Automatic Segmentation and classification of Skin Lesion Images

Automatic segmentation and classification of seven-segment display digits on auroral images

Geoscientific Instrumentation Methods and Data Systems ◽

10.5194/gi-5-305-2016 ◽

2016 ◽

Vol 5 (2) ◽

pp. 305-314 ◽

Cited By ~ 2

Author(s):

Tuomas Savolainen ◽

Daniel Keith Whiter ◽

Noora Partamies

Keyword(s):

Automatic Segmentation ◽

Scientific Information ◽

Spherical Mirror ◽

Automatic Method ◽

Time Period ◽

Film Images ◽

Fully Automatic ◽

Night Sky

Abstract. In this paper we describe a new and fully automatic method for segmenting and classifying digits in seven-segment displays. The method is applied to a dataset consisting of about 7 million auroral all-sky images taken during the time period of 1973–1997 at camera stations centred around Sodankylä observatory in northern Finland. In each image there is a clock display for the date and time together with the reflection of the whole night sky through a spherical mirror. The digitised film images of the night sky contain valuable scientific information but are impractical to use without an automatic method for extracting the date–time from the display. We describe the implementation and the results of such a method in detail in this paper.

Download Full-text

Melanoma Detection by Meta-Heuristically-Optimized MLP Parameters Using Non-Dermatoscopy Images

International Journal of Applied Metaheuristic Computing ◽

10.4018/ijamc.2021100110 ◽

2021 ◽

Vol 12 (4) ◽

pp. 177-200

Author(s):

Soumen Mukherjee ◽

Arunabha Adhikari ◽

Madhusudan Roy

Keyword(s):

Particle Swarm Optimization ◽

Simulated Annealing ◽

Skin Lesion ◽

Multilayer Perceptron ◽

Binary Classification ◽

Ranking Algorithm ◽

Swarm Optimization ◽

Melanoma Detection ◽

Class 1

This paper represents a scheme of melanoma detection using handcrafted feature set with meta-heuristically optimized multilayer perceptron (MLP) parameters. Features including shape, color, and texture are extracted from camera images of skin lesion collected from University of Waterloo database. The features are used in two different ways for binary classification of the data into benign and malignant class. 1) The extracted features are ranked on their relevance using ReleifF ranking algorithm and also converted into PCA components and ranked according to their variance. Best result is obtained with 50 best ranked raw features with accuracy of 87.1%. 2) All 1,888 features are fed to an MLP with two hidden layers, with number of neurons optimized by two different metaheuristic algorithms, namely particle swarm optimization (PSO) and simulated annealing (SA) separately. The latter method is found to be more efficient, and an accuracy of 88.38%, sensitivity of 92.22%, and specificity of 83.07% are achieved by PSO, which is better in comparison with the latest research on this dataset.

Download Full-text

Automatic segmentation and classification of breast masses from mammography using machine learning

Medical Imaging 2021: Computer-Aided Diagnosis ◽

10.1117/12.2582157 ◽

2021 ◽

Author(s):

Prerna Chander ◽

Lina Eilouti ◽

Saubia Khan ◽

Salam Dhou ◽

Michel Pasquier ◽

...

Keyword(s):

Machine Learning ◽

Automatic Segmentation ◽

Breast Masses

Download Full-text

An Ensemble Method for Efficient Classification of Skin Lesion from Dermoscopy Image

Communications in Computer and Information Science - Computer Vision and Image Processing ◽

10.1007/978-981-16-1086-8_15 ◽

2021 ◽

pp. 161-178

Author(s):

B. H. Shekar ◽

Habtu Hailu

Keyword(s):

Skin Lesion ◽

Ensemble Method

Download Full-text

Comparing the capabilities of transfer learning models to detect skin lesion in humans

Proceedings of the Institution of Mechanical Engineers Part H Journal of Engineering in Medicine ◽

10.1177/0954411920939829 ◽

2020 ◽

Vol 234 (10) ◽

pp. 1083-1093 ◽

Cited By ~ 1

Author(s):

Aditi Singhal ◽

Ramesht Shukla ◽

Pavan Kumar Kankar ◽

Saurabh Dubey ◽

Sukhjeet Singh ◽

...

Keyword(s):

Skin Lesion ◽

Transfer Learning ◽

Wiener Filter ◽

Skin Lesions ◽

Automated Classification ◽

Convolutional Networks ◽

Image Quantization ◽

Malignant Skin ◽

Effective Diagnosis

Effective diagnosis of skin tumours mainly relies on the analysis of the characteristics of the lesion. Automatic detection of malignant skin lesion has become a mandatory task to reduce the risk of human deaths and increase their survival. This article proposes a study of skin lesion classification using transfer learning approach. The transfer learning model uses four different state-of-the-art architectures, namely Inception v3, Residual Networks (ResNet 50), Dense Convolutional Networks (DenseNet 201) and Inception Residual Networks (Inception ResNet v2). These models are trained under the dataset comprising seven different classes of skin lesions. The skin lesion images are pre-processed using image quantization, grayscaling and the Wiener filter before final training step. These models are compared for performance evaluation on different metrics. The present study shows the efficacy of the methodology for automated classification of lesion images.

Download Full-text

Automatic Segmentation and Classification of Heart Sounds Using Modified Empirical Wavelet Transform and Power Features

Applied Sciences ◽

10.3390/app10144791 ◽

2020 ◽

Vol 10 (14) ◽

pp. 4791 ◽

Cited By ~ 1

Author(s):

Pedro Narváez ◽

Steven Gutierrez ◽

Winston S. Percybrooks

Keyword(s):

Wavelet Transform ◽

Heart Sound ◽

State Of The Art ◽

Automatic Segmentation ◽

Automatic Classification ◽

The State ◽

Heart Sounds ◽

Empirical Wavelet Transform ◽

Art Methods

A system for the automatic classification of cardiac sounds can be of great help for doctors in the diagnosis of cardiac diseases. Generally speaking, the main stages of such systems are (i) the pre-processing of the heart sound signal, (ii) the segmentation of the cardiac cycles, (iii) feature extraction and (iv) classification. In this paper, we propose methods for each of these stages. The modified empirical wavelet transform (EWT) and the normalized Shannon average energy are used in pre-processing and automatic segmentation to identify the systolic and diastolic intervals in a heart sound recording; then, six power characteristics are extracted (three for the systole and three for the diastole)—the motivation behind using power features is to achieve a low computational cost to facilitate eventual real-time implementations. Finally, different models of machine learning (support vector machine (SVM), k-nearest neighbor (KNN), random forest and multilayer perceptron) are used to determine the classifier with the best performance. The automatic segmentation method was tested with the heart sounds from the Pascal Challenge database. The results indicated an error (computed as the sum of the differences between manual segmentation labels from the database and the segmentation labels obtained by the proposed algorithm) of 843,440.8 for dataset A and 17,074.1 for dataset B, which are better values than those reported with the state-of-the-art methods. For automatic classification, 805 sample recordings from different databases were used. The best accuracy result was 99.26% using the KNN classifier, with a specificity of 100% and a sensitivity of 98.57%. These results compare favorably with similar works using the state-of-the-art methods.

Download Full-text