Authors - Nita Dakhare, Shailesh Gahane Abstract - Kidney disease poses a significant global health challenge, necessitating innovative approaches for early detection and intervention. This study delves into the realm of predictive analytics through the utilization of machine learning algorithms to enhance kidney disease risk assessment. The research employs a comprehensive dataset comprising clinical and demographic variables, fostering a robust analysis of potential risk factors. The initial phase involves a systematic exploration of the dataset, employing statistical methods to identify correlations and patterns within the data. Subsequently, a comparative analysis of various machine learning algorithms, including but not limited to support vector machines, decision trees, and ensemble methods, is undertaken. Development of hybrid algorithm for kidney disease prediction using machine learning involves combining different techniques to improve accuracy, robustness or efficiency in predicting this condition. This evaluation aims to pinpoint the most effective model in terms of accuracy, sensitivity, and specificity in predicting kidney disease onset. The model development phase focuses on the implementation of the chosen machine learning model, incorporating features that contribute significantly to predictive accuracy. The model undergoes rigorous validation using distinct datasets to ensure its generalizability and reliability. Additionally, interpretability and transparency are prioritized to enhance the model's clinical applicability and acceptance. The study's findings provide valuable insights into the identification and understanding of key predictors of kidney disease, offering a potential tool for early diagnosis and intervention. The integration of machine learning in kidney disease prediction not only aids healthcare professionals in risk stratification but also contributes to the broader landscape of predictive analytics in preventive healthcare. The implications of this research extend to improving patient outcomes, reducing healthcare costs, and fostering a proactive approach to managing kidney disease on a global scale.
Authors - Vedant Vaidya, Shailesh Gahane, Prachi Mandade, Deepak S. Sharma, Pankajkumar Anawade Abstract - Pharmaceutical storage management is an important aspect of the health care system It makes sure medicines are on hand and stops fake drugs from spreading, while boosting overall operations. Old ways of tracking stock, like counting by hand or using barcodes, face many issues. These methods tend to be slow, prone to mistakes, and need lots of manual work. This leads to high running costs and inefficiencies. This study looks at how NFC card tech might solve these problems in drug inventory control. We focus on key areas such as accelerating inventory checks, reducing expenses, preventing counterfeit medications, protecting patients, and streamlining the supply chain. NFC cards help stop fake drugs by giving each item a secure tamper-proof ID. NFC cards aid in the fight against fake medicine. This guarantees that genuine medications pass through the supply chain. Additionally, patients are safer when utilizing NFC cards. It reduces drug mix-ups, provides reliable data on drug usage, and enables accurate prescription tracking. We also demonstrate how NFC technology improves supply chain efficiency. It streamlines the entire process of sending medications where they need to go by enabling real-time updates and reducing stock management delays. Besides, NFC calling card boost patient safety. They allow exact prescription monitoring thin down on medicinal drug mistakes, and propose trustworthy datum on drug usage. We too highlight how NFC tech further supply chemical chain productiveness. It activate live updates and cutting off delays in stock management making the whole drug distribution appendage smoother. Our research wraps up by showing that NFC placard tech offers a growth-friendly, budget-friendly fix for the crowing topic in drug inventory control. It impart major gains in precision, f number, costs, and safety. This spend a penny it a hopeful answer to bring drug supply Chain up to date.
Authors - Vanshika Landge, Shailesh Gahane, Deepak S. Sharma, Pankajkumar Anawade Abstract - The relocation to a new city poses significant challenges to the students, especially with the search for safe and relatively affordable accommodation, food service, and transportation. Stress associated with academic demands tends to be amplified in light of these difficulties, indicating the need for a fully integrated solution that would correspond to the needs of a student. This paper explores a web application aimed to help students during their relocation period to new urban environments. Key services include housing listings, food delivery options, community engagement tools, and transportation services while incorporating budgeting features that enable financial responsibility. The application is user-centric and makes relocation easier for students and fosters a sense of community among them. The research indicates that there are critical gaps in the literature. It shows that current digital solutions miss the specific needs of students, especially with regard to affordability, safety, and ease of access to essential services. The methodology includes requirement analysis, exhaustive literature reviews, development in iterations, and rigid testing to ensure that this application will meet the expectations of the users. Utilizing contemporary web technologies and real-time data integration, this project addresses both the logistical problems and emotional support to facilitate students in informed decision making. Ultimately, this research shall contribute to a better understanding of the student experience while alleviating the stress involved in moving to unknown environments and enables the students to focus on their academic pursuit while becoming an integral part of the new community. It is, therefore, an important step toward a comprehensive solution to the multifaceted problems students face from relocation.
