kattamuri venkata chandra sekhar

@adityatekkali.edu.in

Assistant Professor
Aditya Institute of Technology and Management

kattamuri venkata chandra sekhar


                     Download Compact PDF  
https://researchid.co/sekhar.chandra178
I am V. C. S. Kattamuri, a passionate computer science educator with 5+ years of teaching experience in shaping future tech professionals. I hold a B.Tech & M.Tech in CSE and continuously upskill to stay ahead in this fast-evolving field.

🔹 Certifications: Microsoft (AZ-900, AZ-104, SC-900), MongoDB Developer, Python (CISCO), Django (edX), AICTE Universal Human Values
🔹 Skills: Java | Python | SQL | MongoDB | Azure | Django | Web Development

I thrive on integrating industry practices into academics, mentoring students, and enabling them to solve real-world problems through technology. Always open to collaborations in teaching, research, and innovation.

EDUCATION

Earned a Bachelor of Technology (B.Tech) in Computer Science and Engineering in 2013 and a Master of Technology (M.Tech) from JNTUK in 2016.

RESEARCH, TEACHING, or OTHER INTERESTS

Artificial Intelligence, Cancer Research, Computer Vision and Pattern Recognition, Computer Science
4

Scopus Publications

4

Scholar Citations

1

Scholar h-index

Scopus Publications

  • Hybrid Mobile-Spinalnet with feature extraction for brain tumor detection using MRI images
    P. Balashanmuga Vadivu, Telagarapu Prabhakar, Jyothi Mandala, Kattamuri Venkata Chandra Sekhar
    Biomedical Signal Processing and Control, 2026
  • Reinforcement Learning-Based Treatment Recommendation Systems for Infrequent Metabolic Disorders
    G Naga Rama Devi, Balajee Maram, Kattamuri Venkata Chandra Sekhar, Ravula Rajashekhar, Alok Misra, Sathishkumar V E
    Conference Proceedings 1st International Conference on Advancing Sustainable Solutions Through Technologies Icasst 2026, 2026
    The existing abstract: Re reinforcement learning (RL) represents a new paradigm where one can be exposed to individually tailored clinical decision support, that is, when it comes to treating rare metabolic diseases which are heterogeneous in their progression and there are limited studies available on them. The paper would suggest a patient trajector therapeutic Markov decision process of re-enforcement learning model on the basis of clinic, biochemical, and demographic longitudinal characteristics. The framework uses policy optimization and deep-Q-learning to identify optimum sequences of treatments to ensure the maximal clinical utility in the long-run by a safety constraint. It is possible to solve the issue of data sparsity by including experience replay and transfer learning. As a matter of fact, the statistically significant rates of the treatment outcome score (12 to 18 percent) and the minimization of the risk of adverse events (20 percent) are proven to be observed as compared to the protocols used in case of adherence to the guidelines (p < 0.05). The proposed framework helps in the clinical level treatment recommendation using adaptive and data-driven programs and comprehensible suggestion in the case of the rare metabolic disorders.
  • Quantile Aware Causal Framework for Uncertainty Calibrated and Risk Robust Crop Yield Forecasting under Climate Variability
    P.Sirish Kumar, Kattamuri Venkata Chandra Sekhar, A. Himabindu, V. Chandrasekhar Satwik, K. Chiranjeevi, P. Yuvasri
    2025 10th International Conference on Research in Intelligent Computing in Engineering Rice 2025, 2025
    Reliable crop yield forecasting under changing climate conditions remains a major challenge due to uncertainty, data imbalance, and the presence of rare extreme events. Conventional prediction models focus mainly on average accuracy and fail to quantify uncertainty or represent tail-risk behavior, which limits their reliability for real-world decision-making. To address this gap, a quantile-aware causal framework is proposed that integrates Quantile Light Gradient Boosting Machine (Q-LightGBM) for distribution-based yield estimation, Extreme Value Theory (EVT) for modeling rare yield-loss extremes, and a Doubly Robust (DR) causal model for region-wise adaptation analysis. Technically, the framework improves reliability by calibrating the predictive intervals through quantile learning and enhancing tail-risk estimation with a dual-fit approach using Generalized Extreme Value (GEV) and Peak-Over-Threshold (POT) distributions. The causal layer adds interpretability by quantifying treatment effects of adaptation measures across diverse regions, reducing bias through combined propensity and outcome modeling. This multi-layer design enables the system to remain stable across time while maintaining sharp uncertainty bounds and balanced risk sensitivity. Experiments conducted on the CIA-2024 dataset show that the model achieves strong accuracy (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{R}^{\mathbf{2}}= 0.615)$</tex> with reliable calibration (PICP <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$=0.688)$</tex> and improved tail-risk estimation <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\left(\text{CVaR}_{99}=1.876 \text{MT} / \text{ha}\right)$</tex>. The framework maintains consistent reliability across uncertainty, tail-risk, and causal layers, demonstrating advantages over pointprediction models conceptually, although extensive baseline comparison is planned for future work. These results demonstrate that the proposed method provides a dependable, interpretable, and uncertainty-calibrated solution for climatesmart agricultural forecasting.
  • Deep Learning-Based Diagnostic Framework for Early Detection of Jaundice Using Skin and Eye Image Analysis
    Manisha Das, Bura Vijay Kumar, M.V.B. Chandrasekhar, Balajee Maram, Kattamuri Venkata Chandra Sekhar, CH Bhanu Tej
    2025 IEEE International Conference on Advanced Computing Technologies Icact 2025, 2025
    Jaundice, a yellowing of skin and eyes by elevated bilirubin, is an important early sign of liver disease. This paper suggests here a deep learning-based diagnosis system with inspection of images of skin and eyes to detect jaundice in its early stage. A 5,000 annotated image custom-made dataset is used with a specially created database in which the system combines convolutional neural networks (CNNs) for processing and extracting color features. The model developed was shown to have accuracy, sensitivity, and specificity of 94.6 %, 92.3 %, and 95.1 %, respectively, for jaundice detection. Diagnostic accuracy was greatly improved with a combination of skin and scleral image data. The computer-assisted, noninvasive technique promises large-scale screening and telemedicine-based medical treatment, enabling early medical treatment and clinical workload relief.

