Identification of Melanoma through Dermoscopy Image using Learning Vector Quantization

The face is one of the media to identify someone, a human face has a very high level of variability. Many methods have been introduced by researchers and scientists in recognizing one's face, one of the methods introduced is the Feature Extraction of Gray Level Co-Occurrence Matrix (GLCM) and Learning Vector Quantization (LVQ). GLCM feature extraction is used for data extraction/learning process whereas a data analysis process (face recognition, cropping and storing data) the LVQ method is used for the data training process where the data that has been processed in GLCM feature extraction which still has large dimensions are processed to be smaller dimensions. So this test uses data of 190 photos and gets a match of 90%, the authors conclude that the GLCM feature extraction and LVQ method can very well recognize faces contained in the database.

Thrombocyte disease is usually caused by abnormalities, such as abnormalities based on the number and morphological deformities of platelets. Examples of platelet abnormalities include small platelets in Wiskottldrich syndrome, giant platelets in some chronic myeloproliferative diseases, Benard Soulier syndrome and Macrothrombocytopenia in gray platelet syndrome. The usual problem of automatic FBC analysis is that undetectable morphological abnormalities of platelets so the microscopic examination is required using peripheral blood smear. But microscopic examination also has some weakness such as subjective depend on medical analyst/pathologist. We propose an accurate method to classify plateles from digitalization microscopic peripheral blood smear using combination of second order statistic feature extraction and comparing several methods. The comparing methods are K-Nearest Neighbor (KNN) and Learning Vector Quantization (LVQ). In this feature extraction, we use Gray Level Co-occurrence Matrix (GLCM) to get Angular Second Moment (ASM), Invers Different Moment (IDM) and entropi values. Those values will be inserted as input in KNN classifier method to classify blood cell in peripheral blood smear. Classify of cells based on feature extraction values is divided into three classes (leukocytes, normal platelets and giant platelets). Based on the result of experiments, both of methods can classify platelets on all color channels with average accuracy are 83.67% for KNN and 74.75% for LVQ. So, The KNN classification method is better able than LVQ to classify platelets in peripheral blood smear.

Diversity on durian varieties makes it difficult to distinguish between durian varieties. In addition, some durian varieties also have similarities which further complicate the classification process, and are known as typical tropical fruits, these fruits are native to Southeast Asia and have been introduced throughout the world. The ability to recognize objects is needed by humans. Recognition can facilitate human daily activities. Pattern recognition is often applied to a variety of objects, one of which is pattern recognition on fruit. There is some limitations of human memory in remembering object features. Based on this problem, research was conducted to identify fruit plants based on its features using digital image processing techniques, Learning Vector Quantization (LVQ) algorithm and Gray Level Co-occurrence Matrix (GLCM). Each type of plant has different shapes, colors, and textures; this is what makes the durian unique to other durians. The program was created using Microsoft Visual C # 2010 software. The test results achieved an 89% success rate in recognizing fruit plants based on the fruit.

Research on the Identification of Stroke with MRI Imagery Using the Learning Vector Quantization (LVQ) Method Based on Texture Features has been carried out. This study aims to determine the program's best parameters and the highest accuracy level of the stroke identification program. This research was conducted at Haji Sukolilo Hospital - Surabaya by obtaining 57 images of stroke patients and 15 images of regular patients. The study used the intelligence of stroke, tumor, and standard images to determine each category's image characteristics. After knowing the differences in each class, the next process is digital image processing, followed by feature extraction used is the Gray-Level Co-occurrence matrix (GLCM) with four parameters: contrast, correlation, diversity, and homogeneity. These four parameters are the best input parameters with an intelligence rate of 0.100 with a decrease in intelligence rate of 0.100, so the best accuracy value for training is 74.97%, and test data is 78.60%. Regarding the program's ability to correctly identify 11 data from 14 data tested, the program is feasible to be used as a second opinion.

