Journal of Advanced Research in Applied Artificial Intelligence and Neural Network

Recent Advances in AI-Based Heart Disease Prediction: A Review of Machine Learning and Deep Learning Approaches

Rahul Kumar — 2026-06-18

The current modality of diagnosis for heart diseases lacks the sensitivity and specificity characteristics necessary for early and accurate diagnosis of this time-sensitive illness. Artificial intelligence (AI), including machine learning (ML) and deep learning (DL), has demonstrated substantial potential in aiding cardiac health diagnosis and prediction. This review article provides a comprehensive review of ML and DL techniques used in predicting heart diseases. Additionally, it provides an overview of standard ML algorithms using K-Nearest Neighbour (KNN), Support Vector Machine (SVM), Decision Trees, Random Forests, and ensemble methods and advanced deep learning models such as Deep Neural Networks (DNN) and Convolutional Neural Networks (CNN) and hybrid architectures. The importance of data preprocessing, feature engineering, feature scaling, and data augmentation techniques that significantly influence its efficacy in a highly effective manner is brought to the fore in this review. Special attention is given to methods handling class imbalance, such as the Synthetic Minority Oversampling Technique (SMOTE), that improve the detection of minority classes and relevant medical datasets. In addition, optimisation strategies, activation functions, and regularisation techniques used in deep learning models are discussed to understand their role in improving the predictive accuracy and model generalisation. Algorithms and a hybrid algorithm for deep learning-related research show relatively better indications of performance in comparison to regression-based algorithms. To reach an even simpler precision, metrics must include accuracy, sensitivity, specificity, precision, F1-score, and AUC for model evaluation. Presented herein are also particularly creative reflections on further understanding of digital research and envisioning changes for future activities related to AI for the prediction of cardiac dysfunctions – because the fine-tuned discoveries will aid in general terms for evaluation and deployment as clinical feedback mechanisms.

How to cite this article:
Kumar R, Rai A K. Recent Advances in AI-Based Heart Disease Prediction: A Review of Machine Learning and Deep Learning Approaches. J Adv Res Appl Arti Intel Neural Netw 2026; 10(2): 1-9.

AI Readiness in Higher Education: A Study of Faculty Awareness, Adoption Competency, Benefits and Challenges

Shweta Chaudhari — 2026-10-01

Artificial Intelligence (AI) is rapidly changing higher education by supporting teaching, assessment, research, content development, and administrative activities. The readiness of faculty members is a major factor in determining whether these technologies are adopted effectively and responsibly. This study examines AI readiness among higher-education faculty through five dimensions: awareness, adoption, competency, perceived benefits, and challenges. A quantitative descriptive survey design is proposed, using a structured questionnaire and a five-point Likert scale. The study considers faculty use of AI for lesson planning, content creation, assessment, research, programming support, and academic administration. It also considers barriers including insufficient training, accuracy and reliability of AI outputs, data privacy, academic integrity, ethical concerns, institutional policy gaps, and resistance to change. Recent literature indicates that effective AI integration requires not only technical knowledge but also pedagogical, ethical, and professional competencies and supportive institutional environments. The proposed study can provide an evidence-based picture of faculty AI readiness and help institutions design targeted professional development, governance, and support mechanisms for responsible AI adoption.

Artificial Intelligence and Mental Well Being: A Narrative Review

Vivek Chauhan — 2026-10-01

The global burden of mental illness continues to outpace the capacity of health systems to respond to it. According to the latest Global Burden of Disease estimates published in 2026, approximately 1.17 billion people worldwide were living with a mental disorder, highlighting the growing global burden and persistent gaps in access to adequate mental health care [1].In this context, artificial intelligence (AI) is no longer confined to research settings — it now plays an active role in mental health service delivery, including chatbots, digital phenotyping methods, and clinical decision-support systems. This review synthesises evidence from randomised controlled trials, systematic reviews, and meta-analyses published mainly between 2017 and 2026 to assess what AI can and cannot currently deliver for mental well-being. The evidence indicates that AI-based conversational agents produce a moderate, statistically reliable reduction in depressive and anxiety symptoms, with pooled effect sizes in the range of Hedges g = 0.6–0.7 across several meta-analyses, but weaker and less consistent effects on broader psychological well-being. At the same time, a fast-growing body of work documents genuine risks: hallucinated or unsafe responses in crisis situations, algorithmic bias, data privacy exposure, and emotional dependence that can, in some users, deepen rather than relieve loneliness. The review concludes that AI is best understood not as a replacement for professional mental health care but as a conditionally effective adjunct, whose benefits depend heavily on responsible design, transparent limitations, and continued human oversight.

