavi, B., Roberts, C.K., Sarrafzadeh, M. (2013) Designing a Robust Activity Recognition Framework for Health and Exergaming using Wearable Sensors. IEEE Journal of Biomedical and Health Informatics (J-BHI) vol. 18, no. 5., pp 1636-1646 [link] [pdf]

[j2] Mortazavi, B., Nyamathy, S., Lee, S.I., Wilkerson, T., Ghasemzadeh, H., Sarrafzadeh, M. (2014) Near-Realistic Mobile Exergames with Wireless Wearable Sensors. IEEE Journal of Biomedical and Health Informatics (J-BHI) vol. 18, no. 2, pp. 449-456 (March Feature Article) [link] [pdf]

[j1] Lee, S.I., Ghasemzadeh, H., Mortazavi, B., Sarrafzadeh, M. (2013) Pervasive Assiessment of Motor Function: A Lightweight Grip Strength Tracking System. IEEE Journal of Biomedical and Health Informatics (J-BHI) vol. 17, no. 6, pp. 1023-1030 [link] [pdf]

Conferences


[c19] Pourhomayoun, M., Nemati, E., Mortazavi, B., Sarrafzadeh, M. (2015) Context-Aware Data Analytics for Activity Recognition. In Proceedings of the Fourth International Conference on Data Analytics (DATA ANALYTICS 2015) (Accepted - In Press) [link] [pdf] (Best Paper Award)

[c18] Alinia, P., Saeedi, R, Mortazavi, B., Ghasemzadeh, H. (2015) Impact of Sensor Misplacement on Estimating Metabolic Equivalent of Task with Wearables. In Proceedings of the 2015 IEEE Conference on Wearable and Implantable Body Sensor Networks (BSN 2015) (Accepted - In Press) [link] [pdf]

[c17] Mortazavi, B., Pourhomayou, M., Nyamathi, S., Wu, B., Lee, S.I., Sarrafzadeh, M. (2015) Multiple Model Recognition for Near-Realistic Exergaming. In Proceedings of the 2015 IEEE Interantional Conference on Pervasive Computing and Communication (PerCom) St. Louis, Mo., Mar. 2015. [link] [pdf]

[c16] Mortazavi, B., Pourhomayoun, M., Alshurafa, N., Chronley, M., Lee, S.I., Roberts, C.K., Sarrafzadeh, M. (2014) Support Vector Regression for Estimating the Metabolic Equivalent of Task of Exergaming Actions. In Proceedings the Conference on Healthcare Innovations and Point-of-Care Technologies. (HIC) Seattle, Wa., Oct. 2014. [link] [pdf]

[c15] Pourhomayoun, M., Alshurafa, N., Mortazavi, B., Ghasemzadeh, H., Sarrafzadeh, M. (2014) Multiple Model Analytics for Adverse Event Prediction in Remote Health Monitoring Systems. In Proceedings of the Conference on Healthcare Innovations and Point-of-Care Technologies. (HIC) Seattle, Wa., Oct. 2014 [link] [pdf]

[c14] Mortazavi, B., Lee, S.I., Sarrafzadeh, M. (2014) User-Centric Exergaming with Fine-Grain Activity Recognition: A Dynamic Optimization Approach. In Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing: Adjunct Publication (Ubicomp ’14 Adjunct) Seattle, Wa., Sept. 2014, pp. 1233-1240 [link] [pdf]

[c13] Mortazavi, B., Pourhomayoun, M., Alsheikh, G., Alshurafa, N., Lee, S.I., Sarafzadeh, M. (2014) Determining the Single Best Axis for Exercise Repetition Recognition and Counting in SmartWatches. In Proceedings of the 11th IEEE International Conference on Wearable and Implantable Body Sensor Networks (BSN 2014) Zurich, Switzerland, June 2014 [link] [pdf]

