Report an accessibility problem

Engineering | SENSIP

Abhinav Dixit, PhD

Student Name: Abhinav Dixit, PhD

Email address: abhinav.dixit@asu.edu

 

Background and Research Interests:

“I’m a PhD student in the department of Electrical, Computer and Energy Engineering at ASU.

I received by B.E degree in Electrical Engineering from Madhav Institute of Technology and Science, M.P. India in 2013.

My research mainly focuses on Diagnosis of Neurological Disorder using irregularities in speech using speech processing and machine learning.

I am working as an Research Assistant for development and improvement of interactive Digital Signal Processing software named J-DSP. It is used by students and courses at ASU.”

 

Research Work:

1. Early Diagnosis of Neurological Disorder using Speech

The objective of my research is to find degenerative trends related to Parkinson’s disease using speech from the patients. There is no current cure for Parkinson’s. Therefore the early diagnosis is important as early medication can help mitigate the effects and increase the gestation period for over a decade. I use speech processing to extracting relevant speech features for Parkinson’s diseases and then train machine learning models to classify the disease. The problem is motivated to create automatic tools detection of diseases and remote health monitoring. Currently I am working on a project with Michael J. Fox’s speech before he was diagnosed with Parkinson’s. I am looking for tends in his speech, preceding his diagnosis that could have been a sign of the disease.

Research Picture: 

2. Machine Learning with JDSP-HTML5

The award winning J-DSP online software embeds machine learning (ML) functions for training undergraduate students and practitioners. The online ML laboratory is deployed in our classes and outreach programs. Machine learning algorithms are at the core of sensor and Internet of Things (IoT) applications.
JDSP-HTM5 can extract features from raw sensor data and cluster them using machine learning algorithms, such as the k-means, support vector machines and multilayer perceptron classification. Our online user-friendly software train participants to:

Research Picture: 

  • Understand the basics of machine learning and apply DSP methods to IoT platforms.
  • Extracted formants to train machine learning models for speech recognition.
  • Visualize different processes such as supervised and unsupervised learning.

PUBLICATIONS:

  1. A. Dixit, S. Katoch, P. Spanias, M. Banavar, H. Song, A. Spanias, “Development of Signal Processing Online Labs using HTML5 and Mobile platforms,” FIE, October, 2017.
  2. A. Dixit, J. Fan, S. Katoch, U. S. Shanthamallu, G. Muniraju, S.Rao, M. Banavar, P. Spanias, A. Spanias, A. Strom, “Multidisciplinary Modules on Sensors and Machine Learning”, ASEE, June, 2018.
  3. A. Dixit, U. S. Shanthamallu, A. Spanias, V.Berisha, M. Banavar, “Multidisciplinary Online Machine Learning Experiments in HTML5”, FIE, October, 2018.
  4. U. Shanthamallu, A. Spanias, C. Tepedelenlioglu, M. Stanley, “A Brief Survey of Machine Learning Methods and their Sensor and IoT Applications,” Proc. 8th Int. Conf. on Information, Intelligence, Systems and Applications, IEEE IISA, Larnaca, August 2017.

    5. *The work at Arizona State University is supported in part by the NSF DUE award 1525716 and the SenSIP Center. The work at Clarkson University is supported in part by the NSF DUE award 1525224.

Acknowledgements: 

I acknowledge the SenSIP Center of Electrical Engineering at ASU for the advisory and support.