Unsupervised Machine Learning Based Group Head Selection and Data Collection Technique

Nirmala, G; Guruprakash, C D

doi:10.1109/iccmc53470.2022.9753995

Cited by 8 publications

(9 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Distinctive practical works done on ECG leads. Such as 79% of the accuracy is obtained "Lead V2", 85.5% of accuracy is obtained from the" Lead I" Moreover ECG fiducial characteristics are studies by the Venkatesh et al [18] to recognize the 15 subjects. Fiducial points were first located, and then a collection of fiducial features including the ST, QRS, and PR intervals were extracted.…”

Section: Literature Reviewmentioning

confidence: 99%

ECG Biometric for Human Authentication using Hybrid Method

T G,

M N

2023

IJRITCC

View full text Add to dashboard Cite

Recently there is more usage of deep learning in biometrics. Electrocardiogram (ECG) for person authentication is not the exception. However the performance of the deep learning networks purely relay on the datasets and trainings, In this work we propose a fusion of pretrained Convolutional Neural Networks (CNN) such as Googlenet with SVM for person authentication using there ECG as biometric. The one dimensional ECG signals are filtered and converted into a standard size with suitable format before it is used to train the networks. An evaluation of performances shows the good results with the pre-trained network that is Googlenet. The accuracy results reveal that the proposed fusion method outperforms with an average accuracy of 95.0%.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

ECG Biometric for Human Authentication using Hybrid Method

T G,

M N

2023

IJRITCC

View full text Add to dashboard Cite

show abstract

“…Additionally, previously trained deep neural networks are fine-tuned using CNNs showed increased tolerance to the larger size of the dataset. Deep tuning and shallow tuning are challenging to distinguish from one another, though, as there are no universal standards for doing so; instead, it depends on unique circumstances [6]. Here, Deep tuning involves fine-tuning every layer of the deep convolutional network, whereas shallow tuning just involves the final few levels of the deep network.…”

mentioning

confidence: 99%

“…They came out with successful and précised results when compared to classic clustering and thresholding techniques. In other works, like [6], authors have presented a diagnostic system based on ResNet-50 for classifying breast abnormalities. Scaling and Contrast-based Data Augmentation (SCDA) was developed as an extension of scaling and Contrast-based Adaptive Histogram Equalization, an improvement to Existing Data for enhancing the proposed model which when trained with augmented training set resulted in ResNet-SCDA-50.…”

mentioning

confidence: 99%

“…• In [6], as we know that functioning of diagnosing system depends upon on the detection system. Hence, they need to work on implementing automatic detection system like CAD which makes the existing process complete.…”

mentioning

confidence: 99%

“…The key to CNN-based segmentation networks is to learn reliable highresolution features. CNN, a feed-forward neural network, solves the limitations of fully connected neural networks by being simple to train and generalize as shown in figure 1[6].The main components of CNN are Input layer, Convolution Layer, Pooling Layer, Fully Connected Layer and Output Layer. According with the suggested model, the input layer contains image data.…”

mentioning

confidence: 99%

See 2 more Smart Citations

A Novel Transfer Learning Approach to Improve Breast Cancer Diagnosing on Screening Mammography

Murthy,

H V,

E V

et al. 2024

IJEER

View full text Add to dashboard Cite

Segmentation is a technique for separating an image into discrete areas in order to separate objects of interest from their surroundings. In image analysis, segmentation—which encompasses detection, feature extraction, classification, and treatment—is crucial. In order to plan treatments, segmentation aids doctors in measuring the amount of tissue in the breast. Categorizing the input data into two groups that are mutually exclusive is the aim of a binary classification problem. In this case, the training data is labeled in a binary format based on the problem being solved. Identifying breast lumps accurately in mammography pictures is essential for the purpose of prenatal testing for breast cancer. The proposed TLA (Transfer Learning Approach) based CNN (Convolution Neural Network) –TLA based CNN aims to offer binary classification for rapid and precise breast cancer diagnosis (benign and malignant). In order to predict the sub-type of cancer, this exploration as used Deep Learning techniques on the Histogram of Oriented Gradient (HOG) - Feature extraction technique that creates a local histogram of the image to extract features from each place in the image with CNN classifier. This research work employs two well-known pre-trained models, ResNet-50 and VGG16, to extract characteristics from mammography images. The high-level features from the Mammogram dataset are extracted using a transfer learning model based on Visual Geometry Group (VGG) with 16-layer and Residual Neural Network with 50-layers deep model architecture (ResNet-50). The proposed model TLA based CNN has achieved 96.49% and 95.48% accuracy as compared to ResNet50 and VGG16 in the breast cancer classification and segmentation.

show abstract