My group compresses and deploys deep learning models on drone and microcontroller hardware, and applies them to floating plastic in lakes and coasts, fetal ultrasound, and drug safety.
I am an Assistant Professor (Grade-I) in the Department of Computer Science and Engineering at NITK Surathkal, where I joined on 28 October 2026. Before this I was at NIT Srinagar (2021 to 2026), at Madanapalle Institute of Technology and Science (2015 to 2021), at Government College of Technology, Coimbatore as a DST INSPIRE Fellow, and at Karunya as a Lecturer. I completed my Ph.D. at Anna University, Chennai, on personalised information retrieval.
My work sits where model design meets deployment: lightweight deep learning for UAV vision, on-device AI, and deep model compression, alongside a longer-running line in information retrieval and natural language processing. I am interested in how far a network can be shrunk before it stops being useful, and in what that makes possible on a drone or a microcontroller. The applications I return to are environmental monitoring, fetal and maternal health, and language tools for medicine and law.
Compact segmentation and detection models for drone imagery. The anchor application is mapping floating plastic waste in lakes and coastal waters, supported by the FloPWD and AquaPlasticWaste datasets and the EcoEdgeAI project. A newer thread studies vision-only 360° threat detection and trajectory prediction for drones in physics-based simulation.
Running inference directly on drone-mounted and microcontroller-class hardware such as Jetson boards, with no network offloading. The questions are latency, memory and energy budgets, and keeping accuracy usable under them.
Quantisation, pruning and efficient architecture design, and the trade-offs between size and accuracy. I apply them in vision and in healthcare, for example lightweight and federated models for fetal ultrasound and fetal health assessment.
My doctoral area: personalised retrieval, term-association graphs and query reformulation. Today it covers summarisation with large language models, retrieval-augmented question answering, legal and biomedical text, and adverse drug reaction signal mining from clinical text.
Veningston K, Kumar T A, Karthickmanoj R, Kumar K S, Balamurugan S (eds.). Cloud Robotics: Advances in Autonomous Navigation, Data Processing and Communication Mechanisms. Wiley-Scrivener (in preparation, 30+ chapters).
Mushtaq G, Veningston K. Fused activation deep neural network classifier with Cat-Hawk optimization for fetal health classification. Int. J. of Computers and Applications, 690–708, Aug 2026. doi:10.1080/1206212X.2026.2719005
Veningston K, Mushtaq G. A lightweight federated deep learning framework for privacy-preserving fetal ultrasound analysis. Discover Artificial Intelligence 6, 1695, 1–12. doi:10.1007/s44163-026-01827-7
Veningston K, Ronalda M, Sathiyaraj R, Selvan C, Venkateswara Rao P V, Karravula J. Real-time traffic obstacle detection using hybrid FCL for R-CNN and EfficientDet with transfer learning. Discover Artificial Intelligence 6, 186. doi:10.1007/s44163-026-00871-7
Mushtaq S, Veningston K. Biomedical text summarization with large language models: methodologies, challenges, and future directions. Int. J. of Data Science and Analytics 22, 29. doi:10.1007/s41060-025-00956-z
Veningston K, Ronalda M, Kumar S, Mir R N, Singh M. ClinicalBERT-driven adverse drug reaction signal mining with interpretable visual analytics for pharmacovigilance. BDA 2026, BITS Pilani K K Birla Goa Campus; Springer LNCS (to appear).
Veningston K, Selvan C, Ronalda M, Chaudhuri B B, Tatwal A. A multi-stage hybrid retrieval framework for knowledge-augmented question answering. ICPR 2026 International Workshops, Lyon; Springer LNCS 17110. doi:10.1007/978-3-032-39379-1_38 ANRF travel grant
Ajay I, Varshini S A, Ashwin S, Ramasamy M, Veningston K. A high-resolution data fusion framework for water body assessment using Sentinel-2 and gradient boosting. ICCPCT 2026, Kollam, 1112–1117. doi:10.1109/ICCPCT70290.2026.11654963
Mushtaq G, Veningston K. Lightweight neural models for clinical fetal ultrasound assessment. Procedia Computer Science 283, 1668–1677 (ICMLDE 2025). doi:10.1016/j.procs.2026.06.242
Veningston K, Ul Haq A, Mushtaq S, Ronalda M, Rajendran D K. Multi-task UAV dataset for floating plastic waste detection in the Kashmir region (AquaPlasticWaste 2026). IEEE DataPort. 5,000 drone images with classification, regression, semantic and instance labels. doi:10.21227/phkj-qr22
Veningston K, et al.. Real time aerial vision system for localizing and reporting plastic waste. Indian patent application 202611080911 A, published 14 Aug 2026, under examination.
