SVSaavanth Veerumneni

Saavanth Veerumneni

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Saavanth Veerumneni has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Parenting in Roblox

    by Soumya Saraf, Muppu Prathyusha, Saavanth Veerumneni, Mallareddy Adudhodla, Pavan Kumar Reddy Manellore, G. Siva Krishna

    Playing online games is an integral part of the childhood experience as they change the way children play, learn, communicate, and develop their identities. Roblox is an example of this as the platform integrates the following features: user-generated games, avatars, chatting, creation, peer engagement, and virtual spending. The goal of this chapter is to explore the developmental significance of Roblox not in terms of a potential risk but taking into account children's age, motivations, peer group, family policies, parental participation and the platform itself, including its social, cognitive, affective, behavioral, and ethical dimensions. The main concerns addressed include the dangers related to overplay, cyberbullying, unsafe contacts, privacy risks, exposure to offensive content, and family conflicts. The opportunities associated with playing Roblox include creativity, collaboration, problem-solving skills, digital literacy, self-expression, and parent-child communication. A Risk-to-Opportunity Parenting model is applied in the analysis.

  • Hydrophobin-Functionalized Thin Films for Advanced Biosensing and Smart Nanotechnology Platforms

    by K. Gattaiah, Deemed To Be, Saavanth Veerumneni, Vasavi Bande, Mallareddy Adudhodla, Valiveti Dattatreya

    Hydrophobins constitute small and cysteine-rich fungal proteins with an outstanding ability to spontaneously assemble into highly oriented nanometric monolayer structures on hydrophobic-hydrophilic interfaces. Such self-assembled thin films provide a unique basis for chemically addressable immobilization of various biomolecules in defined orientations as recognition elements, such as antibodies, enzymes, nucleic acids, or receptor proteins. This chapter explores the principles underlying self-assembly into hydrophobin thin films, reviews existing and novel approaches to incorporate hydrophobin monolayers into biosensors based on electrochemical, optical, or piezoelectric principles, and addresses the chemical functionality of nanomaterial surfaces through hydrophobin coatings in relation to smart nanotechnology concepts. An approach to designing hydrophobin sensor devices and relevant case studies are outlined. Current technical bottlenecks and important future research topics are discussed as the basis for developing this promising technology into useful biosensing applications.

  • Optimizing Recombinant Expression Systems for High-Yield Production of Functional Biosurfactant Proteins

    by K. Gattaiah, Saavanth Veerumeni, Vasavi Bande, Valiveti Dattatreya, Mallareddy Adudhodla

    Hydrophobins constitute a new class of fungal proteins, which are distinguished by structural variability, low molecular weight, cysteine richness, amphipathic nature, and self-aggregation tendency on the interface of hydrophobic and hydrophilic surfaces. Because of these peculiarities, hydrophobins can be viewed as very prospective biocompatible compounds that may be used for the development of green biosurfactants for biomedical applications, food technology, nanotechnology, and bio-sensors. Nowadays, industrial production of hydrophobins is constrained by difficulties in creating effective recombinant protein expression systems able to produce significant quantities of folded properly proteins. In this chapter, literature dedicated to the topic of recombinant hydrophobin expression is reviewed. Various bacterial, yeast, filamentous fungal, insect cell, and cell-free expression systems suitable for producing hydrophobins are analyzed. Different strategies for improving gene/construct design, hosts' selection, and fermentation process are discussed.

  • From Lab to Edge

    by Saavanth Veerumneni, Vasavi Bande, Dattatreya Valiveti, Mallareddy Adudhodla

    Edge computer vision solutions must adhere to stringent real-world requirements regarding latency, memory, energy usage, and hardware heterogeneity that are overlooked by the benchmark-oriented approach to research. In this chapter, we present the Efficiency-Trust Co-Design Framework, which is a six-step process for deploying vision models that considers efficiency as an essential element of trustworthiness rather than as an afterthought. The process combines steps for defining constraints and risks, choosing efficient architectures, compressing models while maintaining trust, compiling code that respects hardware capabilities, designing streaming pipelines, and performing mandatory verification through five gates involving robustness, calibration, tail latency, subgroup fairness, and thermal stability.

  • Geospatial and Machine Learning Approaches for Optimizing Healthcare Accessibility in Urban Environments

    by M. Sravani, Veerumneni Saavanth, Vasavi Bande, Muppu Prathyusha, Vaishnavi Enuganti, Narasimha Swamy Yalakarajula, Matakala Dinesh

    In this chapter, an integrated framework that uses Geographic Information Systems (GIS), supervised machine learning, and location-allocation optimization is proposed to evaluate, predict, and improve healthcare accessibility in rapidly changing environments. This proposed framework is an advancement over existing conventional methods of evaluating healthcare accessibility by using multidimensional network accessibility measures, supervised learning techniques to predict healthcare service utilization based on various determinants, and mixed-integer programming techniques to optimize equity-oriented healthcare planning. A comprehensive case study of the city of Hyderabad in India shows that 2.4 million people in the city's periphery face travel times of over 60 minutes to seek primary healthcare. Machine learning models were used to achieve an R-squared value of 0.71, whereas optimization scenarios were used to reduce travel time by 31 percent. All the methods were performed using free software to ensure applicability in resource-poor healthcare systems.

  • AI-Driven Digital Twin Framework for Predictive Maintenance in Autonomous Vehicles

    by Vasavi Bande, Mallareddy Adudhodla, Prasadarao Karu, M. Swapna, M. Sravani, Saavanth Veerumneni

    The fusion of Digital Twin (DT) and Artificial Intelligence (AI) is changing the face of maintenance practices for autonomous vehicles. This chapter offers an AI-based digital twin framework for predictive maintenance to ensure timely, efficient, and reliable maintenance decisions. The framework combines various data sources related to the vehicle for continuous monitoring of the condition of the components of the vehicle to detect any signs of failure. The chapter covers DT-based maintenance systems, intelligent data management, and AI-based anomaly detection for improved safety, reliability, and efficiency of the vehicles. The chapter also covers simulation-based analysis and practical scenario-based analysis to demonstrate the efficacy of the framework. The challenges of data heterogeneity, data synchronization, and data interoperability are also covered. The future of digital twin is also discussed for smarter maintenance practices for autonomous vehicles.