| S.No |
Name of the project |
Abstract |
| 1 |
DESIGN AND DEVELOPMENT OF WEARABLE ANTENNA FOR PROACTIVE TUMOR DETECTION. |
In recent years, wearable electronics have gained opportunities and the past decade has become evidence of this growth in Wireless Body Area Network (WBAN). They fulfil the requirements of personalizing healthcare, communication, patient monitoring, tracking, and rescue operations. The major challenge for the WBAN is to handle the coupling of the radiator with the human body. To increase the performance of a microstrip patch antenna, it has been blended and truncated on the diagonal sides to make up the suggested antenna design. Flexible electronics have paved the way for Wireless Body Area Networks (WBAN). This allows for optimal performance. The square patch, measuring 50 × 50 mm, resonates at the 2.4 GHz ISM band, which is frequently utilized in wireless communication applications. The FR-4 substrate, which has a relative permittivity of 4.3, was selected for the antenna. It is a lossy material with a thickness of 1.6 mm and dimensions that match those of the ground plane (50 x 50 mm). Perfect Electric Conductor (PEC) is the material used for the patch. Its dimensions are 30 × 30 mm, or half the wavelength at the resonance frequency to further improves performance, a slot is added to the patch’s center. A secondary ground plane measuring 90 x 90 mm and 2 mm thick is placed beneath the primary ground plane in addition to the main ground plane to improve radiation efficiency and antenna stability. In order to maximize signal strength and coverage, the coaxial feed in the design is carefully calibrated to improve gain. The design seeks to improve impedance matching, radiation pattern, and overall antenna performance by truncating and blending the square patch antenna in addition to adding a slot and an extra ground plane. In order to maximize efficiency and reliability and satisfy the criteria of the 2.4 GHz ISM band, much consideration is given to the choice of materials and dimensions. The proposed antenna is designed and simulated using CST Microwave Studio software. The proposed antenna is an efficient antenna with realized gain of 2.88dBi, low VSWR and wide bandwidth. |
| 2 |
AUGMENTATION OF EMOTIONS – A RESERVOIR COMPUTING BASED EEG ANALYSIS |
Emotion classification based on Electroencephalogram (EEG) signals holds significant potential for enhancing affective computing applications. The proposed work developed an effective emotion classification system utilizing EEG signals and addressed the need for robust methods capable of accurately capturing temporal aspects of EEG data to discern distinct emotional states. Leveraging the capabilities of GRU reservoir computing, the machine learning model was trained and evaluated to classify EEG signals into distinct emotional categories. Through rigorous experimentation and evaluation, the performance of the proposed methodology was assessed using standard metrics such as accuracy, precision, recall, and F1-score. The findings demonstrate a higher classification accuracy of 97.344% with the effectiveness in addressing the challenges associated with the temporal dynamics and non- stationary of EEG data. |
| 3 |
DETECTION OF LUNG DISEASE USING BREATHE RATE ANALYSIS |
This work presented a new approach for the early detection of lung diseases using respiratory rate. The real-time monitoring system is developed using a comprehensive sensor system that includes GSR and heart rate sensors, Node MCU and Internet of Things (IoT) technology. The GSR sensor records physiological changes in skin conductance, while the heart rate sensor measures heart rate variability, providing valuable information about breathing patterns. This data is seamlessly integrated and transmitted in real time via the Node MCU, enabling remote monitoring. The proposed system provides a non-invasive, real-time monitoring, cost-effective and efficient way for early detection, facilitating timely treatment. IoT integration ensures accessibility and immediate response with a help of Machine learning model to get an accuracy of 94%, And also, the detection of lung diseases using X-ray images, CT scan images, and CNN technology is a much costlier and time-consuming complex procedure. To avoid these methods, we use a novel and multisensory approach to detect lung diseases easily and cost-effectively. This method helps increase the potential to develop preventive health strategies in the field of various lung diseases such as asthma, pneumonia, chronic obstructive pulmonary disease, and pulmonary fibrosis |
| 4 |
DESIGN AND DEVELOPMENT OF AN E-NOSE FOR THE DIAGNOSIS OF PULMONARY DISEASES |
The Electronic Nose model introduced a new machine for the diagnosis of lung diseases that changed the field with its approach. The model used artificial intelligence technology, machine learning technology, and the integration of unique medical data to provide a solution to the real and counterintuitive lung problem. By combining the nasal cavity with standard measurements, biochemical markers, and patient-specific clinical parameters, the model provided a comprehensive assessment of lung health, helping to support early diagnosis and self-treatment strategies. Rigorous clinical studies and quality control ensured the reliability and validity of the models in real-world clinical settings, promising to increase accuracy and improve patient outcomes. The integration of the system with existing treatments and its focus on personalized treatments further increased the effectiveness and efficiency of the treatment. The e-nose service represented a significant advance in lung diagnostics and revolutionized the challenges faced by doctors and patients. The model had the ability to increase the accuracy of diagnosis, improve treatment strategies, and improve overall patient care, making it a useful tool in the fight against lung diseases. Through continued research, implementation, and effort, the e-nose program had the potential to revolutionize the field of pulmonary medicine and make a positive impact on the lives of countless people affected by these diseases. |
| 5 |
SPEAK-EASY CLOUD BASED STAMMERING ASSISTANT WITH FREQUENCY MATCHING |
Stuttering, also known as stammering or childhood-onset fluency disorder, presents significant challenges to individuals, affecting the fluency and flow of speech. While it is common among young children as they develop their language skills, it can persist into adulthood for some individuals, impacting their self-esteem and social interactions. This project focuses on the development of a speech therapy device tailored specifically for individuals who stutter. The device aims to improve fluency and communication skills by providing real-time feedback and assistance during speech. By utilising innovative technologies, including speech recognition and artificial intelligence, the device detects instances of stuttering and offers interventions such as delayed auditory feedback or visual cues to help individuals overcome speech disruptions. Moreover, the device incorporates personalised learning algorithms to adapt to the user’s unique speech patterns, maximising effectiveness. Through rigorous testing and refinement, this speech therapy device holds promise in alleviating the challenges associated with stuttering, empowering individuals to communicate confidently and effectively. |
| 6 |
NEURAL NEXUS: PROGNOSTICATION OF ANESTHESIA THROUGH MACHINE LEARNING MASTERY |
This proposed work explored the application of machine learning to personalize and optimize anaesthesia administration, addressing the challenge of patient variability in response to anaesthesia. Machine learning algorithms were employed to analyse extensive datasets encompassing patient information, medical history, and surgical details. These analyses were used to develop personalized anaesthesia prediction models, tailoring medication regimens to individual patient characteristics. The work aimed to minimize risks associated with unpredictable patient responses. By understanding individual nuances through trained algorithms, safer and more effective anaesthesia regimens could be achieved, potentially leading to improved patient outcomes and recovery. Systems continuously analysed data during surgery, adjusting predictions based on evolving conditions. This enhanced patient safety and provided dynamic decision support for healthcare professionals throughout the procedure. Seamless integration with existing healthcare Information systems and electronic health records was crucial. User-friendly interfaces facilitated widespread adoption among healthcare professionals. Integrating machine learning in anaesthesia prediction had the potential to revolutionize patient care. Personalized predictions, improved response to variability, real-time adaptation, and interoperability all contributed to safer procedures, potentially reduced healthcare costs, and an overall higher quality of care. |
| 7 |
DRIVER DROWSINESS DETECTION SYSTEM |