Veningston K, Mushtaq G. A lightweight federated deep learning framework for privacy-preserving fetal ultrasound analysis. Int. Conf. on Advanced Materials, Semiconductor Technology and Start-up Ecosystem, College of Engineering Trivandrum, 9–11 Jan 2026 (invited talk).
Mushtaq G, Veningston K. Fetal MoAt Net: a light-weight deep learning model for fetal diagnostic plane classification. Int. J. of Computers and Applications 47(11), 927–942. doi:10.1080/1206212X.2025.2543550
Mushtaq S, Veningston K. Optimized dynamic multi-head attention for abstractive text summarization. Int. J. of Data Science and Analytics 20, 7233–7255. doi:10.1007/s41060-025-00863-3
Veningston K, Mishra A. Dataset for legal question answering system in the Indian judiciary context. Data in Brief 60, 111647. doi:10.1016/j.dib.2025.111647
Mushtaq G, Veningston K. A comprehensive review on fetal health surveillance using ultrasound imagery. Discover Applied Sciences 7, 478. doi:10.1007/s42452-025-06644-8
Veningston K, Ronalda M, Shaikh T A, Gazi F, Bhagat S. Modeling and evaluating stock market and commodity price fluctuation using time series sequence learning. SN Computer Science 6, 435. doi:10.1007/s42979-025-03931-4
Veningston K, Ajaz S, Mushtaq G, Ul Haq A. FloPWD: a novel UAV-based dataset and lightweight segmentation model for real-time edge deployment. IEEE ICEdge 2025, IISc Bangalore. doi:10.1109/ICEdge67252.2025.11412609
Veningston K, Selvan C, Prabhakar M, Kumar R S, Singh S, Polisetty B. An attention-based lightweight deep network for crack segmentation in pavements and concrete structures. IEEE ICERCS 2025, Coimbatore.
Veningston K, Selvan C, Kumar R S, Prabhakar M, Steffi D S. A comparative analysis of credit card fraud detection under extreme class imbalance. IEEE ICERCS 2025, Coimbatore.
Veningston K, Selvan C, Rao P V V, Ronalda M, Charan K, Sekhar S C, Chowdary M A. Aromaticity and instability index-based prediction of Parkinson’s disease progression using longitudinal peptide data. AICARE 2025, Kolkata. doi:10.1109/AICARE66005.2025.11402688
Madhumathi R, Veningston K, Dhanishtaa R, Elakkiya E, Gunashri R. Fine-tuning light-weight large language model for medical question answering. IEEE ICCCNT 2025, IIT Indore (EPiC Series in Computing).
Verma P, Rasool T, Shaikh T A, Veningston K (eds.). Generative Artificial Intelligence and Large Language Models for Smart Healthcare Applications. Cambridge Scholars Publishing, Dec 2025.
Selvan C, Veningston K. Introduction to generative artificial intelligence. in Generative AI and LLMs for Smart Healthcare Applications, Cambridge Scholars, ch. 1, 1–35.
Selvan C, Veningston K. Systematic view and impact of generative AI in smart healthcare systems. in Generative AI and LLMs for Smart Healthcare Applications, Cambridge Scholars, ch. 2, 36–65.
Mushtaq S, Veningston K. Generative AI and large language models in smart healthcare. in Generative AI and LLMs for Smart Healthcare Applications, Cambridge Scholars, ch. 4, 89–115.
