Skin Lesion Classification Based on Surface Fractal Dimensions and Statistical Color Cluster Features Using an Ensemble of Machine Learning Techniques

LightDPH: Lightweight Dual-Projection-Head Hierarchical Contrastive Learning for Skin Lesion Classification

Effective skin cancer detection is crucial for early intervention and improved treatment outcomes. Previous studies have primarily focused on enhancing the performance of skin lesion classification models. However, there is a growing need to consider the practical requirements of real-world scenarios, such as portable applications that require lightweight models embedded in devices. Therefore, this study aims to propose a novel method that can address the major-type misclassification problem with a lightweight model. This study proposes an innovative Lightweight Dual Projection-Head Hierarchical contrastive learning (LightDPH) method. We introduce a dual projection-head mechanism to a contrastive learning framework. This mechanism is utilized to train a model with our proposed multi-level contrastive loss (MultiCon Loss), which can effectively learn hierarchical information from samples. Meanwhile, we present a distance-based weight (DBW) function to adjust losses based on hierarchical levels. This unique combination of MultiCon Loss and DBW function in LightDPH tackles the problem of major-type misclassification with lightweight models and enhances the model's sensitivity in skin lesion classification. The experimental results demonstrate that LightDPH significantly reduces the number of parameters by 52.6% and computational complexity by 29.9% in GFLOPs while maintaining high classification performance comparable to state-of-the-art methods. This study also presented a novel evaluation metric, model efficiency score (MES), to evaluate the cost-effectiveness of models with scaling and classification performance. The proposed LightDPH effectively mitigates major-type misclassification and works in a resource-efficient manner, making it highly suitable for clinical applications in resource-constrained environments. To the best of our knowledge, this is the first work that develops an effective lightweight hierarchical classification model for skin lesion detection.

Integrating color histogram analysis and convolutional neural networks for skin lesion classification

The color of skin lesions is a crucial diagnostic feature for identifying malignant melanoma and other skin diseases. Typical colors associated with melanocytic lesions include tan, brown, black, red, white, and blue-gray. This study introduces a novel feature: the number of colors present in lesions, which can indicate the severity of skin diseases and help distinguish melanomas from benign lesions. We propose a color histogram analysis, a traditional image processing technique, to analyze the pixels of skin lesions from three publicly available datasets: PH2, ISIC2016, and Med-Node, which include dermoscopic and non-dermoscopic images. While the PH2 dataset contains ground truth about skin lesion colors, the ISIC2016 and Med-Node datasets lack such annotations; our algorithm establishes this ground truth using the color histogram analysis based on the PH2 dataset. We then design and train a 19-layer Convolutional Neural Network (CNN) with different skip connections of residual blocks to classify lesions into three categories based on the number of colors present. