Good day, Karna Vishnu Vardhana Reddy. I'm Arjun. I've had a look through your background and your professional record. I'm interested in hearing more about your work and how you see yourself fitting in here with our faculty. Ready to start?

Yeah, yeah, very good morning. OK. I am Doctor Karna Vishnuvardhan Reddy.

Please start by introducing your professional journey in academia.

Completed my Bachelor of Technology in Electronics and communication engineering from Jawaharlal Nehru Technological University, Hyderabad in 2004.

Professor

Thereafter, I have completed Master of Technology in Communication and Signal Processing from Sri Krishna Devaraya University in 2006.

Then I have joined as a faculty in various in where I have worked as Assistant professor in various institution in Hyderabad.

And then in 2013, I have moved to the Malaysia as a senior lecturer in Quest International University.

And while working as a Senior Lecturer in Quest University, I have registered my PhD in 2018 at University Technological University Technology Petronas in Malaysia itself.

I have worked on UH heart disease risk prediction using machine learning algorithms under the supervision of doctor Professor Iraivan Elamjiti, and I have published various international journals and then UH published my works, presented my works in various international conferences during my PhD journey, and I have completed my doctorate degree in 2023.

And thereafter after completion of my PhD, I moved to India and I have joined as Associate Professor in Aditya Engineering College. And soon after one year, the Aditya Engineering College has become Aditya University and I have worked on various roles in the Aditya Engineering College as head of the department in. Electronics and Communication Engineering for duration of around 10 months.

And thereafter, and I have continued as associate professor, as a faculty in the Electronics and Communication Engineering department.

And then I have completed various courses, especially machine learning course has been introduced as a open elective for the final year students in the ECE department. I have handled that course.

Successfully and with the past percentage of 90% for the first time. OK. And then I have published some various journals in the.

International conferences, OK. And then I have also, uh, applied some, uh, research proposals in various government agencies like uh, DST and ISRO and act lab to market and then, uh, quantum computing lab as as Pi and Co Pi in that.

Yeah, that's all from my side, totally. I have experience about 18 years of academic and then around.

Eight years of research experience. Thank you.

Thank you, Professor, for that detailed overview. Let’s move on to assess some specific areas of expertise relevant to this role. First, can you explain a fundamental image processing technique that you have taught or used in your research?

Fundamental image processing, actually there are several steps involved in image processing like first step is image acquisition. OK, when you are working with any kind of data, first we are going to acquisition, acquisition the images that means collecting the data set and thereafter we are going to process the images, select if there any noise is there in the image, we need to remove that noise.

By applying various filter like a special filters, linear filters OK or some kind of mask, we are going to apply on the image. So image is mainly characterized by the group of pixels and the mask or filters also going to be either three by three mask or 5 by 5 marks are going to be applied on the image to remove the noise OK and to extract the features from the images OK or to segment the images we are going to perform some.

Kind of preprocessing and along with that we can also apply the morphological operations on that to extract some kind of image features also like.

MMM.

Cortana.

Yeah. So after that, normally when we are going to actually, I, we are going to perform actually image classification, I'm going to apply the extracted features to the.

Deep learning algorithms such as CNN like that, LSTM and all, and they're going to perform the classification on the training data. First, we are going to train the algorithm with the training set.

Then after the training, we are going to apply that test set for that trained algorithms. That trained algorithms are going to classify the whether the input image is going to be either belongs to any class like suppose if it is a class one or class 0 like that. So once we the trying to model perform the classification and the test is said, we are going to evaluate the models by computing the performance metrics like cross entropy for the.

Provides machine learning algorithms like accuracy, precision, recall and F1 score and all. And if it is a like a regression task, we are going to predict the mean squared error or loss function.

Yes.

Thank you, Professor. That clarifies your understanding of image processing techniques and their evaluation. Let’s turn our attention to embedded systems and communication. Could you share how you’ve utilized or taught concepts related to embedded and communication systems during your academic career?

Yeah, well, but recently my my main focus is on completely and machine and deep learning algorithm for image processing and I moved to this.

Uh, quantum, uh, I'm going through this latest technology, learning about the quantum computing, uh, basics and fundamentals. I'm going to continue with my, uh, uh, yeah, research work towards the quantum machine learning and the quantum neural networks. I'm not exactly gone through the embedded systems and communication. Normally I am aware about the basics of communications such as.

Communication is, uh, transmitting the information from one point to another point, like from source to destination. There are several methods for.