The Driver Drowsiness Monitoring System (DDMS) is an essential automotive safety innovation designed to combat drowsy driving accidents. It utilizes a combination of sensors and advanced algorithms to continuously monitor the driver’s condition, detecting signs of drowsiness in real-time. Key components include infrared cameras, facial recognition technology, steering angle sensors, and biometric sensors like heart rate monitors. These sensors provide a comprehensive view of the driver’s behaviour and physiological state. The DDMS software analyzes data from these sensors, assessing parameters like eye movement, blink frequency, facial expressions, and steering behaviour. When signs of drowsiness are detected, the system issues audible and visual alerts, preventing accidents caused by driver fatigue. This system is adaptable to various vehicles and driving conditions, making it a vital tool for enhancing road safety by reducing drowsy driving-related accidents. |
| 8 |
IMPLEMENTING IOT MONITORING SYSTEM FOR NEONATAL CARE: COMBATING RETERM BIRTH CHALLENGES |
Premature infants face significant challenges in regulating their body temperature, often leading to serious health complications. In developing countries, where medical resources are limited, addressing this issue effectively while keeping costs low is crucial. This proposed work approach to incubator design aimed at providing essential monitoring and control functions for premature infants at an affordable price point. The proposed system incorporates various sensors such as temperature, humidity, light intensity, CO2 levels, galvanic skin response (GSR), heart rate, and oxygen saturation (SPO2), allowing for comprehensive monitoring of the infant’s environment and vital signs. Reducing infant mortality rates associated with prematurity-related complications, which currently account for a significant portion of global infant deaths. The proposed infant incubator system integrates various sensors to monitor vital parameters such as temperature, humidity, oxygen saturation, and skin conductance. These sensors, coupled with an Arduino microcontroller, facilitate real-time monitoring and control of the incubator environment. Additionally, the inclusion of a Bluetooth module enables seamless communication with caregivers, providing timely alerts in case of any abnormalities. This proposed work prioritizes usability and cost-effectiveness without compromising on the quality of care provided to premature infants. By harnessing the power of innovation and collaboration, we aim to make life-saving medical equipment accessible to all, regardless of economic constraints. |
| 9 |
SKIN CANCER CLASSIFICATION AND SEGMENTATION USING ARTIFICIAL INTELLIGENCE TECHNIQUES |
Skin cancer is one of the most prevalent types of cancer worldwide, and early detection plays a crucial role in improving patient outcomes. Artificial intelligence (AI) techniques have shown promising results in skin cancer classification and segmentation tasks, providing automated and accurate solutions. This proposed work presents an overview of skin cancer classification and segmentation using AI techniques. The classification aspect focuses on distinguishing between benign and malignant skin lesions. Various AI algorithms, including machine learning and deep learning models, are explored for this task. The segmentation aspect addresses the precise delineation of skin lesions from surrounding healthy tissue. AI techniques such as convolutional neural networks (CNNs) and image processing algorithms are utilized for accurate lesion segmentation. The integration of classification and segmentation techniques is explored, and it allowed for a comprehensive analysis of skin cancer. Combined approaches enable accurate identification of cancerous regions within an image, aiding in precise diagnosis and treatment decisions. This paper provides an overview of skin cancer classification and segmentation using AI techniques. It highlighted the potential of AI in improving skin cancer diagnosis and treatment, and discusses the challenges and future directions in the field. Convolutional Neural Networks (CNNs) excel in extracting features from raw data like images, enabling automated learning of hierarchical representations. Their spatial hierarchy and shared weights make them powerful for tasks such as image classification, object detection, and even natural language processing tasks involving sequential data. With the help like semantic segmentation or instance segmentation, CNNs can accurately partition and identify specific areas of interest within complex visual data. CNNs sometimes referred to as convnets use principles from linear algebra, particularly convolution operations, to extract features and identify patterns within images. |
| 10 |
DIABETES PREDICTION WITH MICROVASCULAR IMAGING AND MULTIMODAL BIOMARKERS |
Addressing the global health challenge of diabetes necessitated overcoming limitations in conventional blood tests, which often deterred regular testing due to their invasive nature and discomfort. To enhance diabetes detection accuracy, this study proposed a two-stage Machine Learning (ML) system. In the initial stage, various ML models, including Logistic Regression, Random Forest, Decision Tree, and K-Nearest Neighbors, analyzed user-provided health data such as pregnancies and glucose levels. These models discerned patterns and relationships to predict diabetes risk, achieving notable accuracy rates ranging from 74.25% to 99%. However, further investigation was warranted to tackle potential overfitting observed in the KNN model. Subsequently, Stage 2 offered a non-invasive confirmation process. Users flagged at high risk in Stage 1 could opt for retinal image capture, processed using OpenCV to classify diabetic retinopathy features. Integrating both stages furnished a holistic risk assessment, promising earlier and more accurate diabetes detection. By capturing a broader array of relevant features beyond blood tests, this approach enabled improved preventative strategies and personalized treatment, ultimately alleviating the burden of diabetes on individuals and healthcare systems alike. |
| 11 |
APPLYING DEEP LEARNING TO ACCESS HEART ARRYTHMIA BY ANALYZING ECG |
Arrhythmia classification plays a pivotal role in the early detection and management of heart disease, a leading cause of mortality worldwide. Electrocardiogram (ECG) data, being a widely adopted physiological measurement, serves as the cornerstone for classification endeavors in this domain. Deep learning models have emerged as powerful tools for categorizing arrhythmia classes, exhibiting promising results in automated diagnosis. However, the efficacy of these models is often gauged solely based on performance metrics, which may not fully account for the intricacies of real-world data distributions. The distribution of records within each dataset category profoundly influences the validation of these metrics, especially in scenarios where imbalanced data is prevalent. The disparity between balanced and imbalanced data can significantly impact the assessment of deep learning model performance, potentially leading to skewed outcomes. In light of these challenges it gives a thorough analysis of various architectures of deep learning models for arrhythmia classification, with a keen focus on elucidating pre-processing methods tailored to mitigate the effects of imbalanced data. By delving into the intricacies of Convolutional Neural Networks (CNN), Artificial Neura Networks (ANN), Deep Neural Networks (DNN), AlexNet, ResNet, k-Fold Cross-Validation, VGG-16, Long Short-Term Memory (LSTM), and LeNet, this aims to provide insights into optimizing model performance and assist data imbalances |
| 12 |
ENHANCED PATIENT HEALTHCARE SOLUTION |
This proposed work aimed to revolutionize healthcare delivery, particularly catering to elderly individuals and those with mobility impairments. The work integrated both software and hardware components to provide a seamless and efficient healthcare experience. On the software front, the work utilized a modern web stack comprising HTML, CSS, and React for frontend development, Firebase for robust database management, and Node.js for flexible backend operations. This software infrastructure powered a user- friendly website, “Unified Surgical Telehealth”, which served as the central hub for all healthcare interactions. The hardware aspect of the work involved the creation of a wearable device utilizing advanced components such as microcontroller, AD8232, MLX90614, and MAX30102. These components enabled the device to monitor vital physiological parameters, including heart rate and body temperature, essential for comprehensive health assessment. Central to the work’s functionality is its ability to facilitate remote consultations between patients and healthcare providers. Through live video consultations, patients could connect with their doctors in real-time, eliminating the need for physical travel and ensuring timely access to medical assistance. This feature is particularly beneficial for individuals with limited mobility or those undergoing post-operative care. During remote check-ups, the wearable device can collect crucial health data from the patient, which is securely stored in the Firebase database. Healthcare providers could access this data through the website, enabling them to monitor patient’s health status remotely and make informed medical decisions. |
| 13 |
MACHINE LEARNING ENABLED CARDIAC STROKE ALERT |
The Heart Stroke Alert System emerges as a revolutionary advancement in healthcare technology, poised to redefine the landscape of cardiac care with its proactive approach. Through the seamless integration of Internet of Things (IoT) technology, healthcare providers gain unprecedented insights into patient health statuses, facilitating rapid response to critical alerts. Central to this system are wearable sensors that continuously monitor temperature and heart rate fluctuations, providing vital streams of data for analysis. These data undergo meticulous scrutiny by machine learning models, trained extensively on relevant datasets, to not only identify cardiac events but also to delve into patient-specific attributes such as age, gender, medical history, and physiological markers. By synthesizing this multifaceted information, the system offers personalized care strategies tailored to individual cardiovascular health profiles. The primary objective of this innovative approach is to optimize patient outcomes through early detection and preemptive intervention, transcending traditional reactive models in favor of proactive patient management. Moreover, the Heart Stroke Alert System extends its utility beyond acute event detection, offering a comprehensive solution for continuous remote patient monitoring. Through the integration of sensor technologies, machine learning algorithms, and IoT connectivity, it establishes a robust framework for delivering holistic cardiac care. |
| 14 |
REAL TIME ANALYSIS OF PARKINSON TREMOR USING DATA DRIVEN TECHNIQUES |
Parkinson’s disease (PD) is characterized by motor symptoms such as tremors, rigidity, and gait difficulty, which can significantly impact patients’ quality of life. Traditional methods of evaluating these symptoms rely heavily on subjective patient self-assessment, often resulting in incomplete or inaccurate data. To address this limitation, we proposed a novel approach utilizing wearable accelerometers integrated into a watch for continuous monitoring of PD motor symptoms. This project proposes a monitoring system for Parkinson’s disease (PD) that integrates accelerometers and machine learning. Machine learning algorithms, particularly Support Vector Machine (SVM), analyze this data to classify symptom severity accurately. It provides a more objective and quantitative assessment of PD symptoms, improving diagnostic accuracy and treatment planning. Machine learning enables personalized treatment strategies tailored to individual patients’ symptom profiles, optimizing therapeutic outcomes. |
| 15 |
CYPHER FEEL A SENTIMENTAL COMPANION |
The pervasive integration of technology into our daily lives has led to an unprecedented surge in the generation and storage of personal data, exposing individuals to risks of unauthorized access and breaches. To address this concern, this project introduces “Cipher Feel,” a web application designed not only to provide robust data storage through end-to-end encryption but also to understand user’s current emotional states. Unlike conventional diary applications, Cipher Feel integrates sentimental analysis to comprehend users’ emotions, offering personalized suggestions based on their previous inputs for enhancing well-being. Sentiment analysis refers to identifying as well as classifying the sentiments that are expressed in the text source. The user’s previous text data are useful in generating a vast amount of sentiment data upon the analysis. These data are used to understand the current user’s emotional state and helps to provide the better suggestions for the user. This project report outlines Cipher Feel’s development process, features, security considerations, and the integration of emotional analysis, emphasizing the importance of secure data storage practices in today’s digital landscape. Readers will gain a comprehensive understanding of Cipher Feel and its role in promoting secure data storage practices in our technologically advanced era. |
| 16 |
SKIN DISEASE PREDICTON USING IMAGE PROCESSING |
Skin diseases are most common among the globe, as people get skin disease due to inheritance, environmental factors. In many cases people ignore the impact of skin disease at the early stage. In the existing system, the skin disease is identified using biopsy process which is analyzed and medicinal prescribed manually by the physicians. To overcome this manual inspection and provide promising results in short period of time, we propose a machine learning technique. For this the input images would be microscopic images i.e., histopathological from which features like color, shape and texture are extracted and given to convolutional neural network (CNN) for classification and disease identification. Our objective of the project is to detect the type of skin disease easily with accuracy and recommend the best and global medical suggestions. This paper proposes a skin disease detection method based on image processing and machine learning techniques. In existing approach, the increased skin diseases identified at the later stage using biopsy only. Thus, this process is performed manually which can lead to human errors and takes 1-2 days for the results. Also, the physician finds it difficult to identify the type of skin disease andthe stage of disease at the analysis stage. Thus, making the medicine prescription difficult. This concern can be addressed by usage of machine learning and deep learning techniques by analyzing the microscope image. This approach can provide a promising result by combining computer vision and machine learning techniques. The proposed methodology system is highly beneficial in rural areas where access to dermatologists is limited. For this system, we use PyCharm based python script for experimental results. These results suggest that the proposed system can help effectively diagnose the type of skin disease, thereby reducing further complications. The HAM10000 dataset is used and the proposed method has outperformed other methods with more than 85% accuracy. |
| 17 |
HEALTHCARE AND MEDICINE RECOMMENDATION SYSTEM USING NLP |
In the era of data-driven healthcare, the demand for personalized medical recommendations is paramount. Since corona virus has shown up, inaccessibility of legitimate clinical resources is at its peak, like the shortage of specialists and healthcare workers, lack of proper equipment and medicines etc. The entire medical fraternity is in distress, which results in numerous individual’s demise. Due to unavailability, individualsstarted taking medication independently without appropriate consultation, making the health condition worse than usual. As of late, machine learning has been valuable in numerous applications, and there is an increase in innovative work for automation. This paper intends to presenta drug recommender system that can drastically reduce specialist heap. In this research, we build a medicine recommendation system that uses patient reviews to predict the sentiment using various vectorization processes like Bow, TF-IDF, Word2Vec, and Manual Feature Analysis, which can help recommend the top drug for a given disease by different classification algorithms. The predicted sentiments were evaluated by precision, recall, accuracy and AUC score. The results show that classifier Linear SVC using TF-IDF vectorization outperforms all other models with 93% accuracy. Through extensive experimentation and evaluation, our system demonstrates significant improvements in patient outcomes, fostering a more efficient and effective healthcare delivery model. |
| 18 |
CHRONIC KIDNEY DISEASE PREDICTION USING MACHINE LEARNING |
Chronic Kidney Disease (CKD) is a severe condition affecting millions worldwide, with significant mortality rates. Early detection is crucial for effective treatment and prevention of complications. In this study, we propose a machine learning-based approach to predict CKD, aiming to improve early diagnosis and reduce mortality rates. We employ preprocessing techniques to handle missing data and compare the performance of several machine learning algorithms, including K-Nearest Neighbour, Decision Tree, Gaussian Naïve Bayes, Logical Regression, and Artificial Neural Network. Through dataset selection, preprocessing, algorithm execution, and classification of control metrics, we aim to determine the most effective algorithm for CKD prediction. Our objective is to raise awareness, promote early diagnosis, and ultimately mitigate the impact of CKD through early prediction and proper treatment.In this study, we propose to use a dataset containing a wide range of clinical and demographic variables to train and evaluate machine learning models for CKD prediction. We will employ preprocessing techniques to handle missing data and normalize features to ensure the models’ accuracy and reliability. By comparing the performance of various machine learning algorithms, including K-Nearest Neighbour, Decision Tree, Gaussian Naïve Bayes, Logical Regression, and Artificial Neural Network, we aim to determine the most effective approach for predicting CKD |
| 19 |
MENTAL HEALTH MANAGEMENT |
