“Bioinformatics”, the hot-bed of research community, combines the knowledge of biological systems and computer science. Engineers of today are expanding their knowledge base and implementing their engineering skills of computer science, statistics, mathematics to study biological data. This field of study is increasingly becoming important in automated medical diagnosis. The scope and reach of bioinformatics in solving riddles of extreme complexity pertaining to human body is beyond our thinking capabilities. We caught hold of one of the finest, youngest bio-informatic enthusiast who has spent considerable amount of his time working in his field while pursuing UnderGraduate studies.
Ramesh is currently pursuing his final year BTech from IIT Indore. He has been involved with a lot of research work and publications during his BTech in the domain of NeuroScience, Bio-medical Computing, Genetic Programming. He has also been involved in the communities as a Google Student Ambassador and a Microsoft Student Associate of IIT Indore.
Let’s know more about his research !
Application of Artificial Intelligence and Machine Learning in the areas of Bio-medicine and Neuroscience
What is your research about ? Explain with some analogies and/or examples
- This research includes two phases; the first one is to propose a novel Constructive Genetic Programming classifier and second to demonstrate its use in developing an Android based application in order to build a real time epileptic seizure detection system.This app would run as a background application in the mobile phones and be capable of acquiring EEG signals and implementing a real-time seizure prediction/detection mechanism and contingent delivery of warning stimuli to control epileptic seizures.We have proposed a novel Constructive Genetic Programming (CGP) classifier in order to classify the EEG signals and detect epileptic seizures. We designed this CGP classifier by proposing Dynamic Initialization, Intron deletion, Hybrid Crossover and Mutation operation, Dynamic Fitness Value computation and Duplicate removal. These proposed operations in the CGP classifier are responsible for the increase in classification accuracy and computational speed in classifying the EEG signals.
- This application can be employed as an auxiliary tool for neurologists to determine onsets of seizures in analyzing 24 h EEG recordings and could be helpful in clinical epilepsy diagnostics.
- In Future, our goal is to fabricate our own low cost EEG device which could be utilized by a common man as a house hold application.
How is your research unique ?
- A wide range of methods have been proposed for classifying the EEG signals, where the classification is not upto the mark. The proposed constructive genetic programming classifier has shown remarkable accuracy in this regard. Moreover, we simultaneously developed a Windows Console application and built an Android Application based on this concept.
- Our prototype can unlock a new world of solution for detection of Epilepsy as it is precisely accurate, portable and noise-filtering. Therefore, this prototype can be adopted by all the neurosurgeons for robust clinical diagnosis and investigation of brain functions to reduce seizure frequency in people with epilepsy.
- We believe that there exists a large opportunity to bridge the burgeoning research in Cognitive Science and Soft Computing, and this application attempts to do just that.
How does your research help the people in industry or in academia ?
- Presently existing EEG tests measure the electrical activity of the brain and diagnose or monitor seizures, abnormal changes in body chemistry that effect the brain.
- But there are several limitations that attribute to the working of these existing EEG tests.
- 1) Flat electrodes with sticky paste are placed all over your scalp.
- 2) A change in moment can change in results.
- 3) The brain activity must be monitored for longer period, in that case you will be carrying a special recorder for up to three days and still many instructions on the test day which sheds a feeling of inconsistency at the existing tests.
- Moreover, the symptoms of epilepsy vary from person to person, for instance, some people may have starting spells, while others have violent shaking, and the vital thing to note is that the type of a seizure depends on the cause of epilepsy and is also contingent upon the part of the brain affected. Hence, an effectual real time detection of epileptic seizure is of paramount importance in clinical diagnosis.
- The proposed application has no such limitations. It could be used for continuous monitoring of EEG signals and also it is stress free and portable.
- This application promises to be a boon for epilepsy patients by alerting about an upcoming seizure or informing about an ongoing seizure. This application is focused on the household level and reduces the cost of epilepsy testing.
Where did you get the inspiration about the research (A professor who guided you, or maybe you got inspiration from a peer, or a senior. Describe a little) ?
- The inspiration of this research originated from Dr. Aruna Tiwari, Assistant Professor IIT Indore. One of her PhD students, Arpit Bhardwaj had put forward this idea. I was highly enthusiastic to undertake this as my research project because of its significance in the medical field.
Any advice for newbies in this field of research ? Any good competitions/conferences the students should apply for ? How to take a start with research ?
- The proposed CGP method efficiently classifies the EEG signal to detect seizures and the same can be used for various pattern recognition problems, which also involves Alzheimer’s and Parkinson’s diseases detection and diagnosis. The present scientific advancements in the field of Brain computer Interfaces and medical diagnosis promises a highly optimistic future for diagnosis and detection of neurological disorders, inventions that were once restrained to the realm of scientific fiction come into common usage.
Any outstanding achievement or recognition for the project/research?
- We have published our work in Genetic and Evolutionary Computational Conference (GECCO-14, GECCO-15) and International Conference on Biomedical Computing (BioMedCom-14).
- We are awaiting the decision from Elseiver Journal of NeuroScience Methods and IEEE Transaction on Biomedical Engineering.
- We have also filed a provisional patent for our application through IIT Indore.
Contact details for the people to reach out to you.
- Maddula Ramesh krishna, Email: [email protected], Mobile:+91- 9000055376
- Madaka Vishaal Varma, Email: [email protected], Mobile: +91- 7415169372
We are very sure that Ramesh’s research work would have inspired or motivated many of you. We would be constantly unearthing the hidden talents all across the country. To read more of our interviews, visit our research blog series. If you want your research work to be covered by The College Store, contact us here. And of course, stay tuned for more
To know more about us, visit About Us.
Team College Store