Skin diseases are increasing and becoming a very serious problem. Skin cancer in general there are 2, namely melanoma and non-melanoma. Cases that are often encountered are in non-melanoma types. A critical factor in the treatment of skin cancer is early diagnosis. Doctors usually use the biopsy method to detect skin cancer. Computer-based technology provides convenient, cheaper, and faster diagnosis of skin cancer symptoms. This study aims to identify the type of skin cancer. The data used in the study were 6 types of skin cancer, namely Basal Cell Carcinoma, Dermatofibroma, Melanoma, Nevus image, Pigmented Benign Keratosis image, or Vascular Lesion, with a total of 60 dermoscopy images obtained from the Kaggle site. Dermoscopy image processing begins with a pre-processing process, which converts RGB images to LAB. After that, segmentation is carried out to separate objects from the background. The method of extracting shape and texture features is used to obtain the characteristics of dermoscopy images. As many as 2 types of shape features, namely eccentricity and metric, and 4 types of texture features, namely contrast, correlation, energy, and homogeneity. The result of this study is that it can identify the type of skin cancer based on image features that have been extracted using a program from the Matlab application. The technique of extracting shape and texture features is proven to work well in identifying the type of skin cancer. In the future it is expected to use more data, and add color features in identifying dermoscopy images.

Nowadays, humans tend to carry out activities during the day, both indoors and outdoors. Activities carried out outdoors cause human skin to often receive direct exposure to sunlight, which contains ultraviolet (UV) rays. Direct exposure to UV rays on the skin will harm the skin's health, which is the covering of the human body. Harmful effects on the skin usually include the skin becoming dark and dull, burns, and even causes cancer. One of the skin cancers that may appear on human skin is Actinic Keratosis (AK) cancer. AK cancer is a type of cancer that is classified as benign and can be cured with medical help. However, if this cancer is not caught early, it can become Squamous Cell Carcinoma (SCC), a type of malignant cancer. This research aims to design a system for identifying AK cancer types using color and texture feature extraction. RGB color feature extraction is obtained from image color segmentation and RGB values. The Gray Level Co-occurrence Matrix (GLCM) method is used to determine the texture of the skin cancer. Identification is carried out by a classification process using a Support Vector Machine (SVM), which can recognize the type of AK cancer. This research uses three classification methods: classification with color extraction, classification with texture extraction, and classification with color and texture extraction. Research shows that the highest level of accuracy in cancer recognition reaches 96% by combining color and texture extraction results as classification determinants. So, the system designed has succeeded in recognizing the type of AK cancer early on..

This study aims to develop an AI-based diagnostic tool for classifying dental conditions and tooth types to enhance the accuracy and efficiency of dental diagnostics. Manual documentation and diagnosis in dentistry are often prone to errors, inefficiencies, and delays, leading to adverse patient outcomes. Leveraging digital image processing and machine learning, this research addresses these challenges by automating the classification process. Dental imaging data were collected from the Dental and Mouth Hospital (RSGM) of Nala Husada Surabaya, Indonesia, comprising 3,910 images categorized into dental conditions (1,767 images) and tooth types (2,143 images). The dataset was preprocessed through resizing, grayscale conversion, histogram equalization, and median filtering. Texture features were extracted using the Gray Level Co-occurrence Matrix (GLCM), and classification was performed using Support Vector Machine (SVM), K-Nearest Neighbor, Naïve Bayes, Decision Tree, and Random Forest algorithms. The SVM algorithm achieved the highest accuracy of 54.24% for dental conditions and 41.49% for tooth types, outperforming other methods. However, the overall performance was suboptimal, primarily due to dataset limitations, reliance on GLCM for feature extraction, and insufficient preprocessing. The results highlight the potential of AI-based tools in dentistry but also underscore the need for improvements in dataset diversity, advanced feature extraction methods, and hyperparameter optimization. Future research should focus on expanding the dataset, exploring deep learning-based feature extraction, and employing robust evaluation strategies to enhance model performance. This study lays the groundwork for developing a more reliable and efficient AI-based diagnostic tool, ultimately improving patient outcomes and streamlining clinical workflows in dentistry.

