top of page

Research Area

“Shifting the burden of disease management from patients to machines”

Closed-loop Diabetes Management using AI

With an ever increasing burden of diabetes in India as well as worldwide, we are fully dedicated towards developing technologies to relieve this burden from patients. We leverage on scalable prototyping and manufacturing technologies to develop multianalyte biosensors for recording glucose, and associated biomarkers in a minimally invasive manner using microneedles. We use the sensor data with AI to accurately determine when and how much insulin needs to be injected. In this decision making process, we have developed a software platform, in addition to hardware which automatically syncs physical activity data from wearables to our AI platform.

  1. D. Kalita and K. B. Mirza, "LS-GRUNet: Glucose Forecasting Using Deep Learning for Closed-loop Diabetes Management," 2022 IEEE 7th International conference for Convergence in Technology (I2CT), Mumbai, India, 2022, pp. 1-6, doi: 10.1109/I2CT54291.2022.9824867. 

  2. D. Kalita and K. B. Mirza, "Glucose Prediction using Wide-Deep LSTM Network for Accurate Insulin Dosing in Artificial Pancreas," 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Glasgow, Scotland, United Kingdom, 2022, pp. 4426-4429, doi: 10.1109/EMBC48229.2022.9870983.

  3. D. Kalita and K. B. Mirza, "Physical Activity Classification with Smartphone based Accelerometer, Gyroscope and Device Motion for Personal Diabetes Healthcare Management," 2022 IEEE 19th India Council International Conference (INDICON), Kochi, India, 2022, pp. 1-6, doi: 10.1109/INDICON56171.2022.10039967.

List of Awards

  1. International Travel Support (ITS) (Travel Grant) by Science and Engineering Research Board (SERB), Government of India (Awarded July 2022)

  2. Awarded First Prize for Poster Presentation in 10th Annual Conference for Research Society for the Study of Diabetes in India (RSSDI), Odisha State Chapter held in Cuttack 22th August 2022. 

  3. Travel Grants Award by IEEE India Council in INDICON 2022, Kochi Kerala (Awarded November 2022)

Intelligent Biosensors and Bioelectronics

Chronic diseases require continuous monitoring long-term usage of implantable or wearable sensors. The sensor characteristics drift over time during usage due to degradation of sensing layers, biofouling. Hence, they require frequent calibrations. We are aiming to develop methods and electronic platforms which can enable in situ calibration and reliable operation throughout the period of use in the human body.

  1. H. Sharma, D. Kalita, U. Naskar, B. K. Mishra, R. Panda and K. B. Mirza, "Multimodal Simultaneous Amperometry and Electrochemical Impedance Spectroscopy Measurement for Biosensing Applications," 2022 IEEE 19th India Council International Conference (INDICON), Kochi, India, 2022, pp. 1-5, doi: 10.1109/INDICON56171.2022.10040164.

  2. H. Sharma, D. Kalita, U. Naskar, B. K. Mishra, P. Kumar and K. B. Mirza, "Prediction of Glucose Sensor Sensitivity in the Presence of Biofouling Using Machine Learning and Electrochemical Impedance Spectroscopy," in IEEE Sensors Journal, vol. 23, no. 16, pp. 18785-18797, 15 Aug.15, 2023, doi: 10.1109/JSEN.2023.3289619.

Generative AI models for Healthcare Applications

Lack of reproducible clinical data, leads to poor performance in AI models trained on such datasets for healthcare applications. We are exploring use of Generative AI for data augmentation  with the goal of improving the reproducibility of AI models used for healthcare. In addition to data augmentation, we are aiming for better interpretation and physical regeneration of thoughts/actions. We are also exploring the use of Generative AI to be integrated with biosignals in real time for virtual / augmented reality, medical robotics applications.

  1. S Dash, D Kalita, K B. Mirza, "Analysis of differences in EEG Signal features between Visual Imagery and Perception," 2022 IEEE 8th International conference for Convergence in Technology (I2CT), Pune, India, 2023,

bottom of page