Application of Neural Network and Computer Vision Technologies for Image Analysis of Skin Lesion
Annotation
Opportunity research of using neural networks and computer vision to analyze images of skin lesion and identify features of various pathologies, including oncological neoplasms. A methodology has been developed that makes it possible to evaluate the significance of combinations of color components and spaces in feature extraction using local binary patterns (LBP) and histogram of oriented gradients (HOG) computer vision technologies to extract features of skin changes binary classification of human skin lesions. Optimization of extracted feature makes it possible to more effectively solve the problem of data separability in classification. Research reveals an accessible way to classify skin lesions on a small dataset (less than 1000 images). Research is supposed to be applied to data sequences obtained using a new unique method of multispectral processing of skin lesions. In the course of the work, data from the ISIC-19 and ISIC-20 datasets were used. Samples were formed with a limit of 1000 images for training and validating the models. Additionally, a test sample of 250 images was formed. All images were reduced to 128 × 128 pixels and converted to YCrCb, BGR, Grayscale, HSV color spaces. Features were extracted for each color channel using the HOG and LBP methods. Mathematical models, including neural networks have been used for data classification. The effectiveness of features combinations by color channels and feature extraction methods was evaluated. The preprocessed images were divided into training and validation subsets in a 70/30 ratio. The accuracy, recall, precision and f1-score metrics were used to evaluate the models. The models were evaluated using stratified cross-validation and a test dataset. Optimization of model parameters was carried out based on the loss function represented by the average of cross-validation and evaluation on the validation set. In the process of research, more than 15 000 different optimizations of model parameters were executed. The most stable results on the validation dataset were achieved using ensemble of models, which were trained on a combination of features using local binary patterns (LBP) and histogram of oriented gradients (HOG) technologies. Models which used only local binary patterns technology had the best metrics values, but these models are not recommended to be used in practice without ensemble with stronger models. The results gained can be applied for usage with an ensemble of state-of-the-art convolutional and recurrent neural networks. The proposed approach is universal and applicable both for the analysis of individual images of skin neoplasms and for the analysis of their sequences obtained by the method of multispectral image processing. The technique can be applied to datasets with a limited amount of data. The results obtained will be of interest to specialists in the fields of computer vision and medical images analysis.
Keywords
Постоянный URL
Articles in current issue
- Influence of the dimension, geometry, and orientation of nanostructures on the distribution of the electric field in matters of enhancing of Raman scattering
- Optical properties of planar plasmon active surfaces modified with gold nanostars
- Application of bioradiophotonics methods for the processing of bioelectric signals
- Automatic recognition of internal structures in translucent objects based on hologram-moire interferometry.
- Implementation of digital holographic interferometry for pulsed plasma studies
- Polychromic light source for the realization of multispectral processing method of skin malignant lesions images
- Application of additional high-frequency modulation to reduce influence of residual amplitude modulation LiNbO3 phase modulator on fiber optical gyroscope signal
- Optimization of the optical scheme of a photodetector module operating in the spectral range of 1.3–1.6 μm
- Residue feature analysis with empirical mode decomposition for mining spatial sequential patterns from serial remote sensing images.
- Adaptive nonlinear motion parameters estimation algorithm for digital twin of multi-link mechanism motion trajectory synthesis
- Investigation of spectral-luminescent properties of cesium CsPb(BrCl)3 quantum dots in fluorophosphate glasses
- Investigation of optical phenomena in multispectral matrix photodetector based on silicon
- The impact of yttrium aluminum garnet stoichiometry deviation on the conversion efficiency of tetravalent chromium ions
- Influence of low temperatures and thermal annealing on the optical properties of InGaPAs quantum dots
- Pressure control in material extrusion additive manufacturing
- An enforced non-negative matrix factorization based approach towards community detection in dynamic networks
- Visual display system of changes in physiological states for patients with chronic disorders and data transmission via optical wireless communication.
- Ice reconnaissance data processing under low quality source images
- Korsakov I.N., and othersPrediction of fatal outcome in patients with confirm COVID-19
- Generation of the weakest preconditions of programs with dynamic memory in symbolic execution
- On the possibility of expanding the studied dynamic ranges in thermal anemometry
- The beating effect in uniaxial oriented polymer materials
- Numerical method for calculating the nozzle thrust of a wide-range rocket engine
- Numerical simulation of propulsive aerodynamic profiles