Shilpa Kodli

Work place: Department of CSE (MCA), Visvesvaraya Technological University, Centre for PG Studies, Kalaburagi, India

E-mail: shilpakodli@gmail.com

Website:

Research Interests: Data Structures and Algorithms, Computer Networks, Artificial Intelligence

Biography

Shilpa Kodli, (PhD), is working as Assistant Professor at Department of Computer Science and Engineering (MCA), Visvesvaraya Technological University, Center for PG Studies, Kalaburagi, India. Research areas are Cloud Computing, Networks, and Artificial Intelligence. She has published more than 19 peer reviewed research articles.

Author Articles
Diagnosis of Skin Cancer Using Machine Learning and Image Processing Techniques

By Prashant Kaler Shilpa Kodli Sudhir Anakal

DOI: https://doi.org/10.5815/ijeme.2022.05.05, Pub. Date: 8 Oct. 2022

Skin Lesion is a part of the skin that can be caused by abnormal growth in the epithelium layer on the skin. There are nine types of skin lesion like Actinic Keratoses (AK), Basal Cell Carcinoma (BCC), Dermatofibroma (DF), Melanoma (MEL), Melanocytic Nevi (MV), Benign Keratosis (BK), Vascular Lesions (VASC), Squamous Cell Carcinoma (SCC), and Pigmented Benign Keratosis (PBK). The aim of this study is to spotlight on the problem of skin lesion classification based on early detection of the disease using deep learning techniques. This approach is used to work out the problem of classifying a dermoscopic image. The dermoscopic is a digital device; in this case Smartphone is attached to a lens and collects the images through the device. The proposed spotlight is built in the region of using Convolutional neural network architecture and ResNet-50 module is used to predict Skin-Lesion classification. The dataset used in this research was taken from kaggle repository. The proposed work uses ResNet-50 CNN model which has yielded 93% of accuracy for detecting Skin Cancer, previous work was carried out using Visual Geometry Group model which yielded 73% accuracy. In the proposed work we have considered 25,000 images of skin lesion. Hence we are able to attain this accuracy with more reliable Machine Learning algorithms compared to the previous work.

[...] Read more.
Other Articles