Descending stairs and floors classification as control reference in autonomous smart wheelchair

Melanoma, a type of skin cancer, occurs when melanocytes become cancerous and is a common cause of death in adults. The presence of melanoma can be conclusively proved through biopsies, but these lap reports often take time. Early detection of melanoma could improve mortality rates and reduce costs. AI-based assistive tools can aid early detection. Most studies focus on detection either in dermoscopic images or in non-dermoscopic images, not both. In this paper, we propose a novel generalised framework which can detect melanoma in both dermoscopic and non-dermoscopic images. The framework includes a preprocessing pipeline, data augmentation and resolving class imbalances, followed by a VGG-16 model. The model gives a sensitivity (for melanoma cases) of 87% on non-dermoscopic images and 91 % on dermoscopic images.

One example of the implementations of digital image processing in biomedical field is to identify the gender of the fetus on the ultrasound image. To identify the gender of the fetus, a fetal must attain the age of at least 5 months of pregnancy. Before the process of identification, there are three steps that must be done, i.e. image preprocessing, image segmentation, and feature extraction (shape description). Having obtained the value of the feature extraction stage, the next step is the classification by utilizing one of the artificial neural network (ANN) methods, namely the learning vector quantization (LVQ). Prior to the LVQ process, the training datasets process is conducted beforehand with 3 iterations using the learning rate of 0.05 and the learning rate reduction of 0.02 per iteration. Then the training process is followed by a classification stage. The obtained test results show that the LVQ classification gives poor results. The less optimal results are generated due to the quality of the dataset used. The quality of this dataset is affected by the results of the digitization process, the stage of preprocessing, segmentation, and feature extraction.

Feature extraction is important to obtain information in digital images, where feature extraction results are used in the classification process. The success of a study to classify digital images is highly dependent on the selection of the feature extraction method used, from several studies providing a combination of feature extraction solutions to produce a more accurate classification. Classifying the types of marine fish is done by identifying fish based on special characteristics, and it can be through a description of the shape, fish body pattern, color, or other characteristics. This study aimed to classify coral reef fish species based on the characteristics contained in fish images using Backpropagation Neural Network (BPNN) method. Data used in this research was collected directly from Bunaken National Marine Park (BNMP) in Indonesia. The first stage was to extract shape features using the Geometric Invariant Moment (GIM) method, texture features using Gray Level Co-occurrence Matrix (GLCM) method, and color feature extraction using Hue Saturation Value (HSV) method. The third value of feature extraction was used as input for the next stage, namely the classification process using the BPNN method. The test results using 5-fold cross-validation found that the lowest test accuracy was 85%, the highest was 100%, and the average was 96%. This means that the intelligent model derived from the combination of the three feature extraction methods implemented in the BPNN training algorithm is very good for classifying coral reef fish.

In recent years, computerized biomedical imaging and analysis have become extremely promising, more interesting, and highly beneficial. They provide remarkable information in the diagnoses of skin lesions. There have been developments in modern diagnostic systems that can help detect melanoma in its early stages to save the lives of many people. There is also a significant growth in the design of computer-aided diagnosis (CAD) systems using advanced artificial intelligence. The purpose of the present research is to develop a system to diagnose skin cancer, one that will lead to a high level of detection of the skin cancer. The proposed system was developed using deep learning and traditional artificial intelligence machine learning algorithms. The dermoscopy images were collected from the PH2 and ISIC 2018 in order to examine the diagnose system. The developed system is divided into feature-based and deep leaning. The feature-based system was developed based on feature-extracting methods. In order to segment the lesion from dermoscopy images, the active contour method was proposed. These skin lesions were processed using hybrid feature extractions, namely, the Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) methods to extract the texture features. The obtained features were then processed using the artificial neural network (ANNs) algorithm. In the second system, the convolutional neural network (CNNs) algorithm was applied for the efficient classification of skin diseases; the CNNs were pretrained using large AlexNet and ResNet50 transfer learning models. The experimental results show that the proposed method outperformed the state-of-art methods for HP2 and ISIC 2018 datasets. Standard evaluation metrics like accuracy, specificity, sensitivity, precision, recall, and F-score were employed to evaluate the results of the two proposed systems. The ANN model achieved the highest accuracy for PH2 (97.50%) and ISIC 2018 (98.35%) compared with the CNN model. The evaluation and comparison, proposed systems for classification and detection of melanoma are presented.