Authors - Reena Bhagat, Smita Urkunde, Payal Khode, Shailesh Gahane Abstract - The dynamic interplay between Human Resource (HR) practices and business analytics has emerged as a pivotal factor in driving organizational performance. This research investigates the integration of HR practices with business analytics to enhance operational efficiency and strategic decision-making at Varron Autokast LTD., Nagpur. It also explores the impact of HR Analytics and Performance Management Systems on organizational outcomes at Wipro Limited, Pune. Employing a mixed-methods approach, the study delves into how HR analytics tools and data-driven strategies optimize talent management, improve workforce productivity, and align HR objectives with organizational goals. The research emphasizes the role of advanced analytics in identifying key performance indicators, fostering employee engagement, and enabling predictive insights for proactive HR interventions. Key findings aim to provide actionable frameworks for leveraging HR analytics in diverse corporate contexts, ensuring scalable, adaptive, and measurable improvements in HR processes. This study contributes to the broader understanding of HR analytics as a transformative tool for achieving sustainable competitive advantage in a rapidly evolving business landscape.
Authors - Shrinivas Patwardhan, Shailesh Gahane, Pankajkumar Anawade, Vanshika Landge, Prachi Mandade Abstract - The provision of essential medicines in rural health facilities is a complex issue, primarily influenced by frequent stock repletion, drug wastage, and poor record-keeping. Most of these problems are as a result of limited resources, old organizational systems, and poor infrastructure that characterizes most rural settings. This study evaluates the possible applicability of advanced technologies, like Radio Frequency Identification (RFID), the Internet of Things (IoT), and cloud computing, in meeting the above-mentioned requirements and to better inventory management of rural health facilities. It shall be considered with a mixed-methods approach based on survey and interview methodologies and case studies as well as cost-benefit analysis for testing feasibility, benefits, and drawback regarding the introduction of these technologies into low resource environments. The findings of this study indicate that the implementation of RFID, IoT, and cloud computing technologies possesses the capacity to significantly reduce drug wastage, enhance operational efficiency, and increase inventory accuracy. The primary obstacles to the adoption of these technologies include insufficient internet connectivity, constrained financial resources, and the necessity for specialized training. This study supports stepwise implementation, with key attention to pilot testing, financial assessment, and scalable approaches to these technological innovations. Finally, the investigation determines that, despite the considerable promise these technologies hold in transforming rural healthcare systems, there exists an urgent requirement to address technical, logistical, and financial obstacles to render them feasible and appropriate for application in resource-constrained environments.
Authors - Ritika Tiwari, Shailesh Gahanae Abstract - This research work suggested brain tumor detection and the use of a combination of deep learning and reinforcement studying techniques applied to magnetic resonance imaging (MRI) records. The mixing of deep mastering models, specifically convolutional neural networks (CNN) and reinforcement gaining knowledge of algorithms, aims to enhance the accuracy and performance of brain tumor detection structures. A comprehensive assessment of machine overall performance is carried out using standards such as sensitivity, specificity, accuracy, and computational performance. Early treatment for mind tumors is critical. The only way to identify a tumor is by biopsy, which requires mind surgical treatment. Medical doctors can locate and classify brain tumors with the help of equipment primarily based on Computational algorithms. To help medical doctors perceive early Tumor with high ac-curacy, we are able to suggest deep gaining knowledge of and diverse system studying strategies using magnetic resonance imaging mind and enable the prognosis of numerous varieties of tumors as well as healthy tumors. Massive image files need to be processed and this may be a completely time-eating undertaking. due to the fact brain tumors and normal tissues have similar findings, it is able to be tough to differentiate nearby tumors. Consequently, there's a want for a rather sensitive automatic tumor detection technique. Experimental effects demonstrate the effectiveness of our technique, with vast improvements in accuracy, sensitivity, and specificity in comparison to conventional strategies. Moreover, we discuss the consequences of our findings for scientific practice, highlighting the capacity of deep getting to know-based strategies to beautify the performance and reliability of brain tumor detection. Standard, this research contributes to advancing the sector of clinical photo evaluation and underscores the importance of leveraging deep mastering and MRI within the combat in opposition to mind tumors.