Uh, communication, especially modulation techniques are going to be utilized for, uh, transmitting the, uh, messages. Uh, and that modulation is nothing but them, uh, varying the characteristics of the, uh, carrier signal with respect to the message signal. OK, that means amplitude. Suppose if we consider a amplitude modulation, amplitude of the carrier signal is varies with respect to the envelope of the.

Message signal OK, so like that, uh, we are going to perform the modulation in different uh properties like amplitude modulation, frequency modulation, phase modulation. So especially the main need for modulation is that.

For uh, transmitting the long distances, long distance communication and then, uh, for reducing the height of the antennas, OK, in such case, there are several modulation techniques. I, as I have told you, right analog modulation techniques like amplitude frequency phase modulation, OK And similarly for the digital modulation techniques like amplitude shift keying, frequency shift taking phase shift keying like this in analog modulation coming to amplitude modulation amplitude of the.

Signal is varies with respect to the envelope of the message signal OK and this amplitude modulation again several types we can also have the uh, double sideband suppressed carrier modulation and single sideband modulation. So when we are transmitting the normal amplitude general amplitude modulation, the.

Especially.

The bandwidth, that means the frequency domain OK, it will be twice of the message signal frequency. The amplitude of the the the bandwidth of the amplitude modulator in and power of this transmitting the amplitude modulator is also high in the amplitude modulation. That's why. Carry transmitting the carrier signal OK along with the message is not required. So we are going to suppress the carrier in the double sideband suppressed carrier modulation so that we can reduce the power required for transmitting the.

AMU and but the bandwidth is remains same as twice of the message signal frequency. So we can also reduce the channel bandwidth while transmitting the modulated wave by suppressing the one of the sideband so that it can be called as a single sideband modulation.

Right. So like this, uh, uh, I have good exposure in the analog and digital communication and coming to embedded systems. Uh, but uh, not exactly gone through. I had just, uh, have learned, uh, have learned in my masters degree. OK, that was not exactly sure about it that I came to know the basics of embedded systems is nothing, but it is combining or integrating.

Multiple hardware components become a good system OK which is going to perform a various task.

Like.

Yeah.

Thank you.

Could you elaborate on your approach to teaching both theoretical and laboratory courses effectively? How do you ensure students grasp complex concepts?

Yeah, OK. Right.

Take

Uh, normally. Umm, yeah, yeah, OK.

So during UH teaching, OK, especially for the theory classes, uh, especially from the complex UH problem, uh, while solving for the complex solving, sometimes students may not be able to understand because the classroom consists of around 60 to 70 plus students.

Balancely explaining and and make sure understanding way of the teaching. OK so I going to prefer to use several teaching methodologies. So one thing is one of The thing is talk and talk. OK I used to check and like.

So there may be some students are very good in communication, some students are very good in technological aspects, OK and to make sure balance.

Blackboard in the white chat, like whiteboard with the marker pens. So, so that we can while teaching, I'm going to explain the point by point, uh, for, especially for the students who are, uh, below average, OK, They're going to grasp some points by writing so that I can use the chalk and tag. And when the topics like, uh, some easy topics, we are going to easily understood by all the students, especially like digital.

So any subject normally, uh, when we are going to explain theoretically, OK, we are going to make sure that make sure some kind of practical, uh, application oriented problems so that students can be easily understanding in a good manner. OK, And another teaching methodology, I'm going to prefer to use a PowerPoint presentation, OK, One of the, uh, everybody using PPT, but uh, with the in the PPT also.

Running some simple calculations OK so that things also can be done with the examples.

Especially like images when we are going to. Create some images for easy understanding manner such concept like a complex concepts, OK, so that simply listening OK students by seeing can by seeing the images or videos, they're going to easily understand the concepts. That is my one of the idea so that I used to play some kind of images on the PPT and then also sometimes some videos.

Video lectures also I'm going to like especially some.

Some graphical images, graphical like some simple images, simple images are uh, like videos instead of, uh, having the video lectures, OK, Instead of uh, you can show the, uh, the concept in the form of videos, OK, that can be easily understand by the students. Another thing I'm going to recently a couple of semesters, I'm going to use some AI tools, so especially for lab handling, OK, so.

Tools are very helpful in kind of like recently I'm handling the machine learning lab in which students are going to learn about the machine learning experiments. So normally theory is one part and handling the practical doing the practical it is a different case. OK, during the theory we are going to explain the various machine learning algorithms, how they are work by using the simple examples. But in case of.

