@himudigonda
Welcome to my GitHub profile! I'm a passionate AI enthusiast and a Master's student in Computer Science at Arizona State University. My goal is to revolutionize healthcare and solve real-world problems using the power of Artificial Intelligence. π
- Arizona State University, Aug 2023 - May 2025
- π Proud recipient of the ASU Engineering Graduate Fellowship 2023
- π¬ Member of the cutting-edge JLiang lab at ASU
- π Thesis guided by the brilliant Prof. Jianming Liang
- SRM University, July 2019 - June 2023
- π Served as a board member at SRM E-Cell and SRM student government
- π Thesis guided by the esteemed Dr. Pradyut Kumar Sanki
- π Proud recipient of SRM's Merit Scholarship (2019-23)
- π» Programming Languages: Python (Proficient), Matlab, Java, C/C++
- π§ AI Domains: Deep Learning, Computer Vision, NLP, Machine Learning, LLMs, GenAI
- π Tools: PyTorch, TensorFlow, Sklearn, Keras, TensorRT, Timm, Transformers, Git/VCS, Docker, Kubernetes, VMWare, AWS, GCP
- π οΈ Other Skills: Research & Experimentation, Rapid Prototype Development, Problem Solving
- π©Έ Developed a polynomial kernel-based ridge regression model for non-invasive in vivo detection of random blood glucose using photoacoustic spectroscopy (PAS)
- π Achieved an RMSE of 10.94 mg/dl, MAD of 10.13 mg/dl, and MARD of 8.86%
- π Implemented robust feature extraction methods, including the Mutual Information Gain algorithm, and integrated BMI as a significant feature
- 𧩠Contributed to the development of a portable IoT-based PAS device using a pulsed laser source and piezoelectric transducer, transmitting data securely to the ThingSpeak Cloud platform via MQTT protocol
- π Conducted comprehensive validation using Clarke Error Grid Analysis and Bland-Altman Plot
- π οΈ Implemented realtime data acquisition and processing pipeline on a Raspberry Pi 4
- π Publication: Scientific Reports, Nature Portfolio
- 𧩠Designed and implemented deep learning models processing EEG and 9-axis motion data from a 14-Channel Emotiv EPOC X, achieving a 92.7% accuracy in classifying students based on cognitive learning styles
- π Developed a transformer-based architecture that improved peer performance in exams, quizzes, and assignments by 17%
- π§βπ« Collaborated with cross-functional teams of neuroscientists, psychologists, and English experts
- π Conducted a comprehensive evaluation of 9 machine learning algorithms, with Random Forest achieving the highest accuracy
- π Publication: Learning Styles Based Students Classification Using Neurolinguistics and Natural Language Processing
πΈ Realtime End-to-End Object Detection and Segmentation using Live Drone Feed From DJI Mini SE 2, Spring 2024
- π°οΈ Developed and deployed an object detection pipeline utilizing realtime camera feed for live drone applications
- π Achieved seamless integration with DJI Mini SE 2 drone
- π¨ Delivered a lightweight distilled AI framework capable of processing 22 frames per second, achieving an mAP of 59.32 on the COCO dataset
- π¨ Pretrained DINOv2, a self-supervised, no-label model, on diverse Chest X-ray datasets
- π Achieved 93.11Β±0.23% and 95.32Β±0.06% accuracy on classification tasks
- 𧩠Conducted thorough ablation studies and utilized advanced data augmentation techniques such as CutMix and MixUp
[1] π P N S B S V PRASAD V, Ali Hussain Syed, M. Himansh, Biswabandhu Jana, Pranab Mandal, and Pradyut Kumar Sanki, "Augmenting Authenticity for Non-Invasive In Vivo Detection of Random Blood Glucose with Photoacoustic Spectroscopy using Kernel-based Ridge Regression," Scientific Reports, Nature Journal.
[2] π M. Jha, A. Tiwari, M. Himansh and V. M. Manikandan, "Face Recognition: Recent Advancements, and Research Challenges," 2022 13th International Conference on Computing Communication, and Networking Technologies (ICCCNT), Kharagpur, India, 2022, pp. 1-6.
- π₯ Received gold medal for a paper titled "Learning Styles Based Students Classification Using Neurolinguistics and Natural Language Processing," presented on Research Day (6th edition) at SRM University, AP.
- π Website: himudigonda.github.io
- π«±π½βπ«²π» LinkedIn: linkedin.com/in/himudigonda
- π§ Email: himudigonda@gmail.com
- π GitHub: github.com/himudigonda
Feel free to explore my projects and publications, and let's collaborate on exciting research opportunities that leverage AI for social good. Together, we can make a positive impact on the world! πβ¨