Mental health issues are a growing concern worldwide, with limited access to professional help creating a significant barrier to well-being. This project explores the potential of web app technology integrated with machine learning (ML) to address this challenge. We propose the development of a novel mental health wellbeing app, “Ally – Wellness Companion”. Ally – Wellness Companion leverages ML algorithms to personalize user experience and provide targeted support. Through features like mood tracking, journaling, and interactive exercises, the app aims to promote self-awareness and equip users with coping mechanisms. The ML component analyses user data to identify patterns and suggest relevant resources, personalized recommendations for mindfulness practices, and potential early detection of mental health concerns. Ally – Wellness Companion incorporates an AI-powered chat feature that provides personalized support and insights based on the user’s journal entries. Through natural language processing (NLP) techniques, the AI analyses the user’s journaling patterns and emotional tone. This analysis allows the AI to tailor conversation topics and responses to the user’s specific needs. For instance, if the user frequently expresses feelings of anxiety in their journal, the AI might initiate conversations about relaxation techniques or recommend relevant exercises within the app. The AI can also identify positive entries and offer encouragement, fostering a sense of accomplishment and progress. This personalized interaction fosters a safe space for users to express themselves and receive non-judgmental support, enhancing the overall user experience and potentially leading to a deeper understanding of their own emotions and thought patterns. |
| 20 |
Brain tumor detection using deep networks |
Brain tumor detection plays a crucial role in early diagnosis and treatment planning, significantly impacting patient outcomes. With the advancements in deep learning techniques, automated detection systems have shown promise in enhancing the accuracy and ef iciency of tumor diagnosis. This paper proposes a novel approach for brain tumor detection utilizing deep neural networks (DNNs). The proposed system leverages convolutional neural networks (CNNs) to extract relevant features from magnetic resonance imaging (MRI) scans. Furthermore, a multi- layer perceptron (MLP) is employed to classify the extracted features into tumor and non-tumor classes. Experimental results demonstrate the ef ectiveness of the proposed method in accurately detecting brain tumors, showcasing competitive performance compared to existing approaches. The proposed system not only provides reliable detection but also of ers potential for integration into clinical workflows to aid healthcare professionals in making informed decisions for patient care. |
| 21 |
AUTOMATIC ACCELERATION CONTROL SYSTEM |
Car accident is the major cause of death in which around 1.3 million people die every year. Majority of these accidents are caused because of distraction or the drowsiness of driver. Construction of high-speed highway roads had diminished the margin of error for the driver. The countless number of people drives for long distance every day and night on the highway. Lack of sleep may lead to an accident. Drowsiness and Fatigue of drivers are amongst the significant causes of road accidents. Every year, they increase the amounts of deaths and fatalities injuries globally. To prevent such accidents, we propose a system which alerts the driver if the driver feels drowsy. Facial landmarks detection is used with help of image processing of images of the face captured using the camera, for detection of drowsiness. In this project, a module for Advanced Driver Assistance System is presented to reduce the number of accidents due to drivers’ fatigue and hence increase the transportation safety; this system deals with automatic driver drowsiness detection based on visual information and Artificial Intelligence. This project proposes an algorithm to locate, track, and analyze both the drivers face and eyes to measure EAR (Eye Aspect Ratio), a scientifically supported measure of drowsiness associated with slow eye closure. |
| 22 |
Disease analysis and predictions system using machine learning |
The Disease Analysis and Predictions System (DAPS) represents a groundbreaking solution at the intersection of data analytics and public health. By amalgamating diverse data streams, including electronic health records, environmental indicators, demographic data, and social media activity, DAPS offers a comprehensive platform for disease surveillance and analysis. Through the application of advanced machine learning algorithms, DAPS not only identifies historical trends but also predicts future disease outbreaks with remarkable accuracy. Real-time monitoring capabilities enable prompt response to emerging health threats, while intuitive visualization tools facilitate the interpretation of complex data sets. DAPS stands poised to revolutionize public health practice by empowering stakeholders with actionable insights, facilitating proactive interventions, and ultimately contributing to the advancement of population health outcomes. |
| 23 |
MEDICAL CHAT BOT |
In this project we focus on providing a quick and effective solution to every farmer who is affected with crop damaging pests. Plant diseases cause great damage in agriculture resulting in significant loss in yield, expenditure and farmer’s effort. The recent expansion of deep learning methods has found its application in plant’s disease prediction, offering a robust tool with high accurate results. Our aim is to build a plant disease predicting model using deep learning and deploy it in a website. This model can predict a plant’s disease by seeing its image. And we also provide treatment methods to cure that disease. |
| 24 |
INNOVATIVE APPROACHES TO CANCER INFORMATION: EXPLORING GENERATIVE AI SOLUTIONS |
With the widespread adoption of deep learning and machine learning techniques in healthcare, coupled with the availability of specialized cancer datasets, there has been a surge in research exploring the potential benefits of AI for comprehending the intricate biological mechanisms underlying cancer. This work delves into the emerging AI approaches and their applications in oncology. It also outlines how AI can be leveraged in various aspects of cancer management, such as early detection, prognosis, and treatment administration. Furthermore, this work introduces the use of the latest large language models in oncology clinics, highlighting their potential applications. It provides insights into how these diverse data types can be integrated to create decision-support tools, aiding clinicians in making informed decisions for precision oncology. |
| 25 |
HOSPITAL DATABASE MANAGEMENT SYSTEM USING PHP AND II |
This paper presents a web -based Hospital Database Management System that allows patients, doctors, and administrators to interact with the hospital’s information system through a web interface. The system is built using HTML5/CSS3, JavaScript, Bootstrap, XAMPP, PHP, MySQL, and TCPDF technologies. Web-Based Hospital Database Management System (HMS) enables various hospital and medical processes to be performed online. It consists of registration, login of patients, and booking their appointments with doctors storing their details in the system. It provides a login page for patients, doctors, and admins each have their username and password. Itconsists of three modules. Those are the patient, doctor and admin. This Web Application maintains authentication to accessthe information. Administrators can see patient and doctor information, appointment schedules and add new doctors as part of administrative tasks. A database was created one for the patient and the other for the doctors so that admin can access it. The Patient module includes booking appointments and checking prescriptions. A patient can pay a doctor’s Fee online. The doctor module allows doctors to view appointments, give prescriptions and search for patients. Web based technology provides a wide range of online services in practically every industry. The majority of jobs may be completed online, which helps to minimize the workload, expense, and effort. The paper discusses the concept of a web based platform that would enable various hospital and medical processes to be performed online utilizing Web networking technologies, which could be crucial for implementing the functionality of online medical administration. This will aid in the administration of patients, the management of doctor schedules, and the maintenance of patient data that are accessible throughout the hospital online patient data storage, management, communication, analysis, and updating. Therefore, by implementing this web-based application many tasks that would be time consuming and inconvenient can be accomplished. |
| 26 |
Enhancing healthcare outcomes through machine learning based multi-disease prediction |