Authors - Shrinivas Patwardhan, Shailesh Gahane, Pankajkumar Anawade, Prachi Mandade, Vedant Vaidya Abstract - Pharmaceutical inventory management in health care settings is important to ensure accessibility, access and ability to essential medicines. However, the challenges in rural areas include limited infrastructure, insufficient storage systems, disabled tracking methods, poor visibility in the supply chain and lack of monitoring of real-time portfolio. These factors cause frequent warehouses, drugs and disruption in the patient's care, affecting health results in signed areas. This paper examines the current status of pharmaceutical inventory management in rural health systems, including both manual and automatic systems to determine the efficiency, efficiency and scalability of these approaches. It then examines the effect of poor inventory management on medicines, patient safety and general lack of health care. In addition, the study in existing research and training, especially in the environment with low resources, where cost effective, technology -driven solutions are necessary, intervals within.
Authors - Reena Bhagat, Smita Urkunde, Payal Khode, Shailesh Gahane Abstract - Data driven strategy is already on the rise for better performance of Human Resource in its decision making, thus helping to attract business in the current global market. The escalating growth, hands in glove with human resources, is the transformation of HR analytics within performance management systems, motivating organizations to consolidate the objectives and performance of individuals. The current research is about the integration and impacts of HR Analytics made in Wipro Limited, Pune and aims to identify the role of HR Analytics toward improvement in the performance of the workforce, aligning their goals, and mean to enhance the overall productivity of the organization. This research would cover both the methods: quantitative and qualitative analyses to establish the use and effectiveness of HR Analytics when it introduces quantitative data analysis along with the instrument with qualitative data. Some commonly faced challenges where HR analytics could be used are: silos in data, lack of technological infrastructure, employee resistance, and so on. This research will also embody certain benefits of the HR analytics among some of which: it helps in decision-making, talent management, and allocation of resources in a better manner. It further gives strategic recommendations to organizations for optimum integration of HR analytics and brings out actionable insights to better guarantee performance and subsequent organizational growth. The new findings contribute to HR Analytics and HRM Literature Growth, which can serve as praxis toward the solution for HR professionals and organizational leaders or policymakers.
Authors - Lal Mohan kumar, Shailesh Gahane, Chandan Kumar, Deepak S. Sharma, Pankajkumar Anawade Abstract - This paper does go into the roles cloud computing has in changing the face of online education, but this time, it focuses on its advantages and the flip-side of it all. Advantages reaped from using cloud computing in the education sector include resource access to scalable, flexible, and accessible learning, where students are able to learn through various personalized learning experiences with collaborative learning environments from which the students and their educators interact and share insights in real time. Most importantly, this paper discovers that cloud-based platforms offer many benefits, such as improving access to educational resources and data analytics to achieve personalized learning support for diversity in learning styles. However, despite the widespread benefits, this study also considers inevitable critical challenges that may limit equal access to education, such as creating considerable difficulties related to data privacy issues, digital literacy, and the digital divide. Therefore, research needs to be con-ducted to apply cloud computing solutions in education to improve understanding of its benefits and limitations. Such recognition would lead to better incorporation of cloud computing solutions to facilitate learner engagement, improve educational outcomes, and support inclusive educational ecosystems in those institutions. Thus, this paper suggests more empirical research be conducted to understand the long-term impact of cloud computing on student performance, engagement, and retention in different educational contexts.
