P. S. Manjunath
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
P. S. Manjunath has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Amiya Kumar Sahoo
Department of Computer Science & Engineering, DRIEMS University, Cuttack, India
1 shared publication - Anil Kumar Katta
RVR & JC College of Engineering, Guntur, India
1 shared publication - Gudapati Syam Prasad
Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, India
1 shared publication K. M. Akshay
B.M.S College of Engineering
1 shared publication- Lenin Babu Chilakapati
Research Scholar Department of EEE S. V University College of Engineering Sri Venkateswara University 517502 Tirupati Andhra Pradesh India
1 shared publication - Murugesan Sundaram
Department of Artificial Intelligence and Data Science, J.J. College of Engineering and Technology, Trichy, India
1 shared publication
Latest Publications
- 23 September 2026
- Published by Springer Nature in Molecular Imaging and Biology
- p. 1-10
Hybrid CNN-Transformer Framework for Molecular Imaging-based Early Alzheimer’s Disease Detection
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.
- 23 September 2026
- Published by Springer Nature in Molecular Imaging and Biology
- p. 1-10
Hybrid CNN-Transformer Framework for Molecular Imaging-based Early Alzheimer’s Disease Detection
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.
- Open Access
- 5 July 2025
- Published by National Institute of Professional Engineers and Scientists in NIPES - Journal of Science and Technology Research
- Vol. 7 (3) , 19-30
An Artificial Neural Network Control Based Unified Power Quality Conditioner for a Renewable Energy Coordinated Distribution System
The evolution of renewable energy has been advancing at a rapid pace given the global need for clean and alternative forms of energy. Because of its convenience and ecological advantage, solar energy has been progressed as one of the most efficient of the renewable energy sources. The presented work aims to enhance efficiency of solar power generation and mitigate power quality (PQ) issues such as voltage sag, swell and total harmonic distortion (THD) through the use of Solar PV-fed Unified Power Quality Conditioner (UPQC) with an adaptive leaky least mean square (ALLMS) switching scheme based on Artificial Neural Network (ANN). In order to accomplish the objective, this control technique produce references signals that will define the switching instants in the parallel and series converters of the UPQC. Thus, in addition to PQ enhancement, the projected scheme optimizes usage of solar-PV energy injection into the grid in a cost-effective and eco-friendly manner by regulating the power flow amongst solar-PV, grid and load. Therefore, the effectiveness of the proposed solar-PV-UPQC arrangement is significantly addressed by the ALLMS control scheme and the outcomes are compared with the conventional PI controller and Fuzzy logic controller. This study was conducted using MATLAB/Simulation software, and simulation results showed that the algorithm was proven to be optimum in order to improve PQ while satisfying with the IEEE-519 standards.
- 5 July 2025
- Published by National Institute of Professional Engineers and Scientists in NIPES - Journal of Science and Technology Research
- Vol. 7 (3) , 19-30
Open AccessAn Artificial Neural Network Control Based Unified Power Quality Conditioner for a Renewable Energy Coordinated Distribution System
The evolution of renewable energy has been advancing at a rapid pace given the global need for clean and alternative forms of energy. Because of its convenience and ecological advantage, solar energy has been progressed as one of the most efficient of the renewable energy sources. The presented work aims to enhance efficiency of solar power generation and mitigate power quality (PQ) issues such as voltage sag, swell and total harmonic distortion (THD) through the use of Solar PV-fed Unified Power Quality Conditioner (UPQC) with an adaptive leaky least mean square (ALLMS) switching scheme based on Artificial Neural Network (ANN). In order to accomplish the objective, this control technique produce references signals that will define the switching instants in the parallel and series converters of the UPQC. Thus, in addition to PQ enhancement, the projected scheme optimizes usage of solar-PV energy injection into the grid in a cost-effective and eco-friendly manner by regulating the power flow amongst solar-PV, grid and load. Therefore, the effectiveness of the proposed solar-PV-UPQC arrangement is significantly addressed by the ALLMS control scheme and the outcomes are compared with the conventional PI controller and Fuzzy logic controller. This study was conducted using MATLAB/Simulation software, and simulation results showed that the algorithm was proven to be optimum in order to improve PQ while satisfying with the IEEE-519 standards.
- 21 April 2020
- Published by Springer Nature in Lecture Notes in Electrical Engineering
- p. 243-250
Implementation of Security System and Its Interconnection with Lighting
- 21 April 2020
- Published by Springer Nature in Lecture Notes in Electrical Engineering
- p. 243-250
Implementation of Security System and Its Interconnection with Lighting