[c12] Alshurafa, N., Pourhomayoun, M., Nyamathi, S., Bao, L., Mortazavi, B., Eastwood, J., Sarrafzadeh, M. (2014) Anti-Cheating: Detecting Self-Inflicted and Impersonator Cheaters for Remote Health Monitoring Systems with Wearable Sensors. In Proceedings of the 11th IEEE International Conference on Wearable and Implantable Body Sensor Networks (BSN 2014) Zurich, Switzerland, June 2014 [link] [pdf]

[c11] Lee, S.I., Ghasemzadeh, H., Mortazavi, B., Lan, M., Ong, M., Sarrafzadeh, M. (2013) Remote Monitoring Systems: What Impact Can Data Analytics Have on Cost? In Proceedings of Wireless Health 2013(WH 13) Baltimore, MD, Oct. 2013. pp 4:1-4:8 [link] [pdf]

[c10] Mortazavi, B., Alsharufa, N., Lee, S.I., Lan, M., Chronley, M., Roberts, C.K., Sarrafzadeh, M. (2013) MET Calculations from On-Body Accelerometers for Exergaming Movements. In Proceedings of the 10th IEEE International Conference on Wearable and Implantable Body Sensor Networks (BSN 2013) Cambridge, MA, May 2013. pp 1-6 [link] [pdf]

[c9] Moazeni, M., Mortazavi, B., Sarrafzadeh, M. (2013) High Performance Multi-Dimensional Signal Search with Applications in Remote Medical Monitoring. In Proceedings of the 10th IEEE International Conference on Wearable and Implantable Body Sensor Networks (BSN 2013) Cambridge, MA, May 2013. pp 1-6 [link] [pdf]

[c8] Alsharufa, N., Xu, W., Liu, J., Huang, M.C., Mortazavi, B., Roberts, C.K., Sarrafzadeh, M. (2013) Robust Human Intensity-Varying Activity Recognition using Stochastic Approximation in Wearable Sensors. In Proceedings of the 10th IEEE International Conference on Wearable and Implantable Body Sensor Networks (BSN 2013) Cambridge, MA, May 2013. pp 1-6 [link] [pdf]

[c7] Lee, S.I., Ghasemzadeh, H., Mortazavi, B., Yew, A., Getachew, R., Razaghy, M., Ghalehsari, N., Paak, B., Garst, J., Espinal, M., Kimball, J., Lu, D., Sarrafzadeh, M. (2013) Objective Assessment of Overexcited Hand Movements Using a Lightweight Sensory Device. In Proceedings of the 10th IEEE International Conference on Wearable and Implantable Body Sensor Networks (BSN 2013) Cambridge, MA, May 2013. pp 1-6 [link] [pdf]

[c6] Woodbridge, J., Mortazavi, B., Sarrafzadeh, M., Bui, A. (2012) A Monte Carlo Approach to Biomedical Time Series Search. In Proceedings of the IEEE International Conference on Bioinformatics and Biomedicine (BIBM 2012) Philadelphia, PA, pp 1-6 [link] [pdf]

[c5] Woodbrige, J., Mortazavi, B., Bui, A., Sarrafzadeh, M. (2012) Aggregated Indexing of Biomedical Time-Series Data. In Proceedings of the 2nd IEEE Conference on Healthcare Informatics, Imaging, and Systems Biology (HISB 2012) San Diego, CA, Sep. 2012. pp 23-30 [link] [pdf]

[c4] Suh, D. M. K., Woodbridge, J., Moin, T., Lan, M., Alshurafa, N., Samy, L., Mortazavi, B., Ghasemzadeh, H., Bui, A., Ahmadi, S., Sarrafzadeh, M. (2012) Dynamic Task Optimization in Remote Diabetes Monitoring Systems. In Proceedings of the 2nd IEEE Conference on Healthcare Informatics, Imaging, and Systems Biology (HISB 2012) San Diego, CA, Sep. 2012. pp 3-11 [link] [pdf]