Authors - Vanshika Landge, Shailesh Gahane, Deepak S. Sharma, Pankajkumar Anawade Abstract - Public transportation systems face rising pressure to provide services that are secure, efficient, and accessible to users, while still having a major segment dependent on outdated infrastructure which cannot fulfill the demands of modern-day commuters. Some key challenges include inefficient fare-collection mechanisms, rigid travel routes and poor provision of real-time information. This paper covers the adoption of Radio Frequency Identification (RFID) and Near Field Communication (NFC) technologies within the public transportation system as one of the comprehensive approaches. The proposed solution integrates safe and contactless fare collection along with dynamic travel flexibility through real-time GPS updates with help of smart cards as well as mobile applications. Its multi-phase research approach toward requirement analysis, prototype building, pilot testing, and scaling up ensures the robustness as well as practicality in the system. Modular architectures for scalability, safe use of advanced encryption, as well as intuitive interfaces towards users are integrated into this proposed solution. Pilot implementations show considerable improvements in operational efficiency, transaction accuracy, passenger satisfaction, and system reliability. The results show that RFID and NFC technologies are promising innovations to trans-form public transportation to address essential weaknesses in security, adaptability, and user convenience. This work lays a foundation for introducing innovative, integrated solutions to urban mobility in a manner that promotes sustainable, adaptable, and commuter-centered transit systems.
Authors - Darshan L.M, Nagasundara K.B Abstract - Nowadays, most of the people alters/or conceals their true facial appearance intentionally and/or unintentionally by wearing various disguise accessories such as sunglasses, artificial beard and moustache, face make-up, and many more fancy items. Since, these accessories obscures the prominent facial features, the traditional face recognition systems have not shown a notable recognition performance and thus its performance is challengeable in the various applications fields such as immigration and border control, national security, surveillance, and many more. In literature, IIIT-DDFD, IMFDB, and FDB are disguise datasets are available for Indian ethnicity. Our analysis indicates that, these datasets are not sufficient with enough facial samples to meet the current trends. Therefore, we have introduced an Indian celebrity disguise face dataset (ICDFD), which includes the samples with wide range of complex disguise variations combined with pose, illumination, and expression. Initially, we analyze the performance of these datasets using holistic approaches and followed by deep learning models. From the experimental analysis, it reveals that the deep learning models have shown an optimal performance over holistic approaches. It is observed that, the disguised faces are continue to pose wide open challenges for the researchers in the area of computer vision.
Authors - Harini S, Shamila Ebenezer A Abstract - The document introduces a protected blockchain application for patient care that improves system visibility and operational speed. The application allows patients to book appointments which hospital administrators check before authorization. Following permission by administrators, patients may evaluate their appointment schedule and doctors analyze medical records for medical assessments. Specialist approval must authorize the download of prescription reports. Through blockchain technology physicians can perform decentralized transactions as well as track assets and exchange secure data which leads to lower operational costs and better trust and platform collaboration between patients and specialists and administrators.
Authors - Vishwesh Kumbhre, M. L. Dhore, Pradhynan Lohar, Ajinkya Lende, Om Popade, Shivkumar Padalwar Abstract - Augmented Reality has contributed in many important fields like Education and other fields like Medical and Military, which needs 100% accuracy and focus while performing certain operations. These tasks can be perfected if you have a thorough practice of every situation and gain knowledge about everything being used in that instance. This requirement is fulfilled by Augmented Reality where it creates virtual objects on a real-world background which gives us a real time experience, as if we are actually performing these tasks using solid objects. And the responses are as legit as they would be, in real experiments. Using these features, we can make different software which guides the students and cadets in real life situations.
