Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mert Sevil has a total of 88 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Detection and Classification of Unannounced Physical Activities and Acute Psychological Stress Events for Interventions in Diabetes Treatment

    by Mohammad Reza Askari, Mahmoud Abdel-Latif, Mudassir Rashid, Mert Sevil, Ali Cinar

    Detection and classification of acute psychological stress (APS) and physical activity (PA) in daily lives of people with chronic diseases can provide precision medicine for the treatment of chronic conditions such as diabetes. This study investigates the classification of different types of APS and PA, along with their concurrent occurrences, using the same subset of feature maps via physiological variables measured by a wristband device. Random convolutional kernel transformation is used to extract a large number of feature maps from the biosignals measured by a wristband device (blood volume pulse, galvanic skin response, skin temperature, and 3D accelerometer signals). Three different feature selection techniques (principal component analysis, partial least squares–discriminant analysis (PLS-DA), and sequential forward selection) as well as four approaches for addressing imbalanced sizes of classes (upsampling, downsampling, adaptive synthetic sampling (ADASYN), and weighted training) are evaluated for maximizing detection and classification accuracy. A long short-term memory recurrent neural network model is trained to estimate PA (sedentary state, treadmill run, stationary bike) and APS (non-stress, emotional anxiety stress, mental stress) from wristband signals. The balanced accuracy scores for various combinations of data balancing and feature selection techniques range between 96.82% and 99.99%. The combination of PLS–DA for feature selection and ADASYN for data balancing provide the best overall performance. The detection and classification of APS and PA types along with their concurrent occurrences can provide precision medicine approaches for the treatment of diabetes.


  • Detection of Meals and Physical Activity Events From Free-Living Data of People With Diabetes

    by Mohammad Reza Askari, Mudassir Rashid, Xiaoyu Sun, Mert Sevil, Andrew Shahidehpour, Keigo Kawaji, Ali Cinar

    Background: Predicting carbohydrate intake and physical activity in people with diabetes is crucial for improving blood glucose concentration regulation. Patterns of individual behavior can be detected from historical free-living data to predict meal and exercise times. Data collected in free-living may have missing values and forgotten manual entries. While machine learning (ML) can capture meal and exercise times, missing values, noise, and errors in data can reduce the accuracy of ML algorithms. Methods: Two recurrent neural networks (RNNs) are developed with original and imputed data sets to assess detection accuracy of meal and exercise events. Continuous glucose monitoring (CGM) data, insulin infused from pump data, and manual meal and exercise entries from free-living data are used to predict meals, exercise, and their concurrent occurrence. They contain missing values of various lengths in time, noise, and outliers. Results: The accuracy of RNN models range from 89.9% to 95.7% for identifying the state of event (meal, exercise, both, or neither) for various users. “No meal or exercise” state is determined with 94.58% accuracy by using the best RNN (long short-term memory [LSTM] with 1D Convolution). Detection accuracy with this RNN is 98.05% for meals, 93.42% for exercise, and 55.56% for concurrent meal-exercise events. Conclusions: The meal and exercise times detected by the RNN models can be used to warn people for entering meal and exercise information to hybrid closed-loop automated insulin delivery systems. Reliable accuracy for event detection necessitates powerful ML and large data sets. The use of additional sensors and algorithms for detecting these events and their characteristics provides a more accurate alternative.


  • Open Access

    Meal and Physical Activity Detection from Free-Living Data for Discovering Disturbance Patterns of Glucose Levels in People with Diabetes

    by Mohammad Reza Askari, Mudassir Rashid, Xiaoyu Sun, Mert Sevil, Andrew Shahidehpour, Keigo Kawaji, Ali Cinar

    Objective: The interpretation of time series data collected in free-living has gained importance in chronic disease management. Some data are collected objectively from sensors and some are estimated and entered by the individual. In type 1 diabetes (T1D), blood glucose concentration (BGC) data measured by continuous glucose monitoring (CGM) systems and insulin doses administered can be used to detect the occurrences of meals and physical activities and generate the personal daily living patterns for use in automated insulin delivery (AID). Methods: Two challenges in time-series data collected in daily living are addressed: data quality improvement and the detection of unannounced disturbances of BGC. CGM data have missing values for varying periods of time and outliers. People may neglect reporting their meal and physical activity information. In this work, novel methods for preprocessing real-world data collected from people with T1D and the detection of meal and exercise events are presented. Four recurrent neural network (RNN) models are investigated to detect the occurrences of meals and physical activities disjointly or concurrently. Results: RNNs with long short-term memory (LSTM) with 1D convolution layers and bidirectional LSTM with 1D convolution layers have average accuracy scores of 92.32% and 92.29%, and outperform other RNN models. The F1 scores for each individual range from 96.06% to 91.41% for these two RNNs. Conclusions: RNNs with LSTM and 1D convolution layers and bidirectional LSTM with 1D convolution layers provide accurate personalized information about the daily routines of individuals. Significance: Capturing daily behavior patterns enables more accurate future BGC predictions in AID systems and improves BGC regulation.