Veningston K, Selvan C, Ronalda M, Sangeetha S K B. Applications of AI-enabled drones in agriculture. in Digital Technologies and Tools for Smart Agriculture, CRC Press, ch. 11. doi:10.1201/9781003487005
Veningston K, et al.. Dal Lake floating plastic waste detection dataset (FloPWD 2025). Mendeley Data v3. 2,006 pixel-annotated drone images. doi:10.17632/znxjncgjkc.3
Shaikh T A, Rasool T, Veningston K, Yaseen S M. The role of large language models in agriculture: harvesting the future with LLM intelligence. Progress in Artificial Intelligence 14, 117–164. doi:10.1007/s13748-024-00359-4
Mushtaq G, Veningston K. AI-driven interpretable deep learning based fetal health classification. SLAS Technology 29(6), 100206. doi:10.1016/j.slast.2024.100206
Veningston K, Mushtaq S. LoSNet: a tailored deep neural network framework for precise length of stay prediction in disease-specific hospitalization. Procedia Computer Science 235, 2599–2608 (ICMLDE 2024). doi:10.1016/j.procs.2024.04.245
Mubarakali A, Jaganathan S C B, Veningston K, et al.. Healthcare network system for elderly community based on data analytics and the Internet of Things. MAiTRI 2024, NIT Srinagar; Springer LNNS 1358. doi:10.1007/978-981-96-7614-9_27
Veningston K, Rao P V V, Devi M P, Reddy S P, Ronalda M. Plant protein classification using k-mer encoding. CINS 2023, BITS Pilani Dubai; Springer CCIS 1978. doi:10.1007/978-3-031-48984-6_8
Selvan C, Vidhya K, Kumar R S, Veningston K. Comparative overview of augmented, virtual and mixed reality. in Applications of VR and AR for Health and Wellbeing, IGI Global, 1–16. doi:10.4018/979-8-3693-1123-3.ch001
Veningston K, Rajendran D K. VR and AR in welding technologies. in Automation in Welding Industry, Wiley-Scrivener, ch. 8, 131–144. doi:10.1002/9781394172948.ch8
Veningston K, Mishra A. IndicLegalQA dataset. Mendeley Data v3. 10,000 legal question–answer pairs from Indian courts. doi:10.17632/gf8n8cnmvc.2
Mushtaq G, Veningston K. AlzNET: computational clinical decision support for diagnosing Alzheimer’s disease from MRI images. IEEE NEleX 2023, VIT Vellore. doi:10.1109/NEleX59773.2023.10421608 Best paper
Veningston K, Rao P V V, Ronalda M. Personalized multi-document text summarization using deep learning techniques. Procedia Computer Science 218, 1220–1228 (ICMLDE 2023). doi:10.1016/j.procs.2023.01.100
Sangeetha S K B, Veningston K, Sathya V, Kanthavel R. Design of a novel privacy preservation based cyber security system framework for secure medical data transactions in cloud storage. in Intelligent Edge Computing for Cyber-Physical Applications, Academic Press, 35–43. doi:10.1016/B978-0-323-99412-5.00006-X
Vairaganthan N, Daniel E, Veningston K. Privacy-preserving ranked remote document retrieval using semantic search keyword technique. INFORMATION 25(3), 201–212.
Veningston K, Rao P V V, Selvan C, Ronalda M. Investigation on customer churn prediction using machine learning techniques. ICDSA 2021; Springer LNNS 287, 109–119. doi:10.1007/978-981-16-5348-3_8
Bharathi V C, Veningston K, Rao P V V. Query-based word spotting in handwritten documents using HMM. ICDECT-2K19, Hyderabad; Springer AISC 1079, 31–39. doi:10.1007/978-981-15-1097-7_4
Veningston K, Kadry S, Kalash H S, Balamurugan B, Sathiyaraj R. Intelligent social network-based data modeling for improving health care. Health and Technology 10, 321–332. doi:10.1007/s12553-019-00303-w
Veningston K, Rao P V V. An efficient privacy-preserving search technique for encrypted cloud data. FICTA 2018, Da Nang; Springer AISC 1013, 96–106. doi:10.1007/978-981-32-9186-7_11 Best paper
Veningston K, et al.. System for preferential route control of emergency vehicle by controlling traffic lights. Indian patent application 201941033704 A, published 30 Aug 2019, under examination.
Veningston K, Shanmugalakshmi R. Personalized Information Retrieval Systems: Using Computational Intelligence Techniques (textbook). LAP Lambert Academic Publishing, 260 pp..