Accurate and timely disease prediction is made possible by machine learning techniques, which have completely changed the healthcare industry. Simultaneous prediction of numerous diseases can greatly enhance early detection and treatment, improving patient outcomes and lowering healthcare expenditures. This study examines the use of machine learning algorithms to forecast a variety of diseases, emphasizing the advantages, difficulties, and potential applications. We give a summary of the many machine learning models and information sources that are frequently employed in illness prediction. We also go over the significance of feature selection, model assessment, and combining various data modalities for improved disease prediction. The study’s conclusions demonstrate machine learning’s potential for multi-disease prediction and its possible effects on public health. Once more, I’m using a machine learning model to determine whether or not an individual has a few diseases. This training model trains itself to predict disease using sample data. |
| 27 |
Eye disease detection using deep learning guide |
Eye disorders are the third most prevalent diseases in the world, assuming the positions after cancer and cardiovascular diseases, according to the WHO. Nearly 2.2 billion people worldwide are sight-impaired [1]. Therefore, timely diagnosis and differentiation of eye diseases are necessary to limit their influence on people’s quality of life. This paper describes an original solution to settle this issue in a way of developing an Eye Disease Detection System. Using convolutional neural networks and an open-soure eye fundus image dataset composed of 4 types of eye photos: normal, diabetic, glaucoma and cataract, our system uses cup-to-disc-ratio as a classifier. We use convolutional neural network (CNN) model, named YOLO, known for their speed and accuracy in detecting objects[2]. Through comparison, we observe that YOLOv5 attains an mAP value of 0.985 in eye disease detection, which is a great success as it can be used for quick diagnosis in the medical field. Moreover, the system sets up a user-friendly website that gives out important data and relevant therapy plans for the most common eye diseases, thus helping to save time and to overcome the imbalance between the number of doctors and patients. The project intends to make headways in early diagnosis and management of eye diseases which builds better health services and more accessible health outcomes worldwide. |
| 28 |
Your personalized fitness companion – Multiple exercises with AI-Powered rep counting |
Currently, there has been a boom in demand for personalized fitness solutions, which has necessitated the development of innovative technologies to address these needs. FITLIFE is an all-round fitness companion that offers a collection of over 1000 exercises along with AI-powered rep counting functionality. In this paper, we discuss the architecture, features and technology stack that underpins the development of FITLIFE: A Reacts frontend, NodeJS backend, MongoDB database and Python integration for AI capabilities. This paper describes how users go through authentication, searching for exercise details as well as real-time rep counting according to camera integration. Therefore, FITLIFE not only deepens users’ engagement but also ensures an uninterrupted and effective fitness experience that is consistent with the latest trends in individual health care. |
| 29 |
HEART FAILURE PREDICTION SYSTEM USING MACHINE LEARNING ALGORITHM |
The diagnosis of heart failure in most instances relies on a complex amalgamation of clinical and pathological data. Due to this intricacy, there is a considerable amount of interest among clinical professionals and researchers in efficiently and accurately predicting heart failure. In this project, we have devised a heart failure prediction system that can aid medical professionals in forecasting the heart failure status based on patients’ clinical data. Our methodology encompasses three steps. Firstly, we carefully select 13 crucial clinical features, such as age, sex, anemia, creatinine phosphokinase, diabetes, ejection fraction, high blood pressure, platelets, serum-creatinine, serum sodium, smoking, time, and Death Event. Secondly, we develop Machine Learning algorithms to classify heart failure based on these clinical features. The prediction accuracy is approximately 80%. Lastly, we create a user friendly heart failure prediction system (HFPS) that comprises various features, including an input clinical data section, ROC curve display section, and prediction performance display section (execution time, accuracy, sensitivity, specificity, and prediction result). Our approaches have proven effective in predicting heart failure in patients. The HFPS system developed in this study presents a novel approach that can be utilized for heart failure classification. |
| 30 |
Sleepy driver detection system |
The Sleepy Driver Detection System (SDDS) stands as a crucial advancement in automotive safety, specifically designed to mitigate accidents stemming from drowsy driving. Employing a sophisticated blend of sensors and advanced algorithms, it continuously monitors the driver’s state, identifying signs of drowsiness in real-time. Key elements encompass infrared cameras, facial recognition technology, steering angle sensors, and biometric sensors such as heart rate monitors, collectively offering a comprehensive insight into the driver’s behavior and physiological condition. The SDDS software meticulously analyzes data from these sensors, evaluating parameters like eye movement, blink frequency, facial expressions, and steering patterns. Upon checking and analyzing signs of drowsiness or tiredness, this promptly issues audible and visual alerts, thereby averting of these accidents that are caused by driver tiredness. Its adaptability to diverse vehicles and driving conditions establishes it as an indispensable tool for bolstering road safety, effectively curbing drowsy driving-related incidents. |
| 31 |
EMOGENIUS – AN EMOTIONAL MUSIC RECOMMENDATION SYSTEM |
The emotional music recommendation system is proposed in this paper, which recommends music to users based on their emotional states. It uses a combination of audio features and lyrics analysis to determine the emotional content of music and provides personalized recommendations by matching the opened up new possibilities for music emotional states of users with the emotional recommendation systems to provide personalized recommendations based on the emotional states of users. content of music. It has the potential to revolutionize the way music recommendation systems operate by providing a more personalized and emotionally satisfying music listening experience for users. In our paper, we propose an emotional music recommendation system that recommends music to users based on their emotional states. |
| 32 |
Enhanced type performance classification for advanced vehicle safety monitoring using deep learning technique |
The emotional music recommendation system is proposed in this paper, which recommends music to users based on their emotional states. It uses a combination of audio features and lyrics analysis to determine the emotional content of music and provides personalized recommendations by matching the emotional states of users with the emotional content of music. It has the potential to revolutionize the way music recommendation systems operate by providing a more personalized and emotionally satisfying music listening experience for users. |
| 33 |
MEDICAL INSPECTION USING FIGMA AND DATA SCRIPTING |
The Med Inspect project represents a ground breaking initiative at the intersection of technology and healthcare, aimed at redefining healthcare accessibility for travellers worldwide. Rooted in a meticulously designed modular framework, Med Inspect introduces three core modules that empower travellers to proactively manage their health and well-being while abroad. The Travel Insurance Integration module forms the cornerstone of the system, ensuring travellers have immediate access to comprehensive travel insurance with medical coverage. This module not only offers peace of mind but also connects travellers to a global network of healthcare providers and insurance emergency assistance hotlines, streamlining medical support during emergencies. Complementing this, the Telemedicine Solutions module leverages cutting-edge telemedicine technology to provide travellers with convenient remote consultations for non-emergent medical concerns. It enhances healthcare access, allowing travellers to receive medical advice, prescriptions, and recommendations from the comfort of their location. Moreover, the Diplomatic Missions Assistance module serves as a critical lifeline during crises, enabling travellers to seek support and guidance from their country’s embassy or consulate. This module bridges language barriers and ensures travellers receive essential support in unfamiliar territories. Beyond the current framework, Med Inspect envisions future enhancements, including Emergency SOS integration, Mobile Clinics, and AI. |
| 34 |
NUTRITION RECOMENDAATION SYSTEM |