  • Physical Activity and Psychological Stress Detection and Assessment of Their Effects on Glucose Concentration Predictions in Diabetes Management

    by Mert Sevil, Mudassir Rashid, Iman Hajizadeh, Minsun Park, Laurie Quinn, Ali Cinar

    Objective: Continuous glucose monitoring (CGM) enables prediction of the future glucose concentration (GC) trajectory for making informed diabetes management decisions. The glucose concentration values are affected by various physiological and metabolic variations, such as physical activity (PA) and acute psychological stress (APS), in addition to meals and insulin. In this work, we extend our adaptive glucose modeling framework to incorporate the effects of PA and APS on the GC predictions. Methods: A wristband conducive of use by free-living ambulatory people is used. The measured physiological variables are analyzed to generate new quantifiable input features for PA and APS. Machine learning techniques estimate the type and intensity of the PA and APS when they occur individually and concurrently. Variables quantifying the characteristics of both PA and APS are integrated as exogenous inputs in an adaptive system identification technique for enhancing the accuracy of GC predictions. Data from clinical experiments illustrate the improvement in GC prediction accuracy. Results: The average mean absolute error (MAE) of one-hour-ahead GC predictions with testing data decreases from 35.1 to 31.9 mg/dL (p-value = 0.01) with the inclusion of PA information, and it decreases from 16.9 to 14.2 mg/dL (p-value = 0.006) with the inclusion of PA and APS information. Conclusion: The first-ever glucose prediction model is developed that incorporates measures of physical activity and acute psychological stress to improve GC prediction accuracy. Significance: Modeling the effects of physical activity and acute psychological stress on glucose concentration values will improve diabetes management and enable informed meal, activity and insulin dosing decisions.


  • Discrimination of simultaneous psychological and physical stressors using wristband biosignals

    by Mert Sevil, Mudassir Rashid, Iman Hajizadeh, Mohammad Reza Askari, Nicole Hobbs, Rachel Brandt, Minsun Park, Laurie Quinn, Ali Cinar


  • Open Access

    Detection and Characterization of Physical Activity and Psychological Stress from Wristband Data

    by Mert Sevil, Mudassir Rashid, Mohammad Reza Askari, Zacharie Maloney, Iman Hajizadeh, Ali Cinar

    Wearable devices continuously measure multiple physiological variables to inform users of health and behavior indicators. The computed health indicators must rely on informative signals obtained by processing the raw physiological variables with powerful noise- and artifacts-filtering algorithms. In this study, we aimed to elucidate the effects of signal processing techniques on the accuracy of detecting and discriminating physical activity (PA) and acute psychological stress (APS) using physiological measurements (blood volume pulse, heart rate, skin temperature, galvanic skin response, and accelerometer) collected from a wristband. Data from 207 experiments involving 24 subjects were used to develop signal processing, feature extraction, and machine learning (ML) algorithms that can detect and discriminate PA and APS when they occur individually or concurrently, classify different types of PA and APS, and estimate energy expenditure (EE). Training data were used to generate feature variables from the physiological variables and develop ML models (naïve Bayes, decision tree, k-nearest neighbor, linear discriminant, ensemble learning, and support vector machine). Results from an independent labeled testing data set demonstrate that PA was detected and classified with an accuracy of 99.3%, and APS was detected and classified with an accuracy of 92.7%, whereas the simultaneous occurrences of both PA and APS were detected and classified with an accuracy of 89.9% (relative to actual class labels), and EE was estimated with a low mean absolute error of 0.02 metabolic equivalent of task (MET).The data filtering and adaptive noise cancellation techniques used to mitigate the effects of noise and artifacts on the classification results increased the detection and discrimination accuracy by 0.7% and 3.0% for PA and APS, respectively, and by 18% for EE estimation. The results demonstrate the physiological measurements from wristband devices are susceptible to noise and artifacts, and elucidate the effects of signal processing and feature extraction on the accuracy of detection, classification, and estimation of PA and APS.


