Gopakumar Gopalakrishna Pillai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Gopakumar Gopalakrishna Pillai has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- S. B. Asha
Department of Computer Science and Engineering, Amrita School of Computing, Amrita Vishwa Vidyapeetham, Amritapuri, India
4 shared publications - A. Deepthi
Amrita Vishwa Vidyapeetham
1 shared publication - Akash Kalluvilayil Venugopalan
Amrita Vishwa Vidyapeetham
1 shared publication - Akhbar Sha
Department of Computer Science and Engineering, Amrita School of Computing, Amrita Vishwa Vidyapeetham, Amritapuri, India
1 shared publication - Ashly Ajith
Amrita Vishwa Vidyapeetham
1 shared publication - Ashly Ajith
Amrita Vishwa Vidyapeetham, IN
1 shared publication
Latest Publications
- Open Access
- 16 August 2026
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 6 (1) , 818
Adaptive dual backbone feature fusion with compact latent representation learning for pap smear image classification
- 16 August 2026
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 6 (1) , 818
Open AccessAdaptive dual backbone feature fusion with compact latent representation learning for pap smear image classification
- 7 January 2026
- Published by Springer Nature in SN Computer Science
- Vol. 7 (1) , 64
Discriminative Curvature Regularization Loss for Boundary Segmentation in Microscopy Cell Images
- 7 January 2026
- Published by Springer Nature in SN Computer Science
- Vol. 7 (1) , 64
Discriminative Curvature Regularization Loss for Boundary Segmentation in Microscopy Cell Images
- 16 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (5) , 3833-3848
Integrating LLMs and Knowledge Graphs for Medical AI: Advances, Challenges, and Future Directions
This review synthesizes how integrating large language models (LLMs) with knowledge graphs (KGs) advances medical AI across methods, applications, and evaluation. While LLMs excel at natural language understanding and contextual reasoning, KGs provide structured factual knowledge, ensuring reliability in critical domains like healthcare AI. This review explores recent advances, emphasizing how LLM-KG synergy enhances knowledge extraction, clinical decision support, and explainability in medical applications. We analyze integration methodologies across three key frameworks: (a) KG-enhanced LLMs, where KGs refine reasoning during pre-training and inference; (b) LLM-augmented KGs, where LLMs improve KG construction, reasoning, and query resolution; and (c) Synergistic LLM-KG systems, which enable bidirectional knowledge exchange for more robust AI-driven decision-making. While these models offer substantial improvements in medical diagnostics, personalized treatment, and automated knowledge discovery, key challenges remain. Issues such as data heterogeneity, reasoning transparency, computational scalability, and ethical considerations surrounding patient data must be addressed to enable real-world clinical adoption. This review outlines future directions, including cross-domain knowledge integration, neurosymbolic AI frameworks, causal reasoning for explainable predictions, and multi-agent ensemble models for adaptive decision-making. We emphasize that scalability, real-time KG updates, and privacy-preserving mechanisms are vital for responsible, high-impact AI deployment in medicine.
- 16 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (5) , 3833-3848
Integrating LLMs and Knowledge Graphs for Medical AI: Advances, Challenges, and Future Directions
This review synthesizes how integrating large language models (LLMs) with knowledge graphs (KGs) advances medical AI across methods, applications, and evaluation. While LLMs excel at natural language understanding and contextual reasoning, KGs provide structured factual knowledge, ensuring reliability in critical domains like healthcare AI. This review explores recent advances, emphasizing how LLM-KG synergy enhances knowledge extraction, clinical decision support, and explainability in medical applications. We analyze integration methodologies across three key frameworks: (a) KG-enhanced LLMs, where KGs refine reasoning during pre-training and inference; (b) LLM-augmented KGs, where LLMs improve KG construction, reasoning, and query resolution; and (c) Synergistic LLM-KG systems, which enable bidirectional knowledge exchange for more robust AI-driven decision-making. While these models offer substantial improvements in medical diagnostics, personalized treatment, and automated knowledge discovery, key challenges remain. Issues such as data heterogeneity, reasoning transparency, computational scalability, and ethical considerations surrounding patient data must be addressed to enable real-world clinical adoption. This review outlines future directions, including cross-domain knowledge integration, neurosymbolic AI frameworks, causal reasoning for explainable predictions, and multi-agent ensemble models for adaptive decision-making. We emphasize that scalability, real-time KG updates, and privacy-preserving mechanisms are vital for responsible, high-impact AI deployment in medicine.
