Volume-11, Issue-3

May-June, 2025

Research Paper

1. Advancing Ovarian Cancer Research for Enhanced Subtype Classification and Outlier Detection

Ovarian cancer is challenging due to late diagnosis and diverse subtypes. This study uses CNN-based models (MobileNet, DenseNet) for histopathological image classification and applies machine learning (Logistic Regression, Random Forest, XGBoost) for PCOS outlier detection. The system is supported by a Python backend and an intuitive web interface to assist clinicians. This integrated approach improves diagnostic accuracy and contributes to better patient outcomes.

Published by: Sanika Kashid, Adeetti Khamkar, Sheetal MhatreResearch Area: Medical Imaging, Artificial Intelligence, Healthcare Technology

Organisation: Usha Mittal Institute of Technology, Mumbai, MaharashtraKeywords: Ovarian Cancer, Subtype Classification, PCOS, CNN, MobileNet, DenseNet, Outlier Detection, XGBoost, Machine Learning.

Review Paper

2. The Application of Artificial Intelligence in the Field of Mental Health: A Comprehensive Review

The integration of Artificial Intelligence (AI) into mental health care has ushered in a paradigm shift in how emotional well-being is assessed, monitored, and treated. Among the various AI applications, sentiment and emotion analysis has emerged as a vital tool in extracting psychological insights from unstructured data sources such as clinical notes, therapy transcripts, social media interactions, and mobile health applications. Leveraging advanced models like BERT (Bidirectional Encoder Representations from Transformers) and GPT (Generative Pre-trained Transformer), alongside psycholinguistic tools such as LIWC (Linguistic Inquiry and Word Count) and VADER (Valence Aware Dictionary and sEntiment Reasoner), AI systems are now capable of understanding and interpreting nuanced emotional expressions in text and speech. This review paper presents a comprehensive synthesis of current research and methodologies related to AI-driven sentiment and emotion detection in the context of mental health. We explore classical and deep learning approaches, hybrid models, and multimodal frameworks applied to diverse datasets including clinical conversations, patient self-reports, and public online content. Real-world applications such as AI-powered chatbots, teletherapy platforms, and real-time monitoring tools are examined in detail. In addition, we discuss the ethical implications, including data privacy, algorithmic bias, and interpretability, which are critical for the safe deployment of AI systems in healthcare settings. The paper concludes with a set of recommendations for future research, emphasizing the need for multimodal integration, real-time analytics, and personalized mental health interventions. This work aims to inform researchers, clinicians, and developers about the current landscape and potential of AI in advancing mental health care.

Published by: Revati Sanjay MahajanResearch Area: Artificial Intelligence In Healthcare

Organisation: Tilak Maharashtra Vidyapeeth, PuneKeywords: Artificial Intelligence, Mental Health, Sentiment Analysis, Emotion Detection, NLP, BERT, GPT, LIWC, VADER, Machine Learning, Deep Learning

Research Paper

3. Context Management in Generative AI

Context management is a fundamental challenge in generative AI, directly influencing the coherence, relevance, and quality of AI-generated outputs. This paper explores the concept of context in generative AI, focusing on the difficulties models face in maintaining long-term, dynamic, and global context across interactions. Key challenges include context loss in long-term dialogues, balancing between immediate and overarching context, handling context switching in multi-turn conversations, and addressing ambiguity or incomplete context. Additionally, we examine the impact of contextual drift, scalability issues, and resource constraints. By understanding these challenges, we highlight the importance of developing more sophisticated context management techniques to improve AI's ability to generate consistent, relevant, and user-centered outputs. Finally, we discuss the implications of context management for various applications, including conversational AI, content generation, and personalized recommendations.

Published by: Rushikesh Joshi, Omkar Jainak, Naveena Bhat, Khushal Patil, Dr. Swapnaja UbaleResearch Area: Software Engineering, AI

Organisation: Marathwada Mitra Mandal's College of Engineering, PuneKeywords: Context Management, Generative AI, Artificial Intelligence (AI), Contextual Drift, Long-Term Context, Global Context, Local Context, Context Switching, Ambiguous Context, Incomplete Context, Scalability, Resource Constraints, Attention Mechanisms, Memory Networks, Neural Turing Machines (NTM), Contextual Embeddings, BERT, RoBERTa, Dynamic Context Retrieval, Recency Bias, Forgetting Mechanisms, Reinforcement Learning (RL), Multi-Agent Learning, Fine-Tuning, Transfer Learning, Privacy, Data Security, Transparency, Explainability, Bias, Fairness, Accountability, Misuse, Human-AI Interaction, User Autonomy, Ethical Considerations

Research Paper

4. Application of Optical Communication in FMCW Radar

Frequency-modulated continuous waves (FMCW) radars are long-range, frequency-modulated electromagnetic sensors that can perceive their environment in three dimensions. Recent introductions of RADARs with frequencies ranging from 60 GHz to 300 GHz have expanded their possible applications due to their improved precision in angle, range, and velocity. FMCW RADARs have a better resolution and are more accurate than narrowband and ultra-wideband (UWB) RADARs. They offer several important benefits, such as long-range perception, resistance to rain and lightning, and more, and they are less costly than cameras and LiDARs. Even yet, their outputs are less dense and noisy than those of other RADAR technologies, and their ability to measure target velocities requires the employment of specifically created algorithms. Recently, radar sensors have become more and more common in a variety of industries, such as automotive, defense, and surveillance. This is because radar sensors can withstand a wide range of conditions, such as extreme heat, bright light, and bad weather. The simulation results were performed using Optisystem 22.0 and MATLAB (R2024b). The results demonstrate that 40 mW of power is effectively utilized for target identification, with the best technique for moving targets being direct detection.

Published by: Priyanka Shukla, Priti SinghResearch Area: Computer Science & Electronics

Organisation: Rama University, Kanpur, Uttar PradeshKeywords: Radar, Sensors, Surveillance, Target Detection

Online paper publication is ongoing for the current issue and authors can submit their paper for this issue until Ongoing Submissions.