[c3] Woodbridge, J., Mortazavi, B., Bui, A., Sarrafzadeh, M. (2012) High Performance Biomedical Time- Series Indexing Using Salient Segmentation. In Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC ’12) San Diego, CA, Aug. 2012. pp 5086-5089 [link] [pdf]

[c2] Mortazavi, B., Chu, K.C., Li, X., Tai, J., Kotekar, S., Sarrafzadeh, M. (2012) Near-Realistic Motion Video Games with Enforced Activity. Proceedings of the 9th International Conference on Wearable and Implantable Body Sensor Networks (BSN 2012) London, England, May 2012. pp 28-33 [link] [pdf]

[c1] Mortazavi, B., Hagopian, H., Woodbridge, J., Yadegar, B., Sarrafzadeh, M. (2011) A Body-Wearable Sensor System for Designing Physically Interactive Video Games. Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES 2011) Rome, Italy, Jan. 2011. pp. 62-69 [link] [pdf]

Patents

[p1] (Pending) PCT/US12/21906 Sarrafzadeh, M., Hagopian, H., Mortazavi, J.B., and Garcia, J. System and Methods of Detecting and Reproducing Motions for Video Games [link]

[p2] (Provisional) UCLA Case 2012-106 Sarrafzadeh, M., Mortazavi, J.B., Li, X., Chu, K., Near-Realistic Sports Motion and Analysis [link]

[p3] (Pending) PCT/US20113/30045 Sarrafzadeh, M., Lee, S.I., Mortazavi, J.B., Method and Apparatus for Mobile Rehabilitiation Exergaming [link]

Posters, Demos, Abstracts, etc.

[a2] Naeim, A., VanderWall, K., Lucier, J., Sarrafzadeh, M., Tan, H.J.R., Mortazavi, B., Nemati, E. (2014) Accurate Classification of Performance Status in Elderly Patients: Design, Validation, and Implementation of a Remote Patient Activity Monitoring Device. Journal of Geriatric Oncology. 2014-10;5:S48

[a1] Mortazavi, B., Sarrafzadeh, M. (2012) Soccer Exergaming: A Platform for Energy Expenditure Video Games. 9th International Conference on Wearable and Implantable Body Sensor Networks (BSN 2012) London, England, May 2012. (Best Demo Award) Back to top

Teaching

Lecturer Courses

Teaching Assistant Courses

Back to top

Misc.

Awards

[2] Finalist - funRehab: A Mobile Rehabilitation Gaming System UCLA Business of Science Center Venture Competition June, 2012

[1] Best Demonstration Award - 9th International Conference on Wearable and Implantable Body Sensor Networks (BSN 2012) May, 2012.

Invited Talks

[3] Mortazavi, B.(2015) Challenges in Developing Methods for a Clinical Audience - Tutorial Workshop on Machine Learning and Data Mining with a Focus on Human Studies at the 2015 Wireless Health Conference , NIH, Bethesda, MD, October 2015.

[2] Mortazavi, B., Sarrafzadeh, M. (2013) Real-Time Activity Detection in Sports for Exergaming - Workshop on Pervasive Sensing in Sports and Extreme Environments at 2013 IEEE Body Sensor Networks Conference Cambridge, MA, May 2013.

[1] Mortazavi, B., Sarrafzadeh, M. (2011) Hackers Can Kill You - TakeDownCon 2011 Dallas, Tx

Academic Services

Journal Editor:
T.P.C. Member:
Journal Reviewer:
Conference Reviewer:
Back to top`

Detailed Research Projects

My research investigates all aspects of patient-centered end-to-end medical embedded systems, from various sensors and wearable monitoring techniques to algorithms, data mining, and machine learning to process the data generated for different medical applications. These applications range from monitoring information for healthy individuals to adverse event detection of patients in critical care environments. My work, in particular, investigates the processing of real-time time-series processing, from activity recognition to medical event prediction. This includes supervised and unsupervised techniques to better cluster and identify patient types, evaluate the balance between historical data and real-time data, and classification and risk-stratification.