Authors - Anjali Mahavar, Atul Patel, Ashish Patel Abstract - When given to the human body, different substances and products might present varying health hazards. Concerns regarding toxicity have caused fewer new medicines to access the market throughout the years via the conventional drug development path. The choice of lead compounds and ADMET research depends much on the use of in silico toxicity prediction techniques as ethics, time, money, and other resources often limit in vitro and in vivo approaches. In this regard, we propose a variety of toxicity tools that use structural and physicochemical- based characteristics in the form of molecular descriptors and fingerprints to assess the carcinogenicity of five distinct imidazo[1,2,a]pyridine ligands, including Protox-3, VenomPred, PkCSm, etc. According to the results of the in silico toxicity tool, ligand-1 (IP-1) has a low likelihood of carcinogenicity (0.56%) and excellent accuracy (67.38%) when compared to other ligands. As a consequence, it can be the first choice for medication development in the treatment of cancer.
Thursday August 27, 2026 12:30pm - 2:30pm IST Virtual Room AGOA, India
Authors - Atharv Khunte, Vishakha Dhotare, Divya Pawar, Nikita Dandgavhal, V.M Kokane Abstract - For increasing cyberattacks, web applications require robust and scalable security mechanisms. We suggest a centralized firewall structure that effectively detects and blocks attacks on multiple hosts simultaneously. The system is a command center that collects information about attacks from clients not yet attacked. Information collected is transmitted to the afflicted client only after receiving all necessary details. The system works by detecting and neutralizing various forms of cyberattacks such as SQL injection (SQLi), cross-site scripting (XSS), and distributed denial-of-service (DDoS) attacks. As soon as malicious activity is detected, the system will automatically block the attacking IP address, preventing further intrusion. This process enhances real-time protection, limiting the likelihood of repeated cyberattacks on interconnected web applications. By employing a single defense method, this centralized firewall maximizes threat intelligence sharing, rendering all the connected clients secure. Unlike standalone firewalls in the past, this approach consolidates security policies and enhances cybersecurity resilience on various platforms. Further, this system not only secures web applications against emerging threats but also ensures that organizations meet cybersecurity compliance requirements by hosting a neat and responsive security mechanism. The centralized structure of the fire-wall provides early attack detection, largely reducing downtime, data loss, and monetary loss.
Thursday August 27, 2026 12:30pm - 2:30pm IST Virtual Room AGOA, India
Authors - Prajakta Prasad Kohale, Michael Savariapitchai Abstract - Communication is fundamental for quality healthcare. It is the bridge between the patients and the provider. The basis of any group teamwork and an important factor in an efficient healthcare system is communication. The objective of this paper is to trace the history of communication’s evolution from basic, traditional models to the complex systems that are present in modern-day healthcare. Effective communication helps in establishing trust and confidence which motivates both the patients and the care teams to work together and take accountability for their health. The most favorable health outcomes are shown in patients who feel most sincerely cared for.
Authors - A.Punidha, E.Arul, E.Yuvarani, S.Rajasakaran Abstract - Device drivers are a critical component of modern computing but are increasingly targeted by attackers to gain unauthorized access, execute malicious code, or escalate privileges. Traditional malware detection techniques, such as signature-based and heuristic methods, struggle against advanced threats that employ evasion tactics. To address this, we propose a Graph Neural Network (GNN)-based framework that leverages feature engineering and graph-based learning to detect malicious drivers with high accuracy.By modeling system execution as a graph, our approach captures complex dependencies between API calls, memory accesses, and kernel interactions. We employ Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT) to analyze these relationships, enabling detection of even stealthy and obfuscated malware.Experiments on Windows, Linux, and Android driver datasets demonstrate that our model achieves a 95.8% accuracy, outperforming Random Forest, XGBoost, LSTM, and CNNs. The model is also robust against adversarial evasion techniques, making it a scalable and effective solution for endpoint security, malware sandboxing, and kernel protection.