As of late, food recommender systems have gotten expanding consideration due to their pertinence for solid living. Most existing ponders on the food space center on suggestions that propose appropriate food things for personal clients on the premise of considering their inclinations or wellbeing issues. These systems moreover give functionalities to keep track of wholesome utilization as well as to convince clients to alter their eating behavior in positive ways. In this paper, we show a diagram of proposal methods for people within the sound food space. In expansion, we analyze the existing state-of-the-art in food recommender systems and talk about challenges related to the improvement of future nourishment proposal innovations. |
| 35 |
EARLY STAGE DITECTION AUTISM SPECTRUM DISORDER BY USING MACHINE LEARNING |
ASD is a neurological disorder that affects over 1 in 44 children and this rate has increased. The diagnosis process can be timely and costly. This will make it difficult for the patients to adhere to the prescribed treatments and will hinder the progress of the patient. This project is focused on increasing the efficiency of this diagnosis process through machine learning techniques. The proposed datasets are; ASD Screening Data for Adult, ASD screening Data for Children and ASD Screening Data for Adolescent. These datasets are a categorical, continuous, and binary data type. They have 21 common attributes and a total of 1,100 instances. |
| 36 |
AI-DRIVEN PRECISION HEALTHCARE WITH EFFICIENTNETB3 ALGORITHM: TRANSFORMING DERMATOLOGY FOR ACCURATE INTELLIGENT SKIN DISEASE CLASSIFICATION |
The rise of online shopping has made it a ubiquitous part of modern life. However, individuals with visual challenged often face barriers when trying to navigate e commerce platforms, which are typically not designed with their needs in mind. In this study, we propose a solution aimed at enhancing the accessibility of e-commerce websites for the visually challenged, utilizing Alan AI technology. Our approach involves integrating Face Recognition technology, powered by OpenCV, for tasks such as login, registration, and order placement on these platforms. The face recognition algorithm employed in our solution is based on support vector machine (SVM) technology. Alan AI offers a comprehensive Actionable AI Platform for the rapid development, deployment, and management of AI assistants. These assistants, when built with Alan AI, can engage with users naturally through voice or text, and execute tasks within various applications to boost productivity and enhance user experience. Specifically, our system employs voice AI assistants, which leverage technologies like Natural Language Processing (NLP), Spoken Language Understanding (SLU), Automatic Speech Recognition (ASR), Machine Learning (ML), Speech-to-Text (STT), and Text-to Speech (TTS) to facilitate seamless interaction between users and e-commerce platforms. |
| 37 |
Neural Tapestry: Illuminating Sleep Stages Through EEG Patterns |
This project uses a CNN to accurately classify sleep stages using EEG signals from the physionet dataset. Preprocessing techniques filter and segment EEG recordings, while the CNN extracts features.Training fine-tunes the model’s parameters for classification. Evaluation metrics confirm its effectiveness, outperforming traditional methods.
The CNN is robust to signal variations and computationally efficient for real-time applications, offering promising advancements in sleep science and neurological research. |
| 38 |
Comprehensive Digital Support System for Epilepsy Identification and Management |
Epilepsy patients confront a multitude of challenges that significantly impact their daily lives. The unpredictable nature of seizures disrupts routines and activities, while societal stigma and discrimination exacerbate feelings of isolation. Additionally, the side effects of medications can be debilitating, affecting cognitive function and overall well-being. Safety concerns loom large, with seizures increasing the risk of accidents and injuries. Psychological challenges such as anxiety and depression further complicate management, often compounded by limited access to specialized care and the financial strain of medical expenses . Educational and employment opportunities may be hindered, perpetuating a cycle of disadvantage . Addressing these difficulties requires greater awareness, advocacy, and support to enhance the quality of life for individuals living with epilepsy. |
| 39 |
Experiemental Investigation on Fusion Filament Fabrication of PETG/PLA for Biomedical Application |
This project highlights the comprehensive biochemical testing conducted on PETG/PLA blend, a promising biocompatible
material. Through rigorous experimentation, both PETG and PLA exhibited favorable outcomes in various biochemical assays. Looking ahead, this research aims to delve deeper into the properties of this novel PETG/PLA composite through extensive in vitro and in vivo testing. Such thorough
investigation will pave the way for the potential utilization of PETG/PLA in diverse biomedical applications, including but not limited to body implants and tissue engineering |
| 40 |
Advanced Deep Learning Architectures for Early Detection of Alzheimer’s Disease from MRI Scans with Jetson Nano |
Developing a deep learning-based system to accurately classify different stages of Alzheimer’s disease using MRI images.
Prioritizing early detection of Alzheimer’s stages, including mild, moderate, and very mild dementia, for timely intervention and treatment planning. Utilizing DenseNet121 architecture to extract meaningful features from MRI scans indicative of different disease stages.Training and fine -tuning the model on a curated dataset of MRI images representing various dementia stages.Assess the model’s performance using metrics such as accuracy, loss, precision, recall, F1 -score, and AUC-ROC analysis to ensure robustness and reliability in classification and incorporating with jetson nano. |
| 41 |
3D Reconstruction and Prediction of Diabetic Retinopathy (DR) in Retinal Vessels Using OCTA Images |
Optical Coherence Tomography Angiography (OCT-A) is a imaging modality for visualizing retinal vasculature, offering high- resolution depth-resolved images. However, It primarily provide 2D representations of retinal vessels, limiting our understandingof their intricate 3D architecture. The lack of robust 3D reconstruction methods hinders comprehensive analysis and diagnosis of retinal vascular diseases, such as diabetic retinopathy and age-related macular degeneration.
Therefore, there is an urgent need to develop an advanced 3D reconstruction technique that accurately captures the complex morphology of retinal vessels from OCT-A data. This project aims to address this challenge by leveraging advanced image processing, computer vision, and machine learning algorithms to reconstruct 3D models of retinal vasculature from OCT-A scans. The developed technique will provide clinicians and researchers with invaluable insights into the spatial distribution, connectivity, and abnormalities of retinal vessels, leading to improved diagnosis, treatment planning, and monitoring of retinal vascular diseases. |
| 42 |
Analysis of Skin Cancer Detection Using Dermatology Images |
Revolutionizing Diagnosis : Leveraging CNN to overcome biopsy limitations, expediting diagnosis for improved outcomes.Efficient Machine Learning: Focusing on CNN pattern discernment for precise skin cancer classification . Ethical Implementation : Addressing data anomalies and ethical concerns for responsible ML integration .Expedited Process : Streamlining diagnosis, reducing invasiveness, and enabling early interventions .Transformative Impact: Redefining skin cancer classification, saving lives through efficient, patient-friendly diagnostics .Skin cancer detection from dermatology images to improve diagnosis across different skin types and lesions and integrate seamlessly into clinical practice for early intervention . |
| 43 |
A Wearable EMG Measurement System for Sports Persons Using FPGA |
The project aims to develop an FPGA-based EMG wireless measurement system for athletes, enhancing performance, preventing injuries, and optimizing training strategies. The system includes real-time signal processing, wireless data transmission, adaptive thresholding, portability, a user-friendly interface, real-time feedback, data logging, and integration capabilities. The system enables athletes to monitor EMG signals during training sessions, provide real-time alerts, and evaluate muscle activity patterns and training load. The system also offers comprehensive data logging and analysis tools, enabling post-session evaluation of performance metrics . |
| 44 |
Predictive Precision: Machine Learning for Diabetics Foot Analysis |
Diabetes mellitus is a prevalent chronic disease that can lead to severe health complications if not diagnosed and managed in a timely manner. Foot thermography has shown potential as a non-invasive and cost-effective method for diabetes detection. However, accurate interpretation of foot thermographic images can be challenging due to variations in image quality, anatomical features, and the presence of other foot pathologies. Existing approaches lack robustness and may yield unreliable results. Therefore, the problem is to develop and evaluate a deep learning model, augmented with data augmentation techniques, that can effectively analyze foot thermographic images to accurately detect diabetes mellitus.