RECENT SCHOLAR PUBLICATIONS

  • Reinforcement Learning-Based Treatment Recommendation Systems for Infrequent Metabolic Disorders
    GNR Devi, B Maram, KVC Sekhar, R Rajashekhar, A Misra, S VE
    2026 1st International Conference on Advancing Sustainable Solutions through … , 2026
    2026
  • Hybrid Mobile-Spinalnet with feature extraction for brain tumor detection using MRI images
    PB Vadivu, T Prabhakar, J Mandala, KVC Sekhar
    Biomedical Signal Processing and Control 112, 108571 , 2026
    2026
    Citations: 4
  • Quantile Aware Causal Framework for Uncertainty Calibrated and Risk Robust Crop Yield Forecasting Under Climate Variability
    PS Kumar, KVC Sekhar, A Himabindu, VC Satwik, K Chiranjeevi, ...
    2025 10th International Conference on Research in Intelligent Computing in … , 2025
    2025
  • Deep Learning-Based Diagnostic Framework for Early Detection of Jaundice Using Skin and Eye Image Analysis
    M Das, BV Kumar, MVB Chandrasekhar, B Maram, KVC Sekhar, CHB Tej
    2025 IEEE International Conference on Advanced Computing Technologies (ICACT … , 2025
    2025
  • The Energy Efficient Routing and High Security Transmission in Mobile Ad-hoc Networks
    KVC Sekhar
    International Journal of Computer Science And Technology 9 (2), 22-27 , 2018
    2018

MOST CITED SCHOLAR PUBLICATIONS

  • Hybrid Mobile-Spinalnet with feature extraction for brain tumor detection using MRI images
    PB Vadivu, T Prabhakar, J Mandala, KVC Sekhar
    Biomedical Signal Processing and Control 112, 108571 , 2026
    2026
    Citations: 4
  • Reinforcement Learning-Based Treatment Recommendation Systems for Infrequent Metabolic Disorders
    GNR Devi, B Maram, KVC Sekhar, R Rajashekhar, A Misra, S VE
    2026 1st International Conference on Advancing Sustainable Solutions through … , 2026
    2026
  • Quantile Aware Causal Framework for Uncertainty Calibrated and Risk Robust Crop Yield Forecasting Under Climate Variability
    PS Kumar, KVC Sekhar, A Himabindu, VC Satwik, K Chiranjeevi, ...
    2025 10th International Conference on Research in Intelligent Computing in … , 2025
    2025
  • Deep Learning-Based Diagnostic Framework for Early Detection of Jaundice Using Skin and Eye Image Analysis
    M Das, BV Kumar, MVB Chandrasekhar, B Maram, KVC Sekhar, CHB Tej
    2025 IEEE International Conference on Advanced Computing Technologies (ICACT … , 2025
    2025
  • The Energy Efficient Routing and High Security Transmission in Mobile Ad-hoc Networks
    KVC Sekhar
    International Journal of Computer Science And Technology 9 (2), 22-27 , 2018
    2018