Thursday August 27, 2026 12:30pm - 2:30pm IST Virtual Room AGOA, India
Authors - R Saiprithvi, Sindhu Chandra Sekharan, Summia Parveen, Ajanthaa Lakkshmanan, Jesline D Abstract - Diseases of lungs like asthma, Chronic obstructive pulmonary disease, lung cancer are among the leading causes of death across the world. Ensuring better outcomes for patients with any medical condition requires an early diagnosis which is, unfortunately, technologically impossible in many regions. This work presents a multimodal Conversational Artificial Intelligence approach based on X-ray imaging, respiratory sound processing, and conversational interfaces that can help with the early and easy diagnosis of lung health. A Convolution Neural Network Processes X-ray images and reliably captures abnormalities. A Random Forest classifier examines MFCC features of lung sounds to confirm presence of asthma, bronchitis, and other diseases. The last method involves using a conversational AI chatbot that makes the collection of symptoms more convenient and provides the user with further information. with the capture of volumetric imaging, sound, and text, healthcare accessibility, efficiency, and diagnostics can be achieved using Artificial intelligence in imaging technology. Lung health is of primary importance, but millions of people worldwide live with easily preventable respiratory disease. EWHO alone estimates that more than 300 million people around the globe suffer from chronic lung disorders, including asthma, Lung
Authors - Job Joseph, Shivaprakash S, Rahul, Rojalin Patri Abstract - This research investigates the impact of financial influencers (finfluencers) on investment decisions of students using trust, perceived risk, and ethical concerns. Correlation and regression analysis reveal strong inter-linkages among them. The implications are drawn noting students' increasing utilization of social media as a source of personal finance information and both its benefits and risks. This work informs financial literacy scholarship and provides recommendations for policy change to contain the influence of finfluencers. Additionally, findings from current research show that finfluencers are not only educators but also business entities that act with self-interest, influencing market behaviour and investment decision-making among retail investors.
Authors - Dheeraj Hegde, Aishwarya Kalatippi, Prajwal Shiggavi, Satish Chikkamath, Nirmala S.R Abstract - This study presents a novel approach to image representation, utilizing wavelet transforms to compress image information into a compact latent space. Wavelet transforms offer a multi-resolution analysis, decomposing images into different frequency components at different resolution scales, emphasizing the spatial and frequency attributes. By leveraging the hierarchical structure of wavelet coefficients, we construct a latent space representation preserving essential features while reducing dimensionality. Experimental evaluations on benchmark datasets demonstrate competitive performance in tasks such as compression and classification compared to traditional deep learning approaches. Waveletbased representation offers promise for addressing challenges in highdimensional data while retaining crucial image information for diverse processing tasks.
Authors - SnehalBalasaheb Salve, Harsha Bhute Abstract - Developing a precise and strong model for forecasting energy utilization is importantaim for the management and functionality of smart buildings. The previous studies have researched different models for forecasting various load prediction schemes. The combined effects regarding data enrichment and machine learning approach in energy predictions have not been fully examined. This research proposes a novel approach, an ensemble model enhanced by generative adversarial networks (GANs) for predicting the usage of energy in big buildings that are commercial. This combined system integrates various single models using ensemble method with stacking. Furthermore, a GAN is utilized to capture the distribution of samples from the main dataset, generating top-notch specimens to augment the dataset from the training data. This expanded dataset allows the model to train with a wider range of samples, increasing its resilience. The experimental series evaluate the method that is proposed, using three variants of GAN and assessing performance with metrics such as mean absolute error, root mean square error, and coefficient of variation of root mean square error. This proposed approach demonstrates practical results that develop a model for power utilization prediction in application of real world.
Authors - G Roopa, H.L.Suresh Abstract - The work outlined here provides a new method to handle fault of Power Electronic Traction Transformer (PETT) switch using reverse charging Cascaded H Bridge (CHB) and Dual Active Bridge (DAB) topologies. Precisely, the main purpose is to improve the speed of detection, determining the location, and recovery of faults from the existing system using feature extraction and Machine Learning algorithms. The traditional approaches in achieving fault tolerance are defective in detecting faults in good time, isolating faults inadequately, and using backup hardware. In order to solve these problems, the proposed methodology actively reassigns control signals to backup modules resulting in the exclusion of faulty elements while preserving a stable system performance with moderate loss in efficiency. The feasibility of the suggested approach is confirmed through simulation outcomes for fault detection precision, which is increased to 98 percent; the fault localization time of at most 5-10 ms; and system throughput of 5-8 percent. Furthermore, the work investigates how CHB and DAB function in fault conditions and enshrine a novel reverse charging method for maintaining the DC voltage of the redundant module. The startup process of the PETT system is also managed with optimization of voltage and transient, which leads to enhance the general system initialization. Besides increasing the dependability and fault tolerance of PETT systems, the above methodology also reduces the system’s embedded hardware duplication and elevates system performance and scalability , which consequently leads to the decrease of the total system cost by 15 percent. These results point out that the proposed solution has potential for the development of the next generation of fault-tolerant power electronic systems.