This model will aid in early diagnosis and improve the overall management of the disease, ultimately reducing the risk of complications for patients. |
| 45 |
Heart Attack Detection Using Heart Rate and ECG Sensor |
To develop an heart attack detector using Arduino Uno .The heart attack detection is done with the help of heart rate sensor and ECG sensor . The heart rate and ECG rate is monitored continuously and also intermittently to check the health of the person for any abnormality . The whole data monitored and collected by the LCD Display where the heart attack will be detected The abnormal conditions are detected by checking the heart activity levels . |
| 46 |
Epileptic Zonal Detection Using MRI and EEG Lab |
Develop methodologies to improve the precision and accuracy of identifying epileptic zones within the brain through the integration of MRI and EEG data. Diagnostic process by creating efficient algorithms and workflows that facilitate integration and analysis of MRI and EEG data for epileptic zonal detection. Investigate the underlying neurobiological mechanisms of epileptic seizures by correlating MRI-derived structural abnormalities with EEG-derived functional abnormalities, aiming to enhance the understanding of epilepsy pathophysiology. Investigate data fusion techniques that combine MRI and EEG information, providing the complementary strengths of each modality to improve the accuracy and reliability of epileptic zone detection. |
| 47 |
Non-invasive Hemoglobin Estimation and Analysis Using Classification and Regression Tree |
The project aims to develop a non-invasive method for measuring hemoglobin levels and analyzing its efficiency and accuracy using machine learning models. The goal is to analyze the reliable system that can assist healthcare professionals in quickly and precisely predicting hemoglobin levels, aiding in the early detection and monitoring of anemia and other hematological disorders. In this project, we propose a method for noninvasive measurements of hemoglobin concentration using sensors with IR LED and Photodetector and analysis of the data. |
| 48 |
Wireless Patient Monitoring System for Step-Down ICU Wards |
The objective of this project is to develop a comprehensive wireless patient monitoring system capable of continuously and accurately monitoring vital signs such as heart rate, blood oxygen saturation, respiratory rate, skin temperature, and cuffless blood pressure. The system aims to minimize code blue cases in hospital step-down ICU wards by providing real-time monitoring, early detection of abnormalities, and seamless data transmission to a dedicated platform for trend analysis and clinical decision support |
| 49 |
Disease Prediction and Analysis Using Cell-Level Count Matrix RNA Sequencing of Probe Annotations of Kidney |
The objective of this study is to develop a novel approach for kidney disease prediction and analysis using cell level count Matrix RNA sequencing of probe annotations. We aim to identify hub genes involved in kidney diseases and evaluate their potential as biomarkers for disease diagnosis and prognosis. |
| 50 |
Automated Skin Cancer Detection Using Deep Learning |
Skin cancer is a widespread and potentially life-threatening disease that demands early detection for effective treatment. Our project focuses on leveraging the power of deep learning, specifically the Efficient Net algorithm, to automate the detection of skin cancer from dermatoscopic images. By applying Efficient Net to dermatoscopic images, this project aims to explore and enhance the detection capabilities of skin cancer lesions. The significance of early detection cannot be overstated, as it directly influences the efficacy of subsequent treatments. Dermatoscopic images offer a valuable resource for diagnosis, providing detailed insights into the characteristics of skin lesions. |
| 51 |
Anaesthesia Prediction Using Lean Body Weight |
The objective of anaesthesia prediction is to develop predictive models or systems that accurately anticipate the anaesthetic needs of patients during medical procedures, aiming to enhance patient safety, optimize anaesthesia administration, and improve healthcare resource utilization. The primary objective of anaesthesia prediction is to develop a model or system that accurately forecasts the anaesthetic requirements for patients undergoing medical procedures. The ultimate goal is to enhance patient safety, improve anaesthesia efficacy, and optimize healthcare resource utilization by minimizing the risk of complications, adverse reactions, and unnecessary anaesthesia usage. |
| 52 |
Analyzing the Stages of Diabetic Foot Ulcer |
Create a robust system utilizing Arduino and pressure sensors to continuously monitor foot pressure. Utilize Support Vector Machine (SVM) algorithms to analyze pressure data and identify ulceration risk factors. Train the algorithms to recognize patterns associated with ulcer development, location, and severity. Enable early detection of foot ulcers by promptly alerting users to potential risk factors. |
| 53 |
Cough Monitoring for Respiratory Health |
Identify and extract relevant features from cough sound recordings to develop a comprehensive dataset for machine learning analysis. Train and optimize machine learning algorithms to accurately classify cough sounds associated with different diseases, prioritizing sensitivity and specificity. Evaluate the performance of the predictive model using diverse datasets to ensure generalizability and reliability across various populations and conditions. |
| 54 |
Voice-Based Object Detection for Visually Impaired People |
The objective of your project can be to detect the object with the goal of achieving high accuracy with a real -time performance. Furthermore, the project aims to enhance the quality of life for visually impaired individuals by empowering them with reliable object recognition capabilities, ultimately fostering greater autonomy and confidence in their daily activities . |
| 55 |
Infantile Spasm Monitor Using EMG |
Infantile spasms, prevalent among children under one year, present challenges for timely detection and intervention, typically monitored by Electroencephalography (EEG). Our project aims to improve the detection and management of infantile spasms, a condition common in children under one year old. We developed a wearable device that replaces traditional EEG monitoring for detecting infantile spasms. This device integrates EMG technology and works with a mobile app to alert caregivers instantly during spasms. By tracking muscle activity, heart rate, and temperature in real-time, it detects spasms early, addressing developmental concerns . Our solution overcomes EEG limitations, offering caregivers actionable insights and reducing stress. Plus, machine learning enhances accuracy over time, improving spasms detection . |
| 56 |
Liver Disease Prediction Using Machine Learning |
Gather a diverse dataset of CT scans containing liver tumor annotations. Preprocess the CT images to enhance image quality, reduce noise, and normalize intensities .
Design and implement the U-Net architecture tailored for liver tumor segmentation . Configure the U-Net model with appropriate encoder and decoder layers for feature extraction and segmentation . Extract segmented liver regions from CT scans using the trained U-Net segmentation model. Divide the dataset into training, validation, and testing sets. Train the CNN model using the segmented liver regions for liver disease classification |
| 57 |
Child Activity Recognition Based on Fusion Model |
This paper presents a child activity recognition approach using a single 3-axis accelerometer and a barometric pressure sensor worn on a waist of the body to prevent child accidents such as unintentional injuries at home. Labelled accelerometer data are collected from children of both sexes up to the age of 16 to 29 months. To recognize daily activities, mean, standard deviation, and slope of time-domain features are calculated over sliding windows. Child activities are classified into 11 daily activities which are wiggling, rolling, standing still, standing up, sitting down, walking, toddling, crawling, climbing up, climbing down, and stopping. The overall accuracy of activity recognition was 98.43% using only a single wearable triaxial accelerometer sensor and a barometric pressure sensor with a support vector machine. |
| 58 |
Optimizing Foot Pressure Measurement and Leveraging Precise Adjustment |
Design and construct a wearable device capable of optimizing foot pressure distribution in real-time. Implement Force-Sensitive Resistive (FSR) sensors to accurately measure foot pressure across various regions of the foot sole. Integrate FSR sensors with an Arduino microcontroller to process sensor data and control air pressure adjustments within the footwear. Enable the system to dynamically adjust air pressure in designated areas of the footwear based on real-time sensor readings, ensuring optimal comfort and support.Aim to improve foot health and comfort by minimizing pressure points, reducing the risk of foot- related ailments, and enhancing overall patient well-being. |
| 59 |
Integrating Assistive Technology with Monitoring System for Muscular Dystrophy Patients |
Develop an advanced system integrating GSM technology, EMG monitoring, and muscle sensors for real -time monitoring of muscle activity in muscular dystrophy patients.
