← Back to Archives
VOLUME 13, ISSUE 8, AUGUST 2026
Publication is in progress..
Ethics, Efficiency and Equity: The Triple Challenge of AI in Sustainable Management
Prof.Gangadhar Hugar, Dr.Rajshekhar Basavapattan, Prof.Sharanabasavareddy Patil
Comparative Assessment of Corporate Social Responsibility Initiatives and Government Welfare Schemes emphasising Orunodoi on Women’s Livelihoods and Empowerment in Assam – Study based on Barak Valley
Subrajyoti Dey, Dr. M. Gangabhushan Molankal
A Comparative Study of TabTransformer and Temporal Fusion Transformer for Loan Approval Prediction Using Static and Temporal Financial Features
Dr. Balaji K, K. Sridhar, Rajinish S
Performance Evaluation of M40 Grade Concrete Using Recycled Coarse Aggregate and Plastic Granules as Sustainable Aggregate Replacements
Sandhya Annasaheb Daud, Rahul S Patil
Comparative Study of Latency-Aware Serverless Function Orchestration using XGBoost and Random Forest Regression Models in Edge– Cloud Environments
Bhavana B R, Lekhana HN, Reshma G
Influence of AI-Powered Personalization on Consumer Purchase Intention: The Mediating Role of Consumer Trust
Dr. Prashantha Kumar O, Dr. Shashidhar S Mahantshetti, Ms. Priya.K, Ms. Anuradha H N
GIS-Based Landfill Site Suitability Analysis in Deogarh Municipal Area: A Geographical Study
Deepak Kumar Prajapat, Dr. Hemendra Singh Shaktawat
Experimental Investigation on Concrete with Partial Replacement of Aggregates by Marble Pieces and Plastic Granules
Jayesh Yashwant Gawade, Prof. R. S. Patil
Effect of Aggregate Size and Specimen Dimensions on Mechanical Performance of Concrete
Vishakha D. Ghuge, Dr.D.H. Tupe
INTELLIGENT MEDICAL AI CHATBOT
DR. BHARATHI M P, YOGESH K K, KRISHNE GOWDA G S
An Efficient Diffusion-Based Framework for Accelerated Image Generation Using Fast-DDPM
Y Vidya Indrasena, Dr C Prakasa Rao
Deep Learning and Kinematic Telematics Integration for Real-Time Pothole Detection and Dynamic Navigation Routing
Sahithyaa Krishna Kumar, Sanjay C, Mark Owen A, Yogeshwar P, R.T. Charulatha
Abstract
Studies the acoustic properties of mixed ligand in DMF-water mixture at 303.15K
Y. C. Mavle
Abstract: Acoustical properties have been measured for ligand (Ondansetron) in DMF-water mixture at 303.15K. The measurement have been perform to evaluate acoustical parameter such as adiabatic compressibility (bs), Partial molal volume (fv), intermolecular free length (Lf), apparent molal compressibility (fk), specific acoustic impedance (Z), relative association (RA), salvation number (Sn).
Keywords: Ultrasonic velocity, Salvation number, specific acoustic impedance.
Keywords: Ultrasonic velocity, Salvation number, specific acoustic impedance.
Abstract
Ethics, Efficiency and Equity: The Triple Challenge of AI in Sustainable Management
Prof.Gangadhar Hugar, Dr.Rajshekhar Basavapattan, Prof.Sharanabasavareddy Patil
DOI: 10.17148/IARJSET.2026.13802
Abstract: Artificial Intelligence (AI) has emerged as a transformative force in business and management, promising unprecedented efficiency, innovation, and sustainability. Yet, its rapid integration into organizational practices raises critical questions about ethics, inclusivity, and long-term impact. This paper explores the triple challenge of AI in sustainable management balancing ethical responsibility, operational efficiency, and equitable access. Drawing on secondary data, industry cases, and policy frameworks, the study highlights how AI-driven solutions can optimize decision-making, streamline operations, and support environmental goals, while also exposing risks such as algorithmic bias, workforce displacement, and unequal access for small and rural enterprises. Special attention is given to the role of AI in empowering MSMEs and tourism based entrepreneurship in regions like Karnataka, where digital inclusion can bridge socio-economic divides. The paper argues that sustainable management in the AI era requires not only technological innovation but also strong ethical governance and equitable policies. By proposing actionable recommendations for academia, industry, and government, this research contributes to shaping a future where AI serves as a catalyst for inclusive growth and responsible development.