Authors - Santushti Betgeri, Rohit Rathod, Sakshi Rathod, Sanskar Raut, Anisha Sadanshiv Abstract - AUTOHUB is an integrated solution for essential services regarding vehicles, as well as vehicle expenses. This is a solution with both a native Android app and the response web interface that can easily integrate, especially with dynamic time slot selection for service bookings and an extensive expense tracker for fuel, repairs, tolls/fines, and others. It also has a cloud Online Document Manager for safe document storage, automatic expiration reminders, different dashboards for service providers, and more. It is developed using Android Studio, React, Node.js with Express, Firebase Firestore for real-time data sync, and Razorpay for secure payment processing. The platform has a modular microservices architecture and is scalable and easy to maintain. This integrated solution not just tackles the current challenges posed by fragmented automotive service management, but it also builds the base upon which future extensions can be realized, placing AUTOHUB well ahead of its time as a platform for automotive care.
Authors - Ashwini Matange, Harsha Talele, Pratik Nagare, Vineet Morankar, Aniket Gavkare, Moin Shaikh Abstract - Accurate segmentation and prognostication of brain tumors are critical for effective diagnosis, treatment planning, and patient management in glioma. In this work, we present a unified framework built upon the BRATS2020 challenge data that integrates deep learning-based segmentation with radiomics and machine learning for overall survival prediction. First, we employ a 3D-UNet architecture to perform robust segmentation of brain tumors from multi-modal MRI scans, achieving a mean Intersection over Union (IOU) of 86%. This segmentation not only delineates tumor sub-regions effectively but also provides the basis for subsequent feature extraction. Leveraging the pre-trained 3D-UNet, we extract deep features from the MRI scans, and in parallel, perform radiomics feature extraction on the corresponding tumor masks. These features are then combined with clinical and demographic data provided in the BRATS2020 challenge dataset. A random forest classifier is subsequently trained on this comprehensive feature set to predict overall patient survival, achieving a classification accuracy of 70% in stratifying patients into survival categories. Our approach builds on recent advances in brain tumor segmentation—incorporating ideas such as ensemble learning, multi-modal imaging, and uncertainty quantification—to enhance both the segmentation accuracy and prognostication performance. The promising results demonstrate that the integration of deep learning segmentation with radiomics and traditional machine learning methods can serve as a robust tool for personalized treatment planning and risk stratification in glioma patients.
Authors - Vaishnavi Moorthy, Jagadeesan Moorthy, Shubhradip Saha, Anshuman Kumar Abstract - In the medical field, the readability of important information on medicine packs, like the expiry date, is of prime importance for maintaining patient safety. Yet, a number of reasons like damage, blurring, and printing defects may hide this important information on medicine strips. To solve this problem, we suggest a deep learning-based solution for medicine strip denoising and enhancement, making important information such as expiry dates more legible. Our approach utilizes image denoising, specifically designed to correct blurry or partially readable expiry dates on the packaging of medicines. This solution not only helps healthcare workers and patients validate medicines but also makes a contribution to the pharmaceutical industry.