A Review: Farm Fusion AI for Agriculture

Ramakant Govind Chaudhari — 2026-10-01

Agriculture is one of the most important sectors for global food security, yet farmers face numerous challenges such as unpredictable weather, crop diseases, pest attacks, soil degradation, and inefficient resource management. Farm Fusion AI for Agriculture is an AI-powered smart farming platform that integrates Artificial Intelligence (AI), Machine Learning (ML), the Internet of Things (IoT), and data analytics to improve agricultural productivity and support informed decision-making. The platform provides end-to-end assistance throughout the entire crop lifecycle, including land preparation, crop selection, sowing, irrigation, fertilizer management, weed control, disease and pest detection, crop monitoring, harvesting, and yield prediction. The proposed system utilizes AI and ML algorithms to analyze agricultural data, including soil properties, weather forecasts, crop health, and satellite or IoT sensor data, to generate intelligent recommendations at every stage of farming. Predictive analytics enable farmers to determine the optimal sowing time, optimize irrigation schedules, recommend suitable fertilizers, detect crop diseases and pest infestations at an early stage, and estimate crop yield before harvest. By providing real-time, data-driven insights, Farm Fusion AI enhances productivity, reduces resource wastage and production costs, and promotes sustainable precision agriculture. The proposed solution empowers farmers with intelligent decision-support tools, leading to improved crop yield,

AI in Online Education.

Mangesh Deole — 2026-10-01

The integration of artificial intelligence (AI) into online education has changed the digital learning environment, providing new opportunities to improve learner satisfaction and engagement. This systematic literature review, covering a five-year period from 2020 to 2025, investigates how AI technologies, including chatbots, intelligent tutoring systems (ITS), sentiment analysis, gaze tracking, and predictive analytics, facilitate learner engagement across cognitive, emotional, behavioral, and social dimensions. Drawing from 30 peer-reviewed studies, this review addresses three main research questions: (1) Which aspects of AI positively affect learner satisfaction and engagement in online courses within higher education institutions; (2) What potential challenges may arise from using these technologies; and (3) Which research methods are most frequently employed to evaluate AI's impact in such learning contexts? The results emphasize that adaptive learning, real-time feedback, and emotion-aware systems positively contribute to personalized learning and motivation. Nevertheless, issues remain regarding data privacy, algorithmic bias, over-reliance on automation, and system usability. Experimental and quasi-experimental designs, along with machine learning, mixed methods, and survey-based approaches, are found to dominate the reviewed studies. Based on these insights, this work provides a foundation for future AI-enhanced learning management systems aimed primarily at improving learner engagement across cognitive, emotional, behavioral, and social domains.

A case Study on how Yoga and Artificial Intelligence together improve modern health and support a better life in today’s society

Riddhi Arora — 2026-10-01

The practice of yoga has been considered the ultimate powerhouse for living a happy life since ancient times. However, in today's modern society, with the changes that have occurred in the socio-cultural lifestyle, people are gradually devoting less importance to their health. Once more, in the hectic world of the younger generations, everyone is technologically savvy because artificial intelligence is the revolution that is about to take place. The integration of yoga and artificial intelligence has already been established at the period of covid, and over the course of the past three years, yoga has gradually become an indispensable component of human life only as a result of technological advancements. The number of yoga practitioners in India who rely on virtual classes for their study and practice is expected to exceed 120 million by the year 2025. The whole market for yoga in the country is represented by this number, which represents a sizeable contribution. According to the report titled "Economic Dimensions of Yoga: Market Size Comparison Between India and The World," the market for yoga in India is estimated to be worth approximately USD 5.7 billion by the year 2023. It is projected to grow at a compound annual growth rate (CAGR) of almost 12%, which indicates a continuous rise in the number of participants, including those who participate online.