Melanoma is one of the most common types of skin cancer. A melanoma can be classified into benign and malignant. Fortunately, melanoma is completely curable if caught early. Melanoma, both benign and malignant, can present in a manner that is identical. As a result, distinguishing the two is quite challenging. The early identification of skin cancer presents a significant challenge. To determine which one is benign and which one is cancerous, you need the expertise of a board-certified dermatologist. Automated early detection systems for skin cancer are absolutely necessary for saving lives and reducing time and effort. This paper proposes an automatic skin lesion categorization system that uses a pretrained deep learning network and transfer learning. A detection system based on digital image processing was developed in order to make an early diagnosis of melanoma. In order to make the images of skin more appealing, dermoscopy images were first acquired, then preprocessed and postprocessed to remove noise. A modified k-means clustering algorithm is used to segment the processed image after k-means clustering. The attributes are then extracted using first-order statistics, gray-level co-occurrence matrices, and feature extraction techniques. Comparing the proposed method to the existing approaches, it was discovered that the classification rate of the suggested method exceeded the output of the existing approaches.

Owing to shortage of awareness regarding signs and measures for prophylaxis, skin cancer (SC), among the most severe kinds of carcinoma (Ca), is actively destroying far more people than it used to. Consequently, in order to halt the progression of Ca, timely identification during an early phase is crucial. Melanoma, baseline-cell, as well as squamous cellular Ca are the deadliest subtypes of SC. Under this research, distinct SC kinds will be identified and classified utilizing machine learning and image processing techniques. Dermoscopy imaging are taken into consideration as input during the pre-processing phase. All undesirable hairs on skin lesion are removed using dull razor technique, and the picture is then smoothed using a Gaussian filtering. Median filter is employed to reduce noisiness and protect lesion’s borders. Colour-based k-means grouping is carried out in the segmentation process because colour is a key element in determining the kind of malignancy. With the help of Asymmetric, Boundary, Colour, Diameter (ABCD) and Gray Level Cooccurrence (GLC) matrix, statistic and textural retraction of features is carried out. The ISIC 2019 datasets, which consists of 8 distinct kinds of dermoscopy pictures, is used for investigation conducted. Multi-class Support Vector Machine (SVM) was used for classifying with an accurateness of roughly 96.25 percent.

A brain tumor occurs when abnormal cells form within the brain. Many people suffer from brain tumor, and it is a serious and dangerous disease. The detection and classification of brain tumor is an important and difficult task in the medical field. The brain tumor detection technique in the MRI images is very significant in many symptomatic and cure applications. But, as the data in MRI images are of highquality, tumor detection and classification are very hard in this process. Medical imaging plays a major role in properly diagnosing the disease, wherein an essential part in detecting the tumor is image segmentation. The segmentation provides an automatic brain tumor detection technique in order to increase the precision of the yields while also decreasing the diagnosis time. The goal is to detect the tumor from the MRI images and extract the features from the segmented tumor and finally classify it. The image detection and classification process includes image acquisition, image pre-processing, denoising, image segmentation, feature extraction and classification. The input image is pre-processed using Weiner and median filtering and the noise is removed using Edge Adaptive Total Variation Denoising (EATVD) technique. Once the noise is removed from the image, it is used for segmentation process, where Mean Shift Clustering is used. The segmented tumor undergoes features extraction stage, where Gray Level Co- occurrence Matrix (GLCM) features are used. In the last stage, images are classified either as tumorous or non-tumorous. Classification is done using Genetic Algorithm Support Vector Machine (GA-SVM), Deep Learning with Convolutional Neural Network (CNN). Early detection of the tumor region can be achieved without much time lapse in the calculation by using this efficient classifier model. This system presents a prototype for detecting objects based on GA-SVM that classifies images and assesses whether the image is cancerous. While comparing the accuracy and computational time of these classifiers, CNN would provide high accuracy and GASVM with lesser simulation time. The simulation results obtained for brain tumor detection and analysis are done with minimum computational time and with reasonable accuracy. This proposed system is tested using SPL dataset, which consists of 20 cases with 40 image samples of T2 FLAIR weighted MRI image and implemented using MATLAB software.