Authors - Shweta Kumar, Saru Dhir, Ashish Kumar Mourya Abstract - Healthcare informatics has many difficulties due to the complexity of varied medical data. Clinical notes, imaging, and genomic data are instances of unstructured data that is more flexible and has more depth than organized data, such as digital records, which are easier to use. Combining different healthcare data sources is difficult due to interoperability issues and semantic variability. Despite the emergence of standardization projects such as HL7 (Health Level 7) FHIR (Fast Healthcare Interoperability Resources) and SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms), inefficient processes and unreliable vocabulary continue to impede seamless communication of information. The dispensation of natural language, or NLP (Natural language processing), methods enable the extraction of important information from uncontrolled health information. Furthermore, instantaneous data analysis and scalability are enhanced by online computing, and blockchain technology is being investigated as a safe, independent method of sharing medical data. This study examines the challenges of managing a variety of healthcare information as well as the possible benefits of contemporary technologies. Future research focuses on improving interoperability frameworks, developing AI-driven data analysis, and ensuring confidentiality and security of data in order to provide effective and data-driven healthcare options.
Authors - Arun N, Rithika J Prabhu, DHANYA M Abstract - Sustainability in India has been become a driving force behind the growth of the green energy sector and the economy's transition to cleaner energy. The research paper investigates the use of machine learning models to predict stock prices of green energy companies in India. It deliberates on the rapid growth of the green energy market and the potential for ever-advancing technologies making accurate prediction in finance for supporting the nation's sustainable development goals. Using machine learning, it generates useful insight for stock performance for the benefit of investors and policy makers in arriving at decisions.
Authors - M. Chaitanya Raju, Maddu Reshma, V. Anvesh, Lekha S. Nair Abstract - Waste classification and management are important for healthier planet Earth. In this paper we are proposing an integrated approach for waste detection and classification using object detection along with natural language processing (NLP) techniques. which introduce a YOLO-based model to detect and classify waste in images by using Bootstrap Language-Image Pretraining (BLIP) for scene understanding and contextual analysis. The workflow involves, feeding the waste images into a preprocessing stage (image), captioning image data with Natural Language Processing (NLP) to produce descriptive captions, and analyzing the textual features of detected captions that exist in the waste (waste elements). The classification of the detected object is performed by a custom trained YOLOv8 model which is fine-tuned on a specific waste class dataset. Experiments show that the model recognizes garbage, recyclables and litter with high accuracy. This system showcases the potential of combining visual and textual modalities to enhance waste detection accuracy, offering a robust tool for automated environmental monitoring and management.
Authors - Juttiga Rohita, B Teja Sree, Ibrapatnam Anusha, Mohammad Sharmila Begum, Nirjogi Mahathi Abstract - Social media platforms have become increasingly vulnerable to online threats, making safeguarding the internet an increasingly difficult task. Why? This project showcases an artificial intelligence-powered system that can detect and filter out inappropriate text and images in real-time. Machine learning and natural language processing (NLP) are utilized by the system to detect hate speech, toxic terminology such as slang, and explicit imagery while maintaining document integrity. TF-IDF, LSA, and Word Embeddings are utilized in text filtering to improve the understanding of context. In image filtering, deep learning models using convolutional neural networks (CNNs) and pre-trained NSFW classifiers detect and remove explicit content. This balances scale with accuracy and provides a robust, automated content moderation system that improves both safety and compliance on the Internet.
Authors - Shahedhadeennisa Shaik, Abhinav R B, Chaitra S P, Sagari S M Abstract - Video traffic surveillance has become an essential tool for various applications, including security, transportation planning, and traffic management. Recent advancements in deep learning have opened new possibilities for enhancing the performance of vehicle detection and tracking in these systems. This paper addresses the challenges of online action detection in surveillance scenarios by focusing on enhancing multi-object tracking (MOT) performance. Recognizing the limitations of current MOT methods in handling real-world surveillance complexities, we propose a methodology that integrates appearance model extraction directly from the object detector, adaptive adjustments of confidence thresholds and input resolutions, and the incorporation of color information into ReID embeddings. We aim to bridge the gap between motion-based and ReID-based tracking methods, improving both speed and accuracy. Our proposed techniques, including scene-based and object-based adaptation through reinforcement learning, and advanced feature fusion for ReID, are designed to enhance robustness and efficiency. We evaluate our methodology using publicly available datasets, focusing on surveillance-specific challenges. The enhancement in MOT performance is challenging and paving the way for more reliable and efficient surveillance system.