The purpose of this paper is to investigate the modern tools that are utilized in the training of yoga practitioners and to determine how these tools are facilitating the accessibility of yoga among working professionals and families. Additionally, the paper intends to investigate the modern yoga trainers and business owners who are employing modern technology to educate individuals for the purpose of improving the well-being of community members. The analysis of this case study was carried out with the assistance of secondary data by means of a number of literature reviews.

Predictive Healthcare Using Ayurveda and Machine Learning for Sustainable Bharat 2047

Ms. Sarpreet Kaur Gill — 2026-08-04

The rapid increase in chronic diseases, rising healthcare expenditure, unequal access to medical facilities, and growing environmental and demographic pressures have intensified the need for sustainable and preventive healthcare systems in India. Ayurveda, the traditional Indian system of medicine, offers a holistic framework centered on personalized healthcare, lifestyle regulation, preventive wellness, and mind–body balance. Simultaneously, advancements in Artificial Intelligence (AI), Machine Learning (ML), Internet of Things (IoT), wearable technologies, cloud computing, and predictive analytics have transformed modern healthcare by enabling intelligent diagnostics, real-time monitoring, and data-driven decision-making. This review paper critically examines the integration of Ayurveda with AI and ML technologies to develop predictive healthcare systems aligned with the vision of Sustainable Bharat 2047. The study synthesizes findings from recent research published between 2019 and 2026 related to AI-driven healthcare, Prakriti assessment, predictive disease modeling, IoT-enabled monitoring systems, digital Ayurveda platforms, and sustainable healthcare ecosystems. The review identifies major technological approaches including Random Forest, Support Vector Machines (SVM), Deep Learning, Natural Language Processing (NLP), fuzzy logic, and sensor-based analytics for personalized healthcare prediction and disease prevention. Critical challenges including lack of standardized Ayurvedic datasets, ethical concerns, algorithmic bias, cultural sensitivity, interoperability, and data privacy are also examined. Finally, the paper proposes a conceptual framework integrating traditional Ayurvedic principles with intelligent digital technologies for future-ready healthcare systems.

Impact of Generative Artificial Intelligence on Education

Junaid Khan — 2026-10-01

Generative artificial intelligence (GenAI) has quickly become part of education through tools that can create text, explanations, code, images, feedback, and other learning resources. This paper studies how GenAI affects education by reviewing recent work on teaching, learning, assessment, academic integrity, accessibility, teacher practice, and institutional governance. A structured literature review was used to identify and group peer-reviewed studies and major policy guidance published during the emergence of contemporary GenAI, with more attention to higher education while also considering schools. The reviewed studies show that GenAI can support personalized learning, rapid formative feedback, brainstorming, language development, content generation, tutoring, and teacher productivity. Recent meta-analyses also report benefits for several university learning outcomes, but the results differ depending on learner, tool, task, intervention design, and context. However, hallucinated information, bias, privacy risks, unequal access, overreliance, reduced metacognitive effort, authorship ambiguity, and assessment disruption create important challenges. The paper presents a human-centred framework in which pedagogical purpose, AI literacy, verification, assessment redesign, equity, privacy, and institutional governance together shape the educational value. The paper concludes that GenAI should be treated neither as a replacement for teachers nor as a universally beneficial shortcut, but as an assistive technology whose value depends on responsible pedagogical integration and continued research and evaluation.