Implement GSM-enabled SMS alerts for immediate insights into muscle health status. Empower patients with instant updates via SMS alerts, facilitating prompt medical assistance. Support remote monitoring by healthcare providers, fostering collaborative patient care. Enhance overall quality of life for muscular dystrophy patients with personalized, efficient, and accessible monitoring, utilizing GSM technology alongside other assistive technologies |
| 60 |
Integrating Force Detection and Sound Analysis for Crepitus Evaluation |
This research presents a pioneering approach in the realm of medical diagnostics: the development of a non-invasive bio-acoustics measurement system tailored for assessing the physiological conditions of knee joint articular cartilage. In response, this study proposes a novel knee acoustical method aimed at overcoming these limitations and facilitating early -stage diagnosis by physicians. By harnessing the power of frequency spectral analysis, this innovative approach holds significant potential in revolutionizing early -stage diagnosis and management of knee crepitus. |
| 61 |
Detection of Diabetic Retinopathy Using Vessel Density in OCTA Images |
To develop a software to get OCTA images as input and pre-process the OCTA data like enhancement, then smoothen the image, segmentation of image, calculate the FAZ parameters (circularity, mean area and vessel density). Based on cross validation, we have to provide the results based on Accuracy, Specificity, Sensitivity and AUC to detect DR in diabetic patients. The pre-processing should be done in the OCTA images of the retina. By calculating the FAZ circularity symmetry and the density of the blood vessel in the retina are used to cross validate and to detect the DR in the patients. |
| 62 |
Detection of Dental Caries Using Deep Neural Network |
Develop a robust deep neural network model for dental caries detection with high accuracy and efficiency, leveraging advanced machine learning techniques to analyze and interpret dental images.Enhance the objectivity and consistency of dental caries diagnosis by implementing a deep learning algorithm that can autonomously identify and classify carious lesions, reducing reliance on subjective human interpretation. Improve early detection and intervention of dental caries through the integration of deep neural networks, aiming to revolutionize the efficiency of dental diagnostics and contribute to proactive oral health management. |
| 63 |
Appliance for Enhancing Facial Development Through Correction of Tongue Posture |
Develop a non-invasive orthodontic appliance.Optimize facial harmony through the appliance.Correct and improve tongue posture effectively. Design a convenient and removable orthodontic solution. Address the gap in existing orthodontic care for facial aesthetics and tongue posture.Promote awareness about the importance of proper tongue posture in modern lifestyles.Mitigate adverse effects such as compromised aesthetics and orthodontic issues.Enhance overall oral and craniofacial health through comprehensive orthodontic solutions. |
| 64 |
HealthHub: Food Item Recognition with Calorie Estimation and HealthConscious Product Suggestions |
Accurately measuring the calorie content of food is essential for promoting healthy eating habits and managing dietary intake. However, calorie estimation poses challenges due to the diverse composition of ingredients and variations in cooking methods. This paper presents a novel approach for estimating food calorie content based on ingredient recognition and thermal information. The proposed method utilizes convolutional neural networks (CNN) for image classification to identify food items and extract their corresponding ingredients from a comprehensive database enriched with nutritional knowledge. Additionally, thermal imaging is employed to analyze the heat patterns of food ingredients, aiding in the segmentation and classification process. Fuzzy logic techniques are applied to classify ingredient boundaries based on their thermal signatures and intensity levels. The classified ingredients are then aggregated, and their calorie content is calculated using established nutrition knowledge and area ratios. Comparative analysis against conventional methods demonstrates the efficacy of the proposed approach in accurately estimating food calories. Furthermore, the HealthHub Food Item Recognition system integrates this calorie estimation functionality with health-conscious product suggestions, enhancing its utility for promoting balanced nutrition and facilitating informed dietary choices. |
| 65 |
Prostate Cancer Detection and Prediction System Using Power BI |
Develop an automated system using convolutional neural network (CNN) algorithms for detecting lung diseases from chest X-rays. Achieve accurate identification of various lung conditions such as pneumonia, bacterial pneumonia, and tuberculosis.Assist healthcare professionals in timely diagnosis and treatment planning, enhancing clinical decision-making. Utilize deep learning techniques and transfer learning for efficient feature extraction and classification. Optimize model performance through rigorous training, validation, and performance evaluation. |
| 66 |
WEARABLE FRACTAL ANTENNA FOR FIRE FIGHTERS USING BODY AREA NETWORKS |
Body Area Network technology is rapidly evolving presenting a future where wearable devices seamlessly integrate with our lives, fostering a new era of personalized health care through wearable devices. These devices requires antennas which can withstand the human motion and the immune to noise and produce precise results. The antennas should be flexible and compact in size. The substrate play a deciding role in flexibility of antenna. This paper presents a miniaturized wearable patch antenna for body area network (BAN) applications. The antenna utilizes a crown fractal design technique to achieve a size reduction of 31% compared toconventional designs. Additionally, a flexible Rogers RT Duroid 5880 substrate is employed, making the antenna suitable for wearable biomedical devices. The designed antenna operates in the 2.45 GHz ISM band and exhibits a gain of 4.54 dB and a bandwidth of 145 MHz, covering the entire band. imulations analyze the antenna’s performance through return loss (S11), directivity, radiation pattern, The results making it a strong candidate for wearable BAN applications. |
| 67 |
Driver drowsiness detection system |
In today’s tech world, our project serves as a versatile assistant, integrated with smart devices like Google and Siri. It handles voice input and output for tasks such as medical advice, organization, notes, calculations, and searches. Using microphones, it accesses the web for information, employing Natural Language Processing for communication. |
| 68 |
Enhanced cloud based infrastructure for secure and efficient medical services with ECC |
Smart architecture is the concept to manage the facilities via internet utilization in a proper manner. There are various technologies used in smart architecture such as cloud computing, internet of things, green computing, automation and fog computing. Smart medical system (SMS) is one of the application used in architecture, which is based on communication networking along with sensor devices. In SMS, a doctor provides online treatment to patients with the help of cloud-based applications such as mobile device, wireless body area network, etc. Security and privacy are the major concern of cloud-basedapplications in SMS. To maintain, security and privacy, we aim to design an elliptic curve cryptography (ECC) based secure and efficient authentication framework for cloud-assisted SMS. |
| 69 |
NutriDetect: ML-Powered Analyzer for freshness and nutrition in Fruits and vegetables |
This research presents an innovative approach to classify fruits and vegetables and provide detailed nutritional analysis and freshness assessment. Leveraging OpenCV for image processing and Convolutional Neural Networks (CNN) for machine learning, our system accurately identifies and categorizes produce from images. It also extracts essential nutrient information from a CSV dataset. Integrated with a freshness detection model, it empowers consumers to make informed decisions when selecting fresh and nutritious produce. |
Quick Links
