AI Student Ambassador Panuwat Janwattanapong: Using Deep Learning to Understand Disease

Published: 05/16/2017  

Last Updated: 05/16/2017

The Intel® Student Developer Program was created to work collaboratively with students at innovative schools and universities doing great work in the Machine Learning and Artificial Intelligence space. I had the opportunity to get to know Intel® Student Panuwat Janwattanapong and learn about how he went from a degree in robotics to working towards his PhD in engineering.

Tell us about your background.

I obtained my bachelor’s degree in mechatronics engineering from Assumption University in Thailand. Back then I was working in the field of robotics, which involved programming and control. When I finished my bachelor’s degree, I received a scholarship to pursue a PhD in electrical and computer engineering at Florida International University. Currently I am working with epileptic patients where I explore the electroencephalogram of those patients by using both machine learning and statistical approaches.

What got you started in technology?

 I started getting intensely involved in technology during my second year of my bachelor’s degree. Back at that time, the Internet of Things was one of the major topics in engineering and smart phones were becoming essential. I was working as a chair of education student committee so I gathered information about these topics and requested new courses such as iOS programming and microcontroller to the faculty. I felt that having a hands-on experience with the technology was the best way of learning new things. These past experiences had led me to the selection of my senior project which is the AU Self-Balancing Bicycle. I am excited to see that nowadays we have a much more advanced version of the project with the self-driving car.

What projects are you working on now?

My main research involves in the functional connectivity analysis of electroencephalogram (EEG) obtained from epileptic patients, where functional connectivity is defined as a correlation of events occurring between the regions of the cortex. The extent of the field is being explored by the utilization of machine learning and statistical approaches, where functional connectivity serves as a main feature that will be used in different classification algorithms such as Support Vector Machines (SVM) and neural networks. 

I am also working on text classification based on a Natural Language Process task as a side project. The task is to analyze the similarity and connection between academic papers. The general approaches of this project are transforming the text into representation vectors using Term Frequency – Inverse Document Frequency (TF-IDF) and analyze them by basic K-Mean Cluster and Non-Negative Matrix Factorization (NMF). The project is implemented by using Python with the utilization of Natural Language Toolkit (NLTK) and Scikit.  

Tell us about a technology challenge you’ve had to overcome in a project.

The first challenge that everyone has to overcome in a project is the familiarization of the developing environment. The best way to approach this problem is to conduct a literature review, clarify the problems, and search for the state-of-the-art solutions. I believe that the most challenging task to perform is to be able to switch from the current approach to some alternative methods.

What trends do you see happening in technology?

Now Machine Learning and Artificial Intelligence, especially Deep Learning, have become the buzz words of the world of research and academia. I would guess that in the near future, these tools will utilize the way machines and computers operate in a way that require less human interaction and control.

How are you planning to leverage Artificial Intelligence or Deep Learning technologies in your work?

The general focus of Artificial Intelligence and Deep Learning are in image recognition and computer vision fields. I would like to explore the tools in the time series domain, which can be applied to numerous tasks in my research. These tools can be used to extract the underlying patterns from EEG recordings where the features can be used in the classification systems. Matching generated templates obtained from epileptic patients and comparing to the control population will reveal much more information and increase our understanding of the disease.

As a Student Ambassador, what are you looking forward to doing with Intel?

I am looking forward to gaining more experience with real world applications that Intel is working on and exchange knowledge and opinions among other student ambassadors. I believe that working as a team provides different perspectives and approaches to the solution.

How can Intel help students like you succeed?

Intel is one of the greatest companies leading the world with innovations and ideas. Intel has always been passionate about technology. This student developer program, led by Intel, offers information and state-of-the-art approaches that are currently being used which I believe will help me to continue my research efficiently.

What impact on the world do you see AI having? And do you see yourself as part of it?

Artificial Intelligence is having a great impact towards society and the world. Things that we could only imagine a few years ago are currently happening nowadays and humans are adapting and embracing these technologies without any complications. Self-driving cars or personal artificial assistants are being integrated into everything we use every day. Any person who shares a deep passion in technology definitely wants to be a part of this revolution including me. In my opinion, the only way to achieve this ultimate goal is the sharing of information. 

Want to learn more? Check out our Student Developer Zone, join Intel® Developer Mesh, or learn more about becoming a Student Ambassador

Interested in more information? Contact Niven Singh


Product and Performance Information


Performance varies by use, configuration and other factors. Learn more at