Programming, how to write the program for that, uh, implementing the machine learning algorithms and kind of data, OK, It requires some good knowledge, understanding about the, some, uh, programming skills. OK Nowadays, uh, uh, everyone is going to be familiar with the AI tools like ChatGPT, Gemini, OK. And for machine learning laboratory, I'm going to use the Google collab as one of the tool for implementing the machine learning algorithms.

Datasets, OK, so students, uh, every, uh, platform, uh, there is, uh, a tool, OK, So giving a perfect query, they're going to get the some kind of coding so that they're going to understand what exact the code is going to be helpful, how to read the data set, OK, and how to explore the data set, such kind of things, uh, learning, uh, teaching them. Most monotonically, it is a. Uh, sometimes it is going to be students getting bored, OK? And then, uh, students are going to be sleepy sometimes, OK, Instead, we are going to use some kind of a tools. They are going to be very eager to enter some kind of good queries, uh, to ask this, uh, tools and then getting the response from them and execute that programs in the, their notebooks and then getting the. Output very easily. They are going to do OK so that they are going to feel happy.

Then they are also going through line by line coding, explaining how they're going to implement that coding part and all OK like that. I'm going to use some A tools not only for the lab handling and for the teaching also I'm going to use the AI tools recently.

And, and some like other teaching methodologies, I'm going to implement giving the seminars to the students, assigning some topics at least a couple of weeks before the then student. Students are going to be prepared for that and then they're going to deliver their seminar or they're going to deliver that part of that course and in front of all the students without any fear, OK, and.

Thank

And then another thing is like, uh.

No student seminar and giving assignments. Assignment is the one of the key for them. Actually, they're going to be getting some marks also for the assignments at least every semester. They're going to be for one course, they're going to have the five marks for the assignments, 5 to 10 marks, OK. And then also some prejudice I'm going to conduct. Using online quizzes. They're going, we are going. I'm going to prepare some.

Uh, MCQS are fill in the blanks in that particular course and then I'm going to display on the screen, students are going to log into their accounts and then they're going to enter their numbers and their names, and then they're going to complete the quiz online mode. And then immediately I'm going to display the results on that screen itself. That is one of the another teaching methodology I'm going to implement.

So like this, various methodologies I'm going to use for teaching and elaboratory, yeah.

Yeah, thank.

Can you elaborate on your experience guiding students with their projects or research? How do you support them in developing both technical and research skills?

Yeah, well that is a good question. So normally completing a degree must submission of a research project that is a very important for the student actually. So in our case normally in finally year.

7th semester, the students are going to be defined. They're going to be divided into several groups and at least three to four students are per group actually.

And we're going to what students motive, OK? And how, uh, they're going to do their project things, OK? And what in what area they're going to do their project based on their interests. The, the, the student, we are going to give the preference to the students to choose the, uh, their supervisor. OK, so like that. So many of the students are going very interest in recent technologies like machine learning, deep learning and quantum computing very recently. So most of the students are very easier, very interesting to ask as to meet me to guide them actually. OK, so but mostly at least. I around maximum of two batches are going to be coming for me. For the supervising the final layer projects, OK, during the project guidance, OK, I'm, uh, actually, uh, from the beginning itself, I used to take some classes for the students actually. So like how we are going to perform the research that is very important. Simply, umm, uh, giving some topic. It is very monotonical for them Also, they simply, uh, not able to do, not able to complete. They are going to be very tension actually. So in the first week of the semester.

Sorry

So once the student is uh decided to work on Sup particular area. So I'm going to take some classes especially for the research. How the student is going to perform the research. OK, so that is not about technical first thing is a research. So I asked the student. First, uh, decide one topic or any application area, OK And based on that particular topic or application OK, I request all the students to find the good journals. That means reputed journals have been published on the work OK like especially the databases like I triple E Springer elsewhere or science Direct MDPA, Taylor and Francis like that. OK, so I asked the student uh, uh, you just.

10 journals. 10 journals for for the each student. OK, that means suppose there are four students in that group. OK, I ask the each student to get 10 journals from good good databases. OK, in that particular application. OK, I asked the students after getting the journals, after downloading them. OK, I request all the students. To go through each and every journal, especially what is the abstract of the general OK, abstract of the journal and conclusion in between the methodology part also OK. So as the student, what is the problem and then what is the methodology they have they have implemented in their work and what are the results and what are the limitations of their work so that student able to know how the good, how well the previous research.

Some journals on that particular area, OK, and then extract some at least.

Have been, uh, performed their work on that particular problem like this. I would like to ask the, all the students to get the recent journals around last five years. And then they are going to be gone through each and every paper. And I asked to, to summarize each paper at least one paragraph of the each paper so that they are going to be what is the problem and what is the method they have performed? What is the results of that work and what are the limitations?