Keywords: Artificial Intelligence, Sustainable Management, Ethics, Efficiency, Equity, MSMEs, Rural Entrepreneurship, Digital Transformation, Inclusive Growth, Policy Frameworks
Keywords: Artificial Intelligence, Sustainable Management, Ethics, Efficiency, Equity, MSMEs, Rural Entrepreneurship, Digital Transformation, Inclusive Growth, Policy Frameworks
Abstract
Comparative Assessment of Corporate Social Responsibility Initiatives and Government Welfare Schemes emphasising Orunodoi on Women’s Livelihoods and Empowerment in Assam – Study based on Barak Valley
Subrajyoti Dey, Dr. M. Gangabhushan Molankal
DOI: 10.17148/IARJSET.2026.13803
Abstract: This study examines how women in Assam’s Barak Valley perceive and benefit from private-sector CSR initiatives versus a major government welfare scheme (Orunodoi). A survey of 60 rural women (30 CSR beneficiaries, 30 Orunodoi recipients) was done and conduction of focus-group discussions to compare program accessibility, effectiveness, satisfaction, and livelihood outcomes. Findings suggest Orunodoi is generally seen as more accessible and immediately beneficial for household needs, whereas CSR programs (e.g. skill‐training or microcredit by companies) are valued for longer-term income improvements. Both interventions increased women’s sense of financial security, but differences in targeting and bureaucratic ease emerged. For example, Orunodoi recipients reported higher ease of enrollment (mean score ~3.6/5 vs. 2.8/5 for CSR), while CSR trainees reported slightly higher income gains (Table 2). Policy recommendations include better coordination of CSR and government efforts: for instance, linking CSR skill training with Orunodoi cash support to maximize impact. Despite a small sample, the study highlights that each approach has merits, and that women’s preferences hinge on both economic needs and social factors.
Keywords: Women’s empowerment; CSR initiatives; Orunodoi scheme; Barak Valley; livelihood outcomes.
Keywords: Women’s empowerment; CSR initiatives; Orunodoi scheme; Barak Valley; livelihood outcomes.
Abstract
A Comparative Study of TabTransformer and Temporal Fusion Transformer for Loan Approval Prediction Using Static and Temporal Financial Features
Dr. Balaji K, K. Sridhar, Rajinish S
DOI: 10.17148/IARJSET.2026.13804
Abstract: In the banking and financial services industry, loan approval is an important process that helps minimize credit risk and increase lending efficiency. Traditional machine-learning models can struggle to represent complex interactions among customer financial features and changes in financial activity over time. This paper presents a comparative study of two transformer-based deep-learning models - TabTransformer and Temporal Fusion Transformer (TFT) - for intelligent loan approval prediction. TabTransformer learns meaningful representations from structured customer data, whereas TFT is trained using twelve months of sequential financial data to model long-term financial patterns. For a fair comparison, both models are trained and tested using the same enhanced loan dataset, preprocessing pipeline and experimental setting. Performance is assessed using accuracy, precision, recall, F1-score, ROC-AUC and confusion- matrix analysis. The results show that TFT achieves 94.75% accuracy compared with 93.13% for TabTransformer, indicating that temporal financial information provides additional predictive value for automated loan application decision-making.
Keywords: Loan Approval Prediction, Tab Transformer, Temporal Fusion Transformer, Deep Learning, Credit Risk Assessment, Financial Data Analytics, Transformer Networks, Sequential Learning, Artificial Intelligence.
Keywords: Loan Approval Prediction, Tab Transformer, Temporal Fusion Transformer, Deep Learning, Credit Risk Assessment, Financial Data Analytics, Transformer Networks, Sequential Learning, Artificial Intelligence.