Artificial Intelligence in Education, Gh Raisoni College of Engineering and Management, Jalgaon

Harshal Pramod Patil — 2026-10-01

Artificial Intelligence (AI) is transforming the education sector by providing personalized, efficient and accessible learning experiences. AI-powered systems can analyse students’ learning patterns, identify strengths and weaknesses, provide personalized recommendations, automate assessments, and offer intelligent tutoring support. Teachers can use AI Tools to reduce administrative tasks, monitor student performance, and develop more effective teaching strategies. AI also enables adaptive learning platforms that adjust content and difficulty according to individual student needs. However, challenges such as data privacy, ethical concerns, bias, digital inequality, and overdependence on technology must be carefully addressed. The responsible integration of AI in Education can improve learning outcomes, support teachers, and create more inclusive and student-centred educational environments.

Generative AI in Education: Transforming Personalized Learning

Chetan Pramod Jangid — 2026-10-01

Generative AI (GenAI) is increasingly discussed as a way to bring personalized learning to more students without multiplying the workload of already-stretched teachers. This paper looks at what that promise actually requires: not just a capable language model, but a human-centered tutoring framework that ties together learner data, curriculum-grounded content, adaptive generation, ongoing assessment, and teacher oversight. Drawing on recent empirical research, we find genuine reason for optimism about learning outcomes, but also recurring risks – hallucination, bias, privacy exposure, over-reliance, and academic-integrity concerns – that any deployment has to take seriously. We use these findings to work through applications, evaluation approaches, current limitations, and where the research still needs to go.

AI-Powered Food Label Scanner: Simplifying Healthy Food Choices Through OCR and AI

Aatmika Ramakant Dhake — 2026-10-01

Food adulteration and misleading food labels have become significant public health concerns in today's rapidly growing packaged food industry. Consumers frequently purchase processed food products without fully understanding ingredient lists, nutritional values, preservatives, or artificial additives because food labels are written using technical terminology that is difficult for ordinary people to interpret. Attractive marketing claims such as "healthy," "natural," or "low-fat" often create confusion and influence purchasing decisions without clearly reflecting the actual nutritional quality of the product. As awareness regarding nutrition continues to increase, there is a growing need for intelligent technologies that simplify food information and help consumers make healthier choices.

This paper proposes an AI-Powered Food Label Scanner that enables users to scan a product barcode or nutrition label using a smartphone camera and instantly receive a simplified health assessment. The proposed system combines Optical Character Recognition (OCR), barcode recognition, and Artificial Intelligence-based ingredient analysis to extract product information, identify harmful ingredients, detect allergens, calculate a health score, and recommend healthier alternatives. Instead of reading lengthy ingredient lists, users receive color-coded health ratings and easy-to-understand explanations. The proposed solution improves consumer awareness, saves shopping time, supports preventive healthcare, and encourages informed purchasing decisions. The system demonstrates how Artificial Intelligence can bridge the gap between complex nutritional information and everyday consumer needs

Artificial Intelligence in Higher Education: Benefits, Challenges and Future Directions

Avinash Badrinath Rathod — 2026-10-01

Artificial Intelligence (AI) and Generative AI (GenAI) technologies are fundamentally reshaping tertiary education ecosystems by transforming pedagogical delivery, programming assistance, assignment scaffolding, and institutional administration. While conventional agents and foundation models provide rapid 24/7 academic support, commercial off-the-shelf systems often deliver uncited, generalized responses lacking syllabus alignment, academic contextualization, and transparent pedagogical safeguards. This research paper presents an analytical examination of AI in higher education, assessing its historical evolution, key pedagogical applications, institutional benefits, operational bottlenecks, and ethical challenges. Combining an exhaustive systematic literature review across foundational educational computing studies with a mixed-method empirical methodology, this paper investigates how intelligent tutoring systems, large language models (LLMs), automated formative assessment, and predictive learning analytics can be systematically embedded within university curricula. Crucially, the paper explores dedicated support systems for computer applications curricula, including programming debugging, assignment scaffolding, and placement training. The paper addresses critical risks surrounding academic integrity, cognitive bypass, knowledge hallucinations, data privacy compliance, and algorithmic bias, proposing a multi-tiered Responsible AI Governance Framework for higher education institutions.