GPGudapati Syam Prasad

Gudapati Syam Prasad

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Gudapati Syam Prasad has a total of 16 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Hybrid CNN-Transformer Framework for Molecular Imaging-based Early Alzheimer’s Disease Detection

    by S. Murugesan, P. S. Manjunath, Praful R Pardhi, Anil Kumar Katta, Amiya Kumar Sahoo, Gudapati Syam Prasad

    Purpose Hybrid convolutional neural network (CNN)-transformer models are increasingly proposed for molecular imaging, but scan-level leakage and targets derived from the same image-quantification pipeline can overstate performance. We tested whether an image-only hybrid model could classify florbetapir PET amyloid positivity under participant-disjoint validation. Procedures The Dallas Lifespan Brain Study contributed 551 longitudinal PET scans from 295 participants; 152 scans (27.6%) were amyloid-positive using the released global standardized uptake value ratio (SUVR) threshold of 1.09. Raw PET volumes were rigidly registered without label access and represented by 24 axial slices. CNN-GAP, slice-transformer, and age/sex baselines were evaluated using nested five-fold stratified group cross-validation. Calibration and operating thresholds were learned only within validation partitions, and 95% confidence intervals used 2,000 participant-cluster bootstrap resamples. Results The hybrid achieved ROC-AUC 0.496 (95% CI 0.427–0.569), PR-AUC 0.274, balanced accuracy 0.490, and Matthews correlation coefficient -0.017. It did not outperform CNN-GAP (paired AUC difference -0.006, 95% CI -0.078 to 0.065) and performed below the age/sex baseline (AUC 0.690; hybrid-minus-baseline difference -0.194, 95% CI -0.287 to -0.100). Baseline-only and stable-label sensitivity analyses yielded hybrid AUCs of 0.529 and 0.502. Conclusions Raw registered PET slices did not provide stable participant-generalizable classification of an SUVR-derived amyloid target. Participant-level separation, validated tracer quantification, demographic baselines, and external validation remain necessary before clinical interpretation.

  • Multiplayer reputation-based coalition game model (MRCGM) with dominant strategy analysis for detecting malicious attacks in wireless sensor networks

    by Amara S. A. L. G. Gopala Gupta, Syam Prasad Gudapati, Praveen Tumuluru

    Wireless Sensor Networks (WSNs) are turning into an essential portion of our lives. Devoid of assuring WSNs security (WS), there remain no broad implementations of WSNs. Because of sensor nodes’ constrained capacities concerning computation, communication, and energy, giving protection to WSNs remains competitive. Indeed, the procedure of applying WS remains adaptable and vibrant that develops consistently. Attack-defend’s crux in WS could be conveyed by collaborative schemes of interdependency when Game Theory (GT) could be employed to consider communications amidst schemes of logical decision-makers. Hence, learning WS alongside GT possesses greater logic. This study proffers a Multiplayer Reputation-based Coalition Game Model (MRCGM) for seeking the false data injection amidst the nodes. An attacker attempts to sustain an ideal level of belief by correlating the identifications within the network (NW) to initiate a victorious transfer. Concurrently, the attacker should acquire a few charges for sustaining the trustability of its identifications. This proffered procedure employs this notion and turns the attack expensive by fixing a global threshold for nodes or identifications to be trustable and functional within the NW. Additionally, the beneficial function of the attacker and defender is as well described. The MRCGM has been correlated with 2 advanced methodologies like security-aware routing scheme employing repeated game (SARSRGM) paradigm and a Game-based Fuzzy Q-learning (GBFQL) scheme concerning diverse criteria. Consequently, the proffered MRCGM attains 124 kbps of Packet Drop Rate, 76.14% of energy efficiency, 96.1% of detection accuracy, 25.45% of energy consumption, and 102.45 ms of routing latency.


  • Open Access

    Discernment and Diminution of Black Hole Attack in Mobile Ad-Hoc Network using Artificial Intelligence

    by K Srinivas, V Harsha Shastri, Vinay Kumar Nassa, Gudapati Syam Prasad, Prathipati Ratna Kumar

    The MANET is an interesting research topic that has movable nodes. The mobile ad hoc system is an important research area. The nodes are connected by wireless connections. Thus it is vital to have a stable network because information and interaction are crucial in diverse fields such as security and catastrophe emergency response. Because of the decentralized system’s vibrant existence, these channels are vulnerable to various threats, including BHA, GHA, SHA. One of the well-known security issues in MANET is the black hole assault. Each node has a routing table containing information from the target node. The number of assaults in MANET is usually protocol networking assaults. The study focused on black hole attacks for discernment and decrease. A black hole attack node wrongly picks up all packet and collects them without passing them to an endpoint by taking a fresh path to the destination nodes. This paper presents BHA defence from assaults with the principle of Artificial Neural Network and calculates its efficiency according to the set of elements, such as efficiency, PDR and delay and energy utilization. This paper aims to devise an artificial neural network method for black hole discernment and reduction in MANET.

  • Open Access

    A Unified Frame Work to Integrate Hadoop and IOT to Resolve the Issues of Storage, Processing with Leveraging Capacity of Analytics

    by Gudapati Syam Prasad, P Rajesh, Sk Wasim Akram

    The new trend in the research and real time applications is Internet of Things (IOT). The functional benefits of IOT are ranging from smart house to smart cities. The main purpose of IOT is to integrate various devices logically and interacting between the devices without human intervention. The current discussion mainly focuses on leveraging the capacity of analytics in IOT and resolves the storage issues of the bulk data generated by IOT. The proposed idea gives the usage of Hadoop platform to store the data and from that data performing analytics for the sake of better utilization of IOT communications. The importance is explained with some real time scenarios where there is perfect blend of Hadoop platform and IOT. To store the various categories of the data Hadoop Distributed File System (HDFS) can be used, and to ingest the data from external platforms we can make use of Sqoop or Flume. The data available in HDFS can be used to process with the usage of Map Reduce (MR)technique. Once the data is available in HDFS the analytics can be performed with Hive, Pig or R in the context of Machine learning or data mining techniques. The outcome of the proposed idea is integration of Hadoop and IOT platforms with a unified frame work which accommodates the integration of Hadoop and IOT, storage provisions to handle bulk data, processing of the stored data and applying analytics so as to effectively serve various stake holders.