Sri Shakthi SIET Journals
Not a member yet
211 research outputs found
Sort by
Hip Implant Design using Stainless Steel 316L for Enhanced Stability and Patient Comfort
Get PDFIn a hip replacement procedure, the damaged bone and cartilage are removed and replaced with an artificial component known as prosthesis. Despite advancements in implant sterilization, design, fixation techniques, and the introduction of robotic surgery, a persistent challenge is to identify an optimal, patient-specific hip implant that meets individual criteria. The primary objective of the proposed study was to create a highly accurate patient-specific hip implant by standardizing the existing design. The secondary objective aimed to demonstrate the superiority of a customized design over a conventional one. Geometric measurements of the hip were extracted from CT scans using MIMICS 20.0 software, and the implant design was developed using SolidWorks. Finite Element Analysis (FEA) was employed for meshing and analyzing the planned implant. Comparative research through FEA analysis indicated that a customized implant made from SSL 13 material outperformed the standard implant, showcasing its suitability for the patients studied
Time Series Analysis for Tractor Sales using SARIMAX and Deep Learning Models
Get PDFTime series forecasting is known for playing vital role in many industries to make important decisions and strategies. This study concentrates on providing accurate insights that can help manufactures and stakeholders of agriculture machinery industry on future sales of tractors by applying both traditional and deep learning models like SARIMAX which is extension of SARIMA and deep learning models. Research starts by observing history data which include years of tractor sales then preprocess the data to find its quality and stationarity further applying SARIMAX model to find trends and seasons and cycles in the data and this model is evaluated by famous metrics like Root Mean Squared Error (RMSE).Deep learning models like Gated Recurrent Unit (GRU), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Bidirectional LSTM, CNN LSTM Encoder Decoder, Convolutional Neural networks (CNN). They can help in enhancing the forecasting accuracy by handling all the non-linear relationships and their dependencies in the timeseries and this study will provide comparative analysis of deep learning models and SARIMAX model. Where SARIMAX outperformed the deep learning models with RMSE score 0.01 and provide forecast of next year’s tractor sales using SARIMAX model from the study and use q-q plot, residual plots and ACF and PACF graphs to make sure forecast was done accurately
Modeling of Ion Sensitive Field Effect Transistor for Sensing Application using TCAD
Get PDFHydrogen ion concentration (pH) of a solution can be measured using FET type sensor called Ion sensitive field effect transistor, ISFET. Chemical reactions occur at the electrolyte – insulator interface making the FET sensitive to pH. The objective of this work is to model the electrolyte-insulator structure of a transistor using Silvaco TCAD tool. Sensitivity is measured based on the shift in the threshold voltage which is caused by the effect of pH on the charge and the potential distributions in the gate insulator. Based on the analytical calculation of parameters of the electrolyte region, semiconductor materials are used to model the reference electrode and electrolyte. In this study, Silicon Nitride and Aluminum Oxide are used as gate insulators for ISFET and their performance comparison is made for sensing applications. The transfer and output characteristics of the transistor are obtained by simulation for both the films for various thicknesses. A comparison of the effect of thickness of films on device performance is analyzed since the dielectric constant of Aluminum oxide is higher than Silicon nitride
Dynamic Topic Modeling Techniques for Evolving Medical Texts with UMLS Concepts
Get PDFIn the realm of biomedical research, efficiently extracting and categorising subjects from enormous amounts of medical texts is crucial for knowledge discovery and information retrieval. Traditional topic modelling approaches are useful, but they usually fall short in capturing the intricate semantics of medical terminology. This study investigates the potential benefits of using Unified Medical Language System (UMLS) principles to topic modelling using the MedMentions dataset. We employ four techniques: BERTopic, Latent Dirichlet Allocation (LDA), Hybrid LDA and RNN, and a novel Hybrid BERTopic with Recurrent Neural Networks (RNN). By incorporating UMLS concepts into these models, we hope to improve subject coherence and relevance. According to our research, in terms of clinical relevance and topic coherence
Bitmap Fuzzing: A Comparative Analysis of Libfuzzer and AFL Tools
Get PDFThis research work compares the effectiveness of bitmap fuzzing between two prominent algorithms exploited in LibFuzzer and AFL, aimed at identifying vulnerabilities in software handling bitmap images. Bitmap fuzzing generates varied image-based test cases that rigorously test software, revealing potential security flaws. In this analysis, LibFuzzer, an in-process guided fuzzer, and AFL, an external fuzzer driven by coverage feedback, are both utilized to evaluate their accuracy in detecting errors within bitmap-processing applications. The performance is assessed based on the types and frequency of errors found, offering a layered perspective on error-handling strength in image processing contexts. By recording software crashes and categorizing the faults, this proposed research provides important insights into the comparative strengths and limitations of these fuzzing tools. The experimental results aid significant improvements in fuzzing practices, enhancing security frameworks by enabling early identification of vulnerabilities in multimedia-focused applications. The devised comparative research highlights the critical role of fuzzing tools in building robust, resilient software defenses