Abstract
Performance Evaluation of M40 Grade Concrete Using Recycled Coarse Aggregate and Plastic Granules as Sustainable Aggregate Replacements
Sandhya Annasaheb Daud, Rahul S Patil
DOI: 10.17148/IARJSET.2026.13805
Abstract: The increasing demand for natural aggregates and the disposal of construction and demolition (C&D) waste and plastic waste have created significant environmental challenges. This study investigates the mechanical properties of M40 concrete by partially replacing natural coarse aggregate with recycled aggregate (RA) and fine aggregate with plastic granules (PG). Four concrete mixes were prepared: a control mix and three modified mixes containing 20% RA + 5% PG, 30% RA + 7% PG, and 40% RA + 9% PG. Fresh and hardened concrete properties were evaluated through slump, density, compressive strength, split tensile strength, and flexural strength tests. The results indicate that moderate replacement levels provide satisfactory strength while promoting sustainable construction and reducing environmental pollution.
Keywords: Recycled Aggregate, Plastic Granules, Sustainable Concrete, M40 Concrete, Mechanical Properties.
Keywords: Recycled Aggregate, Plastic Granules, Sustainable Concrete, M40 Concrete, Mechanical Properties.
Abstract
Comparative Study of Latency-Aware Serverless Function Orchestration using XGBoost and Random Forest Regression Models in Edge– Cloud Environments
Bhavana B R, Lekhana HN, Reshma G
DOI: 10.17148/IARJSET.2026.13806
Abstract: The number of Internet of Things (IoT) applications and edge–cloud computing environments has expanded, and efficient and low-latency execution of tasks is desired. Cloud native orchestration solutions are often not well suited to deal with real-time needs due to the frequent movement of cloud workloads, changes in network and resource variations. In this study, we are interested in using two ensemble ML algorithms (Random Forest and XGBoost) to predict the time required for the execution of the tasks in the vmCloud workload dataset in a serverless edge–cloud environment. Seven workload attributes—CPU utilisation, memory usage, network latency, task size, cold-start delay, number of instances, and bandwidth—were used to train the models. RMSE, MAE, MAPE, R2 score and 5-fold cross-validation were used for the performance evaluation. The results of the experiments indicate that the performance of XGBoost is better than that of the RF, with Test R2 of 0.9209, RMSE is 10.08 ms, and MAPE is 14.57%, while RF’s Test R2 is 0.8898, RMSE is 11.89 ms, and MAPE is 18.38%. The feature importance analysis shows that network latency and cold start delay are the most important features that affect the time to execute. The findings indicate that XGBoost is a proper algorithm for latency-aware serverless function orchestration and can be applied to intelligently schedule in the edge– cloud computing environment.
Keywords: Edge Computing, Cloud Computing, IoT, Serverless Computing, XGBoost, Random Forest, Function Orchestration, Latency Prediction.
Keywords: Edge Computing, Cloud Computing, IoT, Serverless Computing, XGBoost, Random Forest, Function Orchestration, Latency Prediction.
Abstract
Influence of AI-Powered Personalization on Consumer Purchase Intention: The Mediating Role of Consumer Trust
Dr. Prashantha Kumar O, Dr. Shashidhar S Mahantshetti, Ms. Priya.K, Ms. Anuradha H N
DOI: 10.17148/IARJSET.2026.13807
Abstract: Artificial Intelligence (AI)-powered personalization has become a cornerstone of modern e-commerce by enabling organizations to deliver customized product recommendations that enhance consumer experiences and influence purchasing decisions. However, the effectiveness of AI-driven personalization depends not only on the quality of recommendations but also on consumers' trust in AI systems. This study examines the influence of AI-powered personalization, perceived transparency, explainability of AI, and perceived privacy protection on purchase intention, with consumer trust serving as the mediating variable. A quantitative, cross-sectional research design was adopted, and data were collected from consumers who had prior experience with AI-enabled personalized recommendations on online shopping platforms. The proposed conceptual model was analyzed using Structural Equation Modelling (SEM) in Jamovi. The findings reveal that AI-powered personalization, perceived transparency, and explainability of AI significantly and positively influence consumer trust, whereas perceived privacy protection does not significantly affect trust. Consumer trust, in turn, exerts a significant positive influence on purchase intention. Furthermore, perceived privacy protection has a significant direct effect on purchase intention, while the direct effects of AI-powered personalization, perceived transparency, and explainability of AI on purchase intention are not statistically significant, indicating that their influence is primarily transmitted through consumer trust. The study highlights the pivotal role of trust as the underlying mechanism linking AI-enabled personalization characteristics to consumers' purchase intentions. The findings contribute to the growing literature on AI-driven consumer behaviour by extending the understanding of trust formation in AI-enabled retail environments and offer practical implications for e-commerce firms seeking to design transparent, explainable, and trustworthy AI recommendation systems that enhance consumer purchase intentions.
