Latest Updates The next call for applications for admission to RVU’s Ph.D. programmes will be announced in March 2026. | 1. KEA/CET/GOVT QUOTA—Fee details for 1st year B.Tech. CSE (Hons) AY 2025–26 | 2. Information submitted to UGC for inspection purpose | 3. ⁠UGC Public Self Disclosure | 4. B.Tech.(Hons.) JEE (Main) admission process time line | CET Code: E285

Prashant Adhikari

Prashant Adhikari

Assistant Professor

About

I am Prashant Adhikari, an Assistant Professor with strong academic and industry experience in Data Science, Artificial Intelligence, and Engineering Systems. I hold an M.Tech in Data Science (CGPA: 9.66) and a B.Tech in Mechanical Engineering, providing me with a solid interdisciplinary foundation that combines core engineering principles with modern machine learning and AI techniques.

I have over three years of industry experience and have been associated with Philips Healthcare (Digital X-ray Division), where I worked on medical imaging systems, large-scale data pipelines, and AI-driven analysis workflows. My industry exposure strengthened my focus on building robust, reliable, and deployable AI systems under real-world constraints such as performance, interpretability, safety, and regulatory requirements.

My research interests include applied machine learning, deep learning, computer vision, explainable AI (XAI), and industrial AI systems. I am a co-author of a peer-reviewed paper published at the IEEE International Joint Conference on Neural Networks (IJCNN) titled “Edge Attention Module for Object Detection”, where my contributions involved deep learning architecture design, attention mechanisms, performance optimization, and experimental evaluation.

I actively work on developing trustworthy, interpretable, and human-centred AI models for applications in healthcare, manufacturing, automation, and cyber-physical systems. My long-term research vision is to bridge the gap between academic AI research and industry-grade deployment, ensuring AI systems are not only accurate but also transparent, reliable, and ethically responsible.

As an educator, I strongly believe in conceptual clarity, rigorous problem-solving, and hands-on learning. My teaching philosophy emphasizes strong fundamentals, practical implementation, and critical thinking, preparing students for both research careers and high-impact industry roles. I actively mentor students in research projects, technical competitions, and career development.

In God we trust. All others must bring data.

    Edge Attention Module for Object Classification
  • IJCNN 2025, Rome, Italy | IEEE | doi: 10.1109/IJCNN64981.2025.11227631 Proposed a novel edge-aware attention mechanism to enhance CNN feature representation, improving object classification accuracy, robustness, and boundary localization, with minimal computational overhead.
    • Fibonacci-Net: A Lightweight CNN for Automatic Brain Tumor Classification arXiv Preprint, 2025 | arXiv: 2503.13928v1
  • Developed a lightweight CNN architecture inspired by Fibonacci growth principles, achieving high accuracy and efficiency for automated brain tumor classification from MRI scans.

  • Applied Machine Learning and Deep Learning for real-world engineering and healthcare applications
  • Computer Vision and Medical Image Analysis, including object detection and disease classification
  • Explainable AI (XAI) and Trustworthy Machine Learning for safety-critical systems
  • Industrial AI and Intelligent Automation for manufacturing and quality inspection
  • Lightweight and Efficient Neural Network Architectures for edge and embedded systems
  • Multimodal Data Fusion and Sensor-driven Decision Support Systems
  • AI for Healthcare Imaging, Diagnostics, and Clinical Decision Support
  • Research Interest
  • Applied Machine Learning and Deep Learning for real-world engineering and healthcare applications
  • Computer Vision and Medical Image Analysis, including object detection and disease classification
  • Explainable AI (XAI) and Trustworthy Machine Learning for safety-critical systems
  • Industrial AI and Intelligent Automation for manufacturing and quality inspection
  • Lightweight and Efficient Neural Network Architectures for edge and embedded systems
  • Multimodal Data Fusion and Sensor-driven Decision Support Systems
  • AI for Healthcare Imaging, Diagnostics, and Clinical Decision Support

  • Predictive Maintenance for Digital X-ray Systems (Philips)

    Developed time-series forecasting models to predict CSM brake usage and failures in X-ray systems, achieving 95% confidence forecasting and enabling a 30% reduction in unscheduled downtime.

  • Failure Prediction for Digital X-ray Detectors (SkyPlate Project)

    Designed machine learning models for early failure detection using highly imbalanced and noisy industrial sensor data, improving diagnostic reliability and enhancing clinical system performance.

  • Industrial AI Impact in Healthcare Systems

    Contributed to predictive analytics solutions that improved overall system reliability by 25% through data-driven maintenance strategies and real-time monitoring pipelines.

  • IEEE IJCNN 2025 Peer-reviewed Publication

    Co-authored a research paper accepted at the 2025 IEEE International Joint Conference on Neural Networks (IJCNN), presenting an edge-aware attention-based deep learning framework for object classification.

  • Best Master’s Academic Performance Award

    Awarded for securing CGPA 9.66/10 in M.Tech Data Science, recognizing consistent academic excellence and strong research aptitude.

Assistant Professor

Institute: School of Computer Science and Engineering

M.Tech (Data Science)

1