- 26 October 2024
- Published by Springer Nature in Applied Spatial Analysis and Policy
- Vol. 18 (1) , 1-18
Machine Learning for Sustainable Development: Ranking Villages for Rural Development Initiatives
- 26 October 2024
- Published by Springer Nature in Applied Spatial Analysis and Policy
- Vol. 18 (1) , 1-18
Machine Learning for Sustainable Development: Ranking Villages for Rural Development Initiatives
- 25 May 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 253
Saliency and boundary guided segmentation framework for cell counting in microscopy images
- 25 May 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 253
Saliency and boundary guided segmentation framework for cell counting in microscopy images
- Open Access
- 17 January 2024
- Published by Springer Nature in Research Square
Advancing Women’s Empowerment through Data-Driven Systems: An Ontological Knowledge Model for Holistic Assessment
Social systems that involve humans as interacting agents are complex due to the massively entangled interactions between the agents and the environment. Social engagement initiatives such as Women’s Empowerment (WE) exemplify a social system actively seeking sophisticated solutions for sharing information, acquiring knowledge, and conducting a comprehensive assessment of its impact on well-informed and effective decision-making. To facilitate this, we propose a Computational Social Science approach to assess WE. This involves integrating computational and data science methodologies into the AWESOME theoretical framework (Accelerating Women’s Empowerment through Systems Model. Despite the existence of valuable digital data, conducting a thorough evaluation of empowerment is challenging due to the variability and heterogene-ity in data source formats and types. To overcome this challenge, we propose a new ontology, AweOnto, tailored to the women’s empowerment domain and anchored in the AWESOME framework. AweOnto serves to bridge the diversity of data sources by establishing relationships between core domain concepts, functioning as an ontological knowledge model. The paper first presents how AweOnto is conceptually modeled based on the existing ontology development methodologies. Next, it discusses the model’s implementation using the protege tool owl format. The paper then demonstrates the usability of AweOnto for its application in information extraction in the form of domain-specific concept detection. The paper discusses the potential of applying ontology for exploring and extracting meta-information from unstructured texts through multi-labeled hierarchical annotations. The resulting annotations in the form of labels/tags are not intended to replace the available keywords but rather provide complimentary detailed information about the different aspects of WE.
- 17 January 2024
- Published by Springer Nature in Research Square
Open AccessAdvancing Women’s Empowerment through Data-Driven Systems: An Ontological Knowledge Model for Holistic Assessment
Social systems that involve humans as interacting agents are complex due to the massively entangled interactions between the agents and the environment. Social engagement initiatives such as Women’s Empowerment (WE) exemplify a social system actively seeking sophisticated solutions for sharing information, acquiring knowledge, and conducting a comprehensive assessment of its impact on well-informed and effective decision-making. To facilitate this, we propose a Computational Social Science approach to assess WE. This involves integrating computational and data science methodologies into the AWESOME theoretical framework (Accelerating Women’s Empowerment through Systems Model. Despite the existence of valuable digital data, conducting a thorough evaluation of empowerment is challenging due to the variability and heterogene-ity in data source formats and types. To overcome this challenge, we propose a new ontology, AweOnto, tailored to the women’s empowerment domain and anchored in the AWESOME framework. AweOnto serves to bridge the diversity of data sources by establishing relationships between core domain concepts, functioning as an ontological knowledge model. The paper first presents how AweOnto is conceptually modeled based on the existing ontology development methodologies. Next, it discusses the model’s implementation using the protege tool owl format. The paper then demonstrates the usability of AweOnto for its application in information extraction in the form of domain-specific concept detection. The paper discusses the potential of applying ontology for exploring and extracting meta-information from unstructured texts through multi-labeled hierarchical annotations. The resulting annotations in the form of labels/tags are not intended to replace the available keywords but rather provide complimentary detailed information about the different aspects of WE.
- Open Access
- 15 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Improved Discrepancy based Domain Adaptation Network using combined loss functions and Feature transformations✱
Domain shifts are a common problem in computer vision. As a result, a classifier trained on a source domain cannot perform well on a target domain. Due to this, a source classifier trained to differentiate based on a specific distribution cannot classify well on new data from a different distribution. In this work, we have studied different domain adaptation networks for their complexity and performance and have shown that a simple discrepancy based network can be modified to produce a performance on par with the more complex GAN based networks[8]. This is achieved by picking and selectively combining loss functions and preprocessing transformations. We have shown that the proposed method has produced accuracies 77.62% and 81.67% on popular datasets Office-31 and MNIST/MNIST-M, surpassing the highest reported accuracies by 6.47% and 16.67% compared to seminal papers [16][1] in this field. 1
- 15 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessImproved Discrepancy based Domain Adaptation Network using combined loss functions and Feature transformations✱
Domain shifts are a common problem in computer vision. As a result, a classifier trained on a source domain cannot perform well on a target domain. Due to this, a source classifier trained to differentiate based on a specific distribution cannot classify well on new data from a different distribution. In this work, we have studied different domain adaptation networks for their complexity and performance and have shown that a simple discrepancy based network can be modified to produce a performance on par with the more complex GAN based networks[8]. This is achieved by picking and selectively combining loss functions and preprocessing transformations. We have shown that the proposed method has produced accuracies 77.62% and 81.67% on popular datasets Office-31 and MNIST/MNIST-M, surpassing the highest reported accuracies by 6.47% and 16.67% compared to seminal papers [16][1] in this field. 1
- 30 November 2023
- Published by Springer Nature
- p. 381-394
Deep Learning-Based Semantic Segmentation of Blood Cells from Microscopic Images
- 30 November 2023
- Published by Springer Nature
- p. 381-394
Deep Learning-Based Semantic Segmentation of Blood Cells from Microscopic Images
- 30 November 2023
- Published by Springer Nature
- p. 187-198
Conjugate Gradient Method for finding Optimal Parameters in Linear Regression
- 30 November 2023
- Published by Springer Nature
- p. 187-198
Conjugate Gradient Method for finding Optimal Parameters in Linear Regression