  • Determining Physical Activity Characteristics From Wristband Data for Use in Automated Insulin Delivery Systems

    by Mert Sevil, Mudassir Rashid, Zacharie Maloney, Iman Hajizadeh, Sediqeh Samadi, Mohammad Reza Askari, Nicole Hobbs, Rachel Brandt, Minsun Park, Laurie Quinn, et al.

    Algorithms that can determine the type of physical activity (PA) and quantify the intensity can allow precision medicine approaches, such as automated insulin delivery systems that modulate insulin administration in response to PA. In this work, data from a multi-sensor wristband is used to design classifiers to distinguish among five different physical states (PS) (resting, activities of daily living, running, biking, and resistance training), and to develop models to estimate the energy expenditure (EE) of the PA for diabetes therapy. The data collected are filtered, features are extracted from the reconciled signals, and the extracted features are used by machine learning algorithms, including deep-learning techniques, to obtain accurate PS classification and EE estimation. The various machine learning techniques have different success rates ranging from 75.7% to 94.8% in classifying the five different PS. The deep neural network model with long short-term memory has 94.8% classification accuracy. We achieved 0.5 MET (Metabolic Equivalent of Task) root-mean-square error for EE estimation accuracy, relative to indirect calorimetry with randomly selected testing data (10% of collected data). We also demonstrate a 5% improvement in PS classification accuracy and a 0.34 MET decrease in the mean absolute error when using multi-sensor approach relative to using only accelerometer data.


  • 704-P: Influence of Acute Stress on Heart Rate and Glycemia during Exercise

    by Nicole Hobbs, Minsun Park, Minseung Chu, Youngkwan Song, Laurie T. Quinn, Mert Sevil, Ali Cinar

    People with type 1 diabetes (T1D) have personal strategies for maintaining euglycemia while exercising which generally serves them well during training. Their strategy often yields wildly differing results during a competition. The influence of competition stress (notably, epinephrine causing prompt elevations in glucose) and high-intensity and/or long-duration efforts can yield dysglycemia. Two different studies were conducted to observe this phenomenon. In a small study of subjects with T1D completing athletic competitions, perceived stress level as reported by the individual is insufficient for determination impact upon glycemia. An elevated heart rate during the competition over what was observed at the same running pace in training was a better indicator of an increase in glycemia. To recreate an athletic stress scenario in a clinical setting, 13 subjects with T1D completed a low intensity treadmill protocol for 3 sessions: non-stress, verbally reported mental math with an audience, and stressful video games to observe the influence on glycemia and heart rate. The treadmill protocol, prior meal, and insulin doses were matched for each individual for the 3 sessions. When compared to the non-stress treadmill session, an elevated average heart rate was observed in 10 of the 13 video game sessions and an elevated average glucose was observed in 9 of the 13 video game sessions. In the mental math sessions, an average heart rate elevated heart rate was observed in 6 of the 13 sessions with 7 of 13 having a higher average glucose compared to the non-stress session. While individual response to acute stress yields a high degree of intersubject variability, the video game sessions may more closely match the competition stress experienced and physiological response to stress observed during an athletic competition. The cause of the stressor may be critical in detection of stress and predicting glycemic response for the management of T1D. Disclosure: N. Hobbs: None. M. Park: None. M. Chu: None. Y. Song: None. L.T. Quinn: None. M. Sevil: None. A. Cinar: Research Support; Self; Dexcom, Inc., JDRF. Stock/Shareholder; Self; Dexcom, Inc., Eli Lilly and Company, Insulet Corporation, Medtronic, Novo Nordisk A/S, Tandem Diabetes Care. Funding: JDRF (A18-0036-001); National Institutes of Health (DP3DK101075, F31DK116524)

  • 48-LB: The Effect of Acute Psychosocial Stress in Adults with Type 1 Diabetes

    by Minsun Park, Youngkwan Song, Minseung Chu, Mert Sevil, Nicole Hobbs, Lisa Sharp, Ali Cinar, Laurie T. Quinn

