Medical Image Classification Using Thermal Images and Diagnosis of Carcinogen Govinda Rajulu. G, L. Sharmila, D. Venkatesan, K Srinivas, Nallam Krishnaiah, et al. 2024 International Conference on Innovation and Novelty in Engineering and Technology Innova 2024 Proceedings, 2024 Lung and Colon (L&C) tumors are lethal sicknesses that can foster in a few organs all the while and, in specific circumstances, jeopardize human existence. Even though it is highly unlikely that these two types of cancer will develop simultaneously, delaying diagnosis significantly increases the likelihood of metastasis between the organs that are affected. To really treat specific sorts of malignant growth, histological determination is fundamental. In the past, doctors had to go through a long and difficult process to look at thermal images and figure out if a patient had cancer; however, this procedure may now be completed much more quickly thanks to the new technology options. Histological images of L&C tumors were classified using a hybrid Deep Learning (DL) model with an attention mechanism and a multipath network in this study. To zero in on the main attributes and negligence the less significant ones, a consideration component was utilized. In a multipath network, data travels over a number of channels before each channel is converted and the output from all of the branches is combined. The multipath network is similar to grouped convolution when simplified. The five thermal image categories of the LC25000 dataset were utilized. Among these classes were two for colon disease and three for cellular breakdown in the lungs. The proposed model was compared to a number of well-known DL models, such as ResNet-50, VGG-16, and AlexNet. The proposed approach showed the best exhibition concerning exactness (99.2%), particularity (99.12%), responsiveness (99.28%), accuracy (99.12%), and F1 score (99.2%), as indicated by the exploratory discoveries
Implementation of Block Chain, IoT and Role-basis Data Access Control (RBAC) for Intelligent Manufacturing Rajanish Kumar Kaushal, S M Ravi Kumar, Pandeeswaran Chellaiah, Sumit Pundir, Ramya Maranan, et al. 7th International Conference on Inventive Computation Technologies Icict 2024, 2024 Various recommendations for predicting the dependability and quality of equipment contribute to the success of intelligent manufacturing systems. Numerous Role basis access control (RBAC) techniques are being investigated for this purpose. Data security and administration is another industry concern that is regarded as crucial. To surmount the challenges, The proposed work integrates block chain and Role basis access control (RBAC) to secure system transactions and manage a dataset to combat the forgery dataset. Big data techniques were used to manage and examine the collected dataset. Based on the hybrid prediction technique, the aspect of defect diagnosis prediction was evaluated on-linear Role basis access control (RBAC) techniques which are mainly used for estimating the system's complex background and figuring out its true positive rate. This was done to evaluate the proposed system's quality control.
Leveraging Machine Learning to Identify Synergistic Drug Combinations for Effective Cancer Treatment P. Sujatha, K Saravanan, Mohammed Ali Sohail, A Basi Reddy, Rohit R Dixit, et al. Proceedings of the IEEE International Conference Image Information Processing, 2023 The potential of drug combinations to treat and overcome medication resistance complex genetic diseases is evident. Synergistic drug combinations offer a promising way to enhance drug therapy efficacy and reduce the required medication dosage. However, developing effective combination medication therapies with synergistic effects has been challenging, despite numerous ongoing clinical investigations. Current models and approaches to detect medication synergy outlined in the literature lack the expected consistency in outcomes. to better comprehend the impact of particular medication combinations, it is essential to be familiar with the vocabulary used to describe synergy. In this study, a combinational drug screen is utilized to identify useful features for locating synergistic or efficient drug combinations. The feature selection algorithm (Boruta) helps select the most relevant features, and machine learning models are then trained using the selected feature dataset. Performance assessment metrics like sensitivity, accuracy, and specificity are used to compare the trained models, and the Random Forest model stands out for its significantly better performance compared to other models.
Applications of Ensuring Security and Privacy Using Block Chain with IoT for Health Record Abdul Shareef Pallivalappil, Sayed Sayeed Ahmad, Yeligeti Raju, Ch. Kishore Kumar, Thamba Meshach W, et al. 2022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering Icacite 2022, 2022 The healthcare system has key security and privacy requirements when considered like an enterprise, such as safeguarding patients' medical records from unwanted access, protected drug tracking, secure connection with transportation such as ambulances, and secure and smart e-health surveillance. With suitable security measures, block chain has brought novel concepts in security and safety of medical data, and it may reconcile the discrepancy among sharing data and confidentiality. We combine the strengths of both block chain and cloud computing in this research to provide a confidentiality method for block chain and IoT. This strategy incorporates IoT and delivers IoT services to block chain nodes; in the meantime, it gathers, examines, operates, and preserves in the identity validation for health information. Interaction and addresses the inadequate computing capabilities of some block chain nodes in order to confirm data validity and feasibility. The proposed approach is efficient, as demonstrated by the simulation experiment. It can preserve and verify the integrity of medical data while also addressing issues such as high computer complexity, data exchange, and privacy protection.
Machine Learning Approach to Patient Health and Stress Monitoring System P. Madhuri, Nallam Krishnaiah, P. Anandan, U Nilabar Nisha, Ashish Kumar Tamrakar, et al. Mysurucon 2022 2022 IEEE 2nd Mysore Sub Section International Conference, 2022 Over 90% of people have too many feelings of stress, and there is no external remedy to relieve stress other than being diverted to other cases. Stress can be good in helping the less fortunate, or it could be injurious, but it continues to suffer. If the demanding model of a human keeps going, he or she should die. After conducting research, the scientists discovered that using Internet of Things (IoT) devices and sometimes a Neural Network (NN) model to eradicate a person's stressful behavior. The gadgets used to gather information from the body of the patient would have a connection to the internet and be programmed to analyze the data using a Machine Learning(ML) algorithm. Now the whole system has been tested in several scenarios. Different parts sent out with standards have been managed and monitored. The goal of this study is to eliminate stress with continuous statistics and go through the whole process.
Automatically prospecting feature for queries from their search impact Dr. Krishnaiah Nallam, B. Ganga Bhavani, B S N Murthy, G.L.N.V.S Kumar, and International Journal of Engineering and Advanced Technology, 2019 We recommend that you compile the duplicate lists in the top search engine results to track the aspects of the query and implement a method known as QDMiner. More specifically, QDMiner extracts free text lists, HTML tags and re-regions the top search engine results, combining them with groups according to the products they contain, then line up the blocks and products, depending on how the conversation and products are included in the best results. The recommended approach is generic and does not depend on understanding any area. The main purpose of the extraction side differs from the query recommendations. We recommend a structured solution, described as QDMiner, to trace query aspects immediately by removing and grouping repetitive lists in free text results and HTML tags and repeating search engines. We continue to evaluate the support of the list and discover better search queries by looking for exact similarities between menus and penalizing duplicate lists. Experimental results reveal that there are many listings available and QDMiner can find useful queries. The proposed approach is general and does not depend on understanding a particular area. As a result, it can handle open-domain queries. The query supports. Instead of a static system for your problems, we extract the sides of the uploaded document above each query.
A unique class prediction classifier for redundant multi-label values to support efficient clustering International Journal of Applied Engineering Research, 2017