My work is a bridging of statistical inference/computer science and system design techniques with the medical research realm. As these clustering, classification and risk-stratification techniques are improved upon, I look to investigate the user-specific nature of such information, from the data processing to patient identification and adverse event prediction. Ultimately, the end-to-end monitoring systems for preventative medicine need to be personalized to provide the strongest possible care and information to the patient, the clinician, and the clinical care environment.

Machine Learning with Clinical Data: Patient-Centered Risk and Prediction in Clinical Data, Registry Data, and EMR Data

My postdoctoral position in the medical school investigates the application of data analytic techniques to better identify patient types, heterogeneous illness identification, and risk-stratification and prediction on various data sets and on numerous clinical problems. This varies from pristine clinical trial data sets, to comprehensive national cross-sectional data registries, to the real-time data available from the electronic medical records system at the Yale New Haven Hospital system. Data ranges from standard, baseline, cross-sectional data to streaming real-time information at the patient level. This streaming can vary from daily telephone calls, monthly physical exams, and minute-by-minute ICU/EMR data. This project is a componenet of the Big Data To Knowledge innitiative here at Yale to bridge the data sciences with clinical practice and research. We are investigating the implementation of novel techniques to identify sub-phenotypes of patients/disease types to better risk-stratify patients and identify/address adverse events. Our projects investigate risk prediction, comparative effectiveness, and a deeper understanding and evaluation of existing national models on adverse events in registry data and electronic medical record data.


Realistic-Motion Activity Recognition

I am also interested in the application of user-specific models for wearable sensors of various applications. While the medical and heart failure classification project looks at ill patients and readmissions information, patient-centered wearable systems also allow for new systems for healthy individuals. My work investigates systems and algorithms for real-time and near-realistic activity recognition for user-centered sports training, exercise-based video game (exergaming), and the classification of large, multi-class environments of similar movements in a fine-grain, non-cyclical manner. Further, we have investigated, through the use of a pilot clinical trial, the actual caloric expenditure of such a system, developing various regression techniques to determine the exercise effectiveness of the movements.

I am also interested in the application of user-specific models for wearable sensors of various applications. While the medical and heart failure classification project looks at ill patients and readmissions information, patient-centered wearable systems also allow for new systems for healthy individuals. My work investigates systems and algorithms for real-time and near-realistic activity recognition for user-centered sports training, exercise-based video game (exergaming), and the classification of large, multi-class environments of similar movements in a fine-grain, non-cyclical manner. Further, we have investigated, through the use of a pilot clinical trial, the actual caloric expenditure of such a system, developing various regression techniques to determine the exercise effectiveness of the movements.

Rehabilitation Devices, Remote Monitoring of Patients, and Exercise Monitoring

These movements can range from multiple sensors and a gaming environment to exercise repetition counting from a wrist-wearable sensor in a gym environment. Further, we have investigated the use of custom sensors, smart phones, and smart watches in tracking people from healthy individuals to patients in a clinical setting, such as elderly oncology patients and patients in a large heart failure clinical trial, which was a project named Wanda. Further, we looked at trade offs between false positive and false negative rates in such systems, alarm fatigue, and nursing costs in implementing such remote monitoring systems.

Biomedical Signal Search

Another aspect of evaluating medical embedded systems and analytics is information gain in finding similar patients. We looked at ECG patterns and developed a signal search method to find similar heart beat patterns to categorize patients' signals. This form of signal search allows for finding similar but not exact matches in a database without resulting in a significant pruning time for erroneous matches.

Nutrition Monitoring

Our wearable sensors work also extends to nutrition monitoring where we developed a necklace that can detect the types of food you ingest while swallowing, in order to better estimate your caloric intake. While most systems focus on determining caloric expenditure, the need to better automate intake is high.

Back to top