    The effects of acute psychological stress, heart rate variability (HRV), an indicator of autonomic function and glycemic variability (GV) have not been extensively researched in type 1 diabetes (T1D). The aim of this study is to explore the effect of acute psychosocial stress on HRV and GV in people with T1D. Methods: Eleven participants with T1D (33.00 ± 11.25 years) participated in the study. To assess acute mental stress, the subjects 1) participated in the Trier Social Stress Test (TSST); and 2) played competitive video games (Nintendo Switch™). The TSST followed standardized procedures whereby stress was induced asking participants to provide a job talk and arithmetic in front of two judges. Participants played the Mario Kart to go-kart races (10 minutes) and rotated their controllers to untangle the Treasure Chest (10 minutes). Resting HRV and GV were obtained. The subjects wore a continuous glucose monitor and a Zephyr Bioharness™ to record GV and HRV, respectively. Nonparametric, paired samples T-tests were conducted using SPSS 24. Results: During TSST, there were significant differences between the resting and TSST testing period in the sympathetic nervous sympathetic nervous system index (.57 ± 1.33 vs. 1.36 ± 1.40; z = -2.845, respectively; p <.01; and stress index (8.17 ± 3.69 vs. 10.27 ± 5.69; z = -2.578, respectively; p <.05). When subjects were playing video games, there were significant differences in the parasympathetic nervous sympathetic nervous system index (-.09 ± 1.75 vs. -.87 ± 1.05; z = -2.134, respectively; p <.05) compared to a resting period. However, there were no significant differences in GV compared to a resting period in during TSST and playing video game. Conclusion: This study suggests that the TSST induced stress in people with T1D, increased sympathetic activity; however, the TSST did not change glycemic variability. Playing competitive video game did not appear to neither induces stress nor affected glycemic variability. The competitive video game did not cause stress compared to TSST. Disclosure M. Park: None. Y. Song: None. M. Chu: None. M. Sevil: None. N. Hobbs: None. L. Sharp: None. A. Cinar: Research Support; Self; Dexcom, Inc., JDRF. Stock/Shareholder; Self; Dexcom, Inc., Eli Lilly and Company, Insulet Corporation, Medtronic, Novo Nordisk A/S, Tandem Diabetes Care. L.T. Quinn: None. Funding JDRF


  • 43-LB: Glycemic Control Affects Heart-Rate Variability in Adults with Type 1 Diabetes

    by Minsun Park, Minseung Chu, Youngkwan Song, Mert Sevil, Ali Cinar, Laurie T. Quinn, Ulf G. Bronas

    People with type 1 diabetes (T1D) have an increased risk for autonomic dysfunction and those with poorly controlled blood glucose levels have an increased risk for cardiovascular disease. The purpose of this study was to determine heart rate variability (HRV) and glucose variability (GV) parameters in less than optimally controlled patients (Group 2: A1C ≥ 7.0%, n = 6]) compared to well-controlled patients (Group 1: A1C ≤ 6.9% [n = 7]) during steady state treadmill exercise took and rest. Methods: Thirteen T1D participants, (32.42 ± 10.46 years), performed steady state treadmill exercise (40% VO2Peak) for 30 minutes. Prior to exercise, participants rested for 30-minutes for determination of resting HRV and GV. Participants wore continuous glucose monitors and a Zephyr Bioharness™ to determine GV and HRV, respectively. Nonparametric, paired samples T-tests were conducted using SPSS 24. Results: In group 2, there were significant differences between the resting period and treadmill exercise in the (1) root mean score of successive differences (RMSSD) (27.71 ± 13.43 vs. 8.01 ± 3.90; z = -2.201, respectively; p <.05); (2) standard deviation of RR intervals (SDNN) (36.54 ± 16.04 vs. 11.68 ± 6.19; z = -2.201, respectively; p <.05); and (3) the stress index (11.17 ± 5.48 vs. 23.07 ± 6.31; z = -2.201, respectively; p <.05). However, during treadmill exercise, there were no significant differences in GV, compared to a resting period. In group 1, there were no significant differences in HRV, during treadmill exercise compared to the resting period. However, there were significant differences in GV (.26 ± .18 vs. .54 ± .38; z = -2.197, respectively; p <.05)., between resting and exercise period. Conclusion: This study suggests that during steady state aerobic exercise, people with poorer glucose control exhibit a pattern of metabolic inflexibility; however, HRV decreased with steady state exercise in both groups suggesting that exercise training may help to normalize HRV in patients with poorly controlled glucose levels. Disclosure M. Park: None. M. Chu: None. Y. Song: None. M. Sevil: None. A. Cinar: Research Support; Self; Dexcom, Inc., JDRF. Stock/Shareholder; Self; Dexcom, Inc., Eli Lilly and Company, Insulet Corporation, Medtronic, Novo Nordisk A/S, Tandem Diabetes Care. L.T. Quinn: None. U.G. Bronas: None. Funding JDRF