So like that each student is going to complete around 10 papers and then after going through the 10 papers, I asked them to complete a summarized table so that they're going to perform some summary of the reviewed works. So after performing summary, I would like to ask them to get the some limitations and then get the problem statement from that existing literature. OK, So what are the based on the limitations that they're?

Going to create the problem statements so for each problem statement like at least, uh, after by combining all these 40 journals, OK, they're going to, umm, gone through all the papers and particular area and then they're going to get I asked them 3 problems statements and I asked that students to complete to work on that problem statement to execute by framing the objects. OK, so objectives are going to be prepared and.

Based on that objectives, I'd like to ask them to pay the methodology, OK, and that methodology is going to be, uh, checked, OK, whether it is going to be feasible to develop or design that particular model. OK, So that if it is OK. And then after completing the methodology part, I ask the students to work on that designing and implementation of that model. OK? So that, and every week it is going to be. Supervising OK, so from the week 1 to week 16, OK total 16 weeks, I'm going to ask the students, meet the student. I ask the students to meet me every day around 3:30 to 4:30, OK, during the free hours OK And then I'm going to get the progress from them OK, so every week I'm going to get the progress from the repo progress from the students and then. Like this step by step wise I'm going to implement the research project so once they completed the.

Now designing of a model, training them, OK, and then test the unseen data sets, OK, and then, uh, getting the results, comparing the results with respect to the existing works, then, uh, if the, uh, performance of the our model is good, then, uh, it is going to be preparing like a journal article or conference presenting in an international conferences like that. I'm going to add the before completing. The final Viva was they're going to be submit at least either international conference paper or going to be at this corpus general OK.

After that, uh, final thesis is going to be compiled by all the batch mates and then, uh, it should be submitted to me at least two weeks before the final Viva was, so that I'm going to evaluate the corrections in the thesis and then sending to the external before sending to the external.

Then finally they're coming across the Viva was, then they're going to complete their final project.

To get the degree right.

Thank you.

Now, let's turn to your research publications. Could you summarize the key areas of focus in your published works and their contribution to the field?

Yeah. So during my, uh, doctorate degree, OK, I have mainly focused on the heart disease risk prediction using machine learning algorithms in which, uh, from the beginning I have, uh, collecting the data set from the UCI machine learning repository. And then I have gone through the various data sets related to the heart disease and thereafter I'm. I started learning as actually during that time in 2018 or 2019, OK, the machine learning was very new to me and that time I have gone through various courses in a Coursera, LinkedIn and all. And then I have gone through learning attending some kind of workshop. Then I have started implementing my own models on my research work, OK. So especially that's my research area is going to be. Predicting the heart disease risk prediction on the. Heart disease data set, whether the incoming patient is going to be disease or not OK for that I have gone through exhaustive literature review OK around more than 150 to 200 international journals and publications. Then thereafter I have started preparing a review on the each paper and then getting the summary of each all the papers and then finally created some three problem statements in that especially my during my work I have especially.

Found that there is a lack of, uh, data. Actually most of the researchers worked on, uh, since small data says so that I used to, uh, find the uh, uh, used to get the big data set. That means by integrating several samples, OK, uh, there are different data sets in the UCM machine learning repository related to heart disease. So I have integrated them and then get the large data set. Then there is a. The feature extraction and that means selection of the features. I have implemented features selection methods regarding my work and then, uh, perform the hyperparameter tuning on the various machine learning algorithms, OK, And finally, I have gone through the feature extraction using the principal component analysis and then I have performed the feature selection on the principal component. OK, so like that I have gone through the my during my PhD OK and complete.

Worked on machine learning algorithms, OK developing a big large data sets and feature selection method, uh, various machine learning algorithms training OK Then again, I have gone through the hyperparameter tuning, then feature extraction and feature selection methods. So like this, I have gone through it OK uh, then uh, I have published around 3 sci journals and four international conferences during my degree OK then I have.

Did submission of my thesis and OK and after coming back to work OK.

Con, uh, continue my heart disease risk prediction with ECG signals also. OK, So at first I have taken the uh, like uh dot CSV format data. Then I have moved to the ECG signals, OK. And then I worked on ECG signals to extract in the features from the ECG there ECG data and then, uh, trying them using various, uh, deep learning models, OK, and then getting the results.

I have collaborated with various researchers in our institute, OK, in and around nearby institutes. And then I have gone through uh, next, after going through the heart disease data set, I would like to.

Thank