Keywords: AI-Powered Personalization; Consumer Trust; Purchase Intention; Explainability of Artificial Intelligence; Perceived Transparency
Keywords: AI-Powered Personalization; Consumer Trust; Purchase Intention; Explainability of Artificial Intelligence; Perceived Transparency
Abstract
GIS-Based Landfill Site Suitability Analysis in Deogarh Municipal Area: A Geographical Study
Deepak Kumar Prajapat, Dr. Hemendra Singh Shaktawat
DOI: 10.17148/IARJSET.2026.13808
Abstract: The rapid increase in municipal solid waste (MSW) resulting from population growth, urbanization, and changing consumption patterns has emerged as a major environmental challenge in developing countries. Scientific landfill site selection is essential to minimize adverse environmental and public health impacts while ensuring sustainable waste management. The present study aims to identify suitable landfill sites in the Deogarh Municipal Area, Rajsamand District, Rajasthan, using Geographic Information System (GIS) and Multi-Criteria Decision Analysis (MCDA). Spatial datasets, including land use/land cover, road networks, water bodies, built-up areas, agricultural land, railway lines, and slope, were acquired through field surveys and secondary data sources and processed using QGIS. Buffer analysis and weighted overlay techniques were applied in accordance with the Solid Waste Management Rules, 2016, and the guidelines of the Central Pollution Control Board (CPCB) to evaluate landfill site suitability. The integrated analysis classified the study area into different suitability zones ranging from unsuitable to highly suitable. The results indicate that the existing dumping site is environmentally less suitable because of its proximity to residential areas, agricultural land, and public institutions, thereby posing potential environmental and public health risks. In contrast, the proposed alternative sites are located on barren land with favourable environmental conditions, adequate road accessibility, and greater compliance with landfill site selection criteria, making them more suitable for future landfill development. The study demonstrates that the integration of GIS and MCDA provides a robust, reliable, and scientifically sound decision-support framework for landfill site selection. The proposed methodology can assist municipal authorities and planners in developing sustainable solid waste management strategies and may serve as a replicable model for other urban areas facing similar waste management challenges. The final suitability analysis showed that 12.4% of the study area falls under highly suitable class while 28.6% was moderately suitable.
Keywords: Municipal Solid Waste (MSW), Multi-criteria Decision Analysis (MCDA), Buffer Analysis, Landfill Site Selection, (GIS), QGIS, Spatial Analysis
Keywords: Municipal Solid Waste (MSW), Multi-criteria Decision Analysis (MCDA), Buffer Analysis, Landfill Site Selection, (GIS), QGIS, Spatial Analysis
Abstract
Experimental Investigation on Concrete with Partial Replacement of Aggregates by Marble Pieces and Plastic Granules
Jayesh Yashwant Gawade, Prof. R. S. Patil
DOI: 10.17148/IARJSET.2026.13809
Abstract: Concrete is the most widely used construction material because of its strength, durability, versatility, and availability. At the same time, the construction sector consumes large quantities of natural aggregates and generates or receives large quantities of waste materials. Marble processing produces solid waste in the form of pieces and fragments, while plastic waste creates long-term environmental concerns because of its persistence and difficult disposal. This study investigates the possibility of using these two waste materials in concrete as partial replacements for conventional aggregates.