Possibility of formation of stationary structures in relativistically degenerate magnetized quantum plasma with exchange-correlation energy
Get PDFIn the present paper we have studied the possibility of stationary structure formation in ion acoustic wave in a relaivistically degenerate quantum plasma in presence of magnetic field quantum diffraction parameter and localized exchange correlation energy. Recent authors include exchange correlation term in many plasma configurations including quantum and relativistic regime. We have analyzed the applicability of certain mathematical tools like the Sagdeev pseudo-potential method in dealing with the analysis of the formation and properties of large amplitude solitary structures, double layers, shocks etc. The findings of this paper will help future researchers to select analytical methods while studying wave phenomena in plasma
Physicochemical and Biological Properties of Land and Water Bodies Surrounding Major Dumpsites in Kolkata
Get PDFIn this paper we have investigated the data acquired from the analysis of the soil and water samples from and around the dump sites in Kolkata and North 24 Paragana district of West Bengal where the population density is extremely high. The treatment of disposal of municipal solid wastes and waste water has been inadequate to negligible in these areas and as a result the quality of soil and water bodies is subject to deterioration. We have made use of GPS enabled Satellite acquired images and its associated softwares to identify and demarcate the areas that come under the direct impact of the dumping sites, and which ultimately are the areas prone to diseases and degradation in the coming years. The pH, salinity, Total Dissolved Solvents and oxidation reduction potential has been investigated for the basic characterization of the samples. An estimate of heavy metals has also been made. Estimation of salts and oxides from the various sediments and soil samples were acquired. Identification of bacteria, under the purview of biological studies and the dependence of their growth on physiochemical parameters of the surrounding has portrayed an alarming result. Adjacent to these areas there are agricultural fields where leached water from the dumping site directly drain into and cause biomagnifications. Contamination of the ground water is also sizable. This research will help to control pollution and biological outbreaks as well as suggest areas where immediate care should be taken to set up environmental restoration. The findings of the paper will further enlighten the planning and designing of waste disposal in urban areas and assist in its policy making in urban areas and thereby improve the quality of life of the scavengers who are left to equate their survival with the garbage mounds
Design & Prototype development of 2.2 kW Axial Flux PCB Stator Motor
Get PDFThis paper presents the design and prototype development of 2.2 kW Axial flux PCB stator motor, compared to the conventional permanent magnet type of electric motors. This construction eliminates the iron core, as a result there is a reduction of weight, cogging torque, noise and increase in efficiency. Analytical design was done, and prototype is developed with 8-layer PCB board with 6 Oz of cu is used to build the required power
Skin Cancer Classification using Deep Learning
Get PDFAccording to world health organization skin cancer is the one of the most common cancer types in the world. The abnormal growth of skin cells most often develops on the skin when exposed to the sun and occurs when there is a mutation in the DNA of skin cells, it begins at the top of the skin. More than five million people are affected by skin cancer each year. The proposed method aim at analyzing and detecting the significant class of skin cancer variant such as Melanoma, Basal cell Carcinoma, Nevus. Melanoma is the most dangerous form of skin cancer when compared to the other types. In this paper we have developed a webapp that could differentiate skin cancer. The data set has been taken from ISIC and the model is trained using Gcollab. The proposed work has used convolution neural network (CNN) as algorithm for deep learning as it has higher accuracy and flask is used to develop the web app and the class of cancer is classified based on historical data of dermoscopic images
Crop Prediction using Supervised Learning
Get PDFAgriculture is a key driver of a nation's economy, supplying raw materials, employment, and essential food. In India, the world's second most populated country, a significant portion of the population depends on agriculture for their livelihood. However, farmers face numerous challenges, including crop diseases, poor soil quality, unpredictable weather, and water scarcity. Additionally, repeated cultivation of the same crops and the indiscriminate use of fertilizers deplete soil nutrients, further reducing crop yields. Modern technology adoption may minimize these issues and improve the quality and production of agriculture. In agriculture, machine learning (ML), a branch of artificial intelligence (AI), makes automation, classification, and prediction easier. It supports well-informed decision-making for food security and effective crop management by assisting in the optimization of crop selection, fertilization, and irrigation. Using the Kaggle crop recommendation dataset, this paper presents a strong machine learning architecture for crop prediction. Important input characteristics including soil pH, temperature, humidity, and nutrient levels are included in the dataset. Using classification techniques like Decision Tree (DT) and Support Vector Machine (SVM), the system identifies the most appropriate crop for a specific type of soil based on its weather and soil data. It also offers information on the amount of fertilizer, seeds, and soil nutrients needed for production. By using this system, farmers can explore new crop varieties, increase agricultural productivity, enhance profit margins, and reduce soil pollution