Lavanya T, Pamila J C M J, Veningston K. Online review analytics using word alignment model on Twitter data. IEEE ICACCS 2016. doi:10.1109/ICACCS.2016.7586388
Veningston K, Shanmugalakshmi R, Nirmala V. Semantic association ranking schemes for information retrieval applications using term association graph representation. Sādhanā 40, 1793–1819. doi:10.1007/s12046-015-0413-3
Veningston K, Shanmugalakshmi R. Personalized location aware recommendation system. IEEE ICACCS 2015, Coimbatore. doi:10.1109/ICACCS.2015.7324140 Best paper
Madhumathi R, Radhakrishnan R, Veningston K. Cloud resource allocation schemes using drip irrigation approach. in Advances in Innovative Engineering and Technologies (ICIET 2015, Bangkok), 114–124. doi:10.18797/CAASR/ICIET/2015/11/27/09
Veningston K, Shanmugalakshmi R. Computational intelligence for information retrieval using genetic algorithm. INFORMATION 17(8), 3825–3832.
Veningston K, Shanmugalakshmi R. Information retrieval by document re-ranking using term association graph. ACM ICONIAAC 2014, Amrita University. doi:10.1145/2660859.2660927 Best paper
Veningston K, Shanmugalakshmi R. Combining user-interested topic and document topic for personalized information retrieval. BDA 2014; Springer LNCS 8883, 60–79. doi:10.1007/978-3-319-13820-6_6
Veningston K, Shanmugalakshmi R. Efficient implementation of web search query reformulation using ant colony optimization. BDA 2014; Springer LNCS 8883, 80–94. doi:10.1007/978-3-319-13820-6_7
Veningston K, Shanmugalakshmi R. Personalized grouping of user search histories for efficient web search. WSEAS ACACOS 2014, Kuala Lumpur, 164–172.
Veningston K, Shanmugalakshmi R, Ruksana N. Context-aware personalization for web information retrieval: a large scale probabilistic approach. ICCBDA 2013, PSG College of Technology, Coimbatore.
Veningston K, Shanmugalakshmi R. Statistical language modeling for personalizing information retrieval. IEEE ICACCS 2013. doi:10.1109/ICACCS.2013.6938717 Best paper
Veningston K, Shanmugalakshmi R. Enhancing personalized web search re-ranking algorithm by incorporating user profile. IEEE ICCCNT 2012. doi:10.1109/ICCCNT.2012.6396036
Veningston K, Simon M. Collaborative filtering for sharing the concept-based user profiles. IEEE ICECT 2011, 187–191. doi:10.1109/ICECTECH.2011.5941884
Veningston K, Rajkumar N. A system to re-rank retrieved images from image search engines using visual similarity scores. ISSSIS 2009, CIT Coimbatore, 31–36.
Veningston K, Geetha S, Kottilingam K. Optimization of average network paging latency for remote memory in network memory system. ISSSIS 2009, CIT Coimbatore, 93–98.
Veningston K, Ghayathri J, Kottilingam K. A study on approaches to find faults in multithreaded programs. MNGSA 2008, Karunya University, 377–384.
Veningston K, Ghayathri J, Kottilingam K. Securing wireless sensor network using pairing-based identity-based encryption (IBE). MNGSA 2008, Karunya University, 385–392.
Veningston K, Rajkumar N. Design of dynamic proxy server using distributed aspect oriented programming. MNGSA 2008, Karunya University, 393–403.
Veningston K, Kottilingam K, Prasanna S. A non-server centric digital voice communication system for pilot–controller communication using VoIP. Int. Conf. on Advanced Computing and Communication Technologies for High-Performance Applications, FISAT, Angamaly, 939–942.
Journal list follows my CV, which includes the two 2014 BDA papers published in Springer LNCS. Full record on ORCID.
I teach at undergraduate and postgraduate level and like to take on courses I have not taught before, because preparing them keeps my own understanding current. I prepare complete lecture material for every course and keep it available to students.
Current-semester courses and material are listed on the department pages.
I welcome enquiries from prospective Ph.D. and M.Tech. students interested in on-device AI, language models for domain question answering, healthcare AI or graph learning. Admissions follow the institute’s official process; a short note about your interests and background is a good start.
More than 70 invited, keynote and workshop sessions since 2010, on information retrieval, deep learning, federated learning, large language models, edge AI and ethical AI in healthcare. The twelve most recent are shown below.