The experimental study considers M30 grade concrete. Marble pieces are used as a partial replacement for natural coarse aggregate, while plastic granules are used as a partial replacement for natural fine aggregate. The project methodology includes material testing, concrete mix proportioning, casting and curing of specimens, and strength evaluation after 7, 14, and 28 days. The synopsis specifies compressive strength, split tensile strength, and flexural strength as strength characteristics to be evaluated.
The reported results show that concrete strength generally increases with curing age. Among the replacement levels considered, the mixes containing 30% marble pieces and 7% plastic granules produced the highest reported strength performance. The synopsis concludes that higher replacement levels of 40% marble pieces and 9% plastic granules resulted in a reduction in strength, attributed to weaker bonding and increased voids.
The study therefore identifies 30% marble pieces with 7% plastic granules as the optimum combination among the investigated mixes. The findings indicate that appropriately proportioned marble and plastic waste can be incorporated into concrete while reducing dependence on natural aggregates and supporting more sustainable construction practices.
Keywords: M30 concrete, marble waste, plastic granules, coarse aggregate, fine aggregate, compressive strength, split tensile strength, sustainable concrete.
The experimental study considers M30 grade concrete. Marble pieces are used as a partial replacement for natural coarse aggregate, while plastic granules are used as a partial replacement for natural fine aggregate. The project methodology includes material testing, concrete mix proportioning, casting and curing of specimens, and strength evaluation after 7, 14, and 28 days. The synopsis specifies compressive strength, split tensile strength, and flexural strength as strength characteristics to be evaluated.
The reported results show that concrete strength generally increases with curing age. Among the replacement levels considered, the mixes containing 30% marble pieces and 7% plastic granules produced the highest reported strength performance. The synopsis concludes that higher replacement levels of 40% marble pieces and 9% plastic granules resulted in a reduction in strength, attributed to weaker bonding and increased voids.
The study therefore identifies 30% marble pieces with 7% plastic granules as the optimum combination among the investigated mixes. The findings indicate that appropriately proportioned marble and plastic waste can be incorporated into concrete while reducing dependence on natural aggregates and supporting more sustainable construction practices.
Keywords: M30 concrete, marble waste, plastic granules, coarse aggregate, fine aggregate, compressive strength, split tensile strength, sustainable concrete.
Abstract
Effect of Aggregate Size and Specimen Dimensions on Mechanical Performance of Concrete
Vishakha D. Ghuge, Dr.D.H. Tupe
DOI: 10.17148/IARJSET.2026.13810
Abstract: Concrete is a heterogeneous material. Past research has established the effect of its primary constituents namely water, cement, aggregates on the behavior at the micro level as well as the macro level. In particular, several researchers have investigated the role of aggregates on the behavior of the concrete mix, the goal being to design aggregate gradations which results in higher concrete quality. The goal of this is to study the effect of specimen size on strength in unconfined concrete.
An experimental program to determine the compressive strength of a set of unconfined concrete cylinders of varying size (100mm, 150mm, 190mm diameter) and slenderness ratio of 2 is adopted. However, while studying the effect of meso-structure on strength and size effect, care has to be taken to ensure similar levels of workability and compaction in all specimens, since otherwise no meaningful conclusions can be drawn from the test results.
Average strength of specimen is taken and It is concluded from the experimental study that with increase in size of specimen its compressive strength is observed decreased.
Keywords: compressive strength Test, size effect, meso geometry
An experimental program to determine the compressive strength of a set of unconfined concrete cylinders of varying size (100mm, 150mm, 190mm diameter) and slenderness ratio of 2 is adopted. However, while studying the effect of meso-structure on strength and size effect, care has to be taken to ensure similar levels of workability and compaction in all specimens, since otherwise no meaningful conclusions can be drawn from the test results.
Average strength of specimen is taken and It is concluded from the experimental study that with increase in size of specimen its compressive strength is observed decreased.
Keywords: compressive strength Test, size effect, meso geometry
Abstract
Digital Influence on Economic and Lifestyle Dynamics of Working Women
Dr. P. Karthika
DOI: 10.17148/IARJSET.2026.13811
Abstract: Digitalization has increasingly transformed the economic, social and personal lives of working women by facilitating access to digital services, electronic payments, online information and technology-enabled decision-making. The present study examines the relationship between digitalization and lifestyle dynamics among working women, with particular emphasis on digital usage, digital services, digital payment practices and perceived digital influence on lifestyle. The study is based on primary data collected from 542 working women. The analysis incorporates digitalization variables such as devices used, frequency of digital usage, purpose of usage, digital services, device purpose, digital payment usage, security awareness, security level and perceived digital influence. Lifestyle dynamics are examined through living standard, consumption change, lifestyle decision-making, balance of life and quality of life. Descriptive statistics and Pearson's Chi-square test were employed to identify significant associations between digitalization and lifestyle-related variables. The findings reveal a significant association between digital influence and balance of life (χ² = 11.853, p = 0.003), usage purpose and consumption change (χ² = 8.005, p = 0.018), and usage frequency and balance of life (χ² = 7.365, p = 0.025). Digitalization was also significantly associated with selected economic outcomes, including salary satisfaction, income change and income control. The findings indicate that digitalization is becoming an important component of the contemporary lifestyle of working women, although its effects vary across different dimensions of economic and personal life. The study recommends strengthening digital literacy, secure digital payment practices and women-centred digital financial services to promote inclusive economic participation and improved quality of life.
Keywords: Digitalization, Working Women, Lifestyle Dynamics, Digital Payments, Economic Empowerment, Digital Financial Services, Work-Life Balance
Keywords: Digitalization, Working Women, Lifestyle Dynamics, Digital Payments, Economic Empowerment, Digital Financial Services, Work-Life Balance
Abstract
INTELLIGENT MEDICAL AI CHATBOT
DR. BHARATHI M P, YOGESH K K, KRISHNE GOWDA G S
DOI: 10.17148/IARJSET.2026.13812
Abstract: Artificial intelligence and natural language processing have opened up new possibilities for healthcare delivery, yet patients in rural and underserved regions still struggle to get timely medical consultations, which often means delayed diagnoses and complications that could have been avoided. This paper presents an Intelligent Medical Chatbot that pairs an interactive 3D human body visualization system with a custom deep learning model, aimed at giving users real-time symptom assessment and preliminary diagnosis support. The framework brings together a fine-tuned Bidirectional Encoder Representations from Transformers (BERT) model for intent classification and entity recognition, a Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) hybrid for multi-label disease prediction, and a WebGL-rendered 3D anatomical interface where users can simply click on the body region that's bothering them. On a curated medical symptom-disease dataset, the system reaches high diagnostic accuracy while keeping response latency under a second. Our experiments show it outperforming existing rule-based and single-modal chatbots on prediction accuracy, user engagement, and clinical relevance. Overall, the work offers a scalable, accessible way to provide preliminary healthcare assistance across web and mobile platforms, with no specialized hardware required.
Keywords: Medical chatbot, deep learning, BERT, CNN-LSTM, 3D human visualization, symptom checker, healthcare AI, natural language processing, real-time diagnosis.
Keywords: Medical chatbot, deep learning, BERT, CNN-LSTM, 3D human visualization, symptom checker, healthcare AI, natural language processing, real-time diagnosis.
Abstract
Medication Management and Reminders
Dr.P.S. Vachaspati, M. Pallavi
DOI: 10.17148/IARJSET.2026.13813
Abstract: Medication adherence is an important challenge among patients, as missed or incorrect doses can lead to poor treatment outcomes and medication-related errors. This paper proposes a Smart Medication Reminder System enhanced with Machine Learning to help patients take their prescribed medicines correctly and on time. The proposed system provides timely medication reminders, maintains medication history, and analyzes patient medication patterns to predict the possibility of missed doses. It also enables family members and healthcare providers to monitor medication adherence and support patients when necessary. By combining automated reminders, medication tracking, and machine learning-based prediction, the system aims to improve medication adherence, reduce missed doses and medication errors, and provide a reliable solution for different groups of patients.
Abstract
An Efficient Diffusion-Based Framework for Accelerated Image Generation Using Fast-DDPM
Y Vidya Indrasena, Dr C Prakasa Rao
DOI: 10.17148/IARJSET.2026.13814
Abstract: Denoising Diffusion Probabilistic Models (DDPMs) have emerged as a leading class of generative models for image synthesis, yet their reliance on long Markov chains of up to one-thousand-time steps makes them impractical for many medical imaging applications, where training a single model can take days and generating one image volume can take minutes to hours. This paper presents a comprehensive analysis and implementation framework built around Fast- DDPM, an efficient diffusion-based approach that aligns the training and sampling procedures of DDPMs so that both stages operate over only ten time steps rather than the conventional one thousand. Two complementary noise-scheduling strategies, a uniform time-step scheduler and a non-uniform time-step scheduler, are described in detail, along with the modified forward and reverse processes that allow the denoising network to be trained efficiently without sacrificing sample fidelity. The framework is evaluated on three representative medical image-to-image generation tasks: multi- image super-resolution of prostate MRI, denoising of low-dose lung CT scans, and cross-modality translation of brain MRI. Across all three tasks, the accelerated framework matches or exceeds the image quality of the full DDPM baseline while reducing training time to roughly one-fifth and sampling time to roughly one-hundredth of the original cost. These results indicate that carefully aligning the number of training and sampling steps, rather than relying solely on post hoc fast samplers or expensive distillation procedures, is sufficient to close the efficiency gap that has limited the clinical adoption of diffusion models. We further discuss the implications of this efficiency gain for real-time clinical workflows, the trade-offs revealed by an ablation over the number of time steps, and directions for extending the framework to three- dimensional and four-dimensional medical imaging data.
Keywords: Diffusion models, Denoising Diffusion Probabilistic Models, medical image generation, image super- resolution, image denoising, image-to-image translation, fast sampling, computational efficiency.
Keywords: Diffusion models, Denoising Diffusion Probabilistic Models, medical image generation, image super- resolution, image denoising, image-to-image translation, fast sampling, computational efficiency.
Abstract
Deep Learning and Kinematic Telematics Integration for Real-Time Pothole Detection and Dynamic Navigation Routing
Sahithyaa Krishna Kumar, Sanjay C, Mark Owen A, Yogeshwar P, R.T. Charulatha
DOI: 10.17148/IARJSET.2026.13816
Abstract: Potholes represent a severe hazard to vehicular safety, traffic efficiency, and municipal road infrastructure sustainability. Traditional manual road inspection paradigms are inefficient, expensive, and structurally non-scalable. This paper presents a multimodal framework integrating vision-based edge deep neural networks and mobile kinematic sensor fusion for real-time pothole detection, localized classification, and dynamic navigation routing. The optical perception pipeline incorporates Contrast Stretching Adaptive Gaussian Star Filtering (CAGF) with POT-YOLOv8 and ResNet50-YOLOv8n architectures to isolate road anomalies under variable environmental conditions. Concurrently, dynamic vehicular impact is captured using three-axis accelerometer and gyroscope telemetry modeled via Quarter Vehicle Dynamics. Geotagged detection telemetry is transmitted via Application Programming Interfaces (APIs) to cloud routing servers, enabling dynamic path recalculations in Google Maps API platforms and generating automated notifications for public works departments. Empirical evaluations demonstrate benchmark object detection accuracies up to 𝟗𝟗. 𝟏𝟎 ± 𝟎. 𝟑𝟏% and real-time inference speeds of 18.4 ms/frame (≈ 𝟓𝟒. 𝟑 FPS) on edge hardware. Multimodal sensor fusion achieves a 77.5% reduction in false positives (lowering the false-positive rate to 3.2%) and field deployment accuracies between 𝟖𝟎% and 𝟖𝟕% (𝟖𝟑. 𝟓𝟎 ± 𝟑. 𝟐𝟎%), validating the framework for intelligent transportation systems.
Keywords: Pothole Detection, YOLOv8, Sensor Fusion, Dynamic Routing, Edge Computing, Geographic Information System
Keywords: Pothole Detection, YOLOv8, Sensor Fusion, Dynamic Routing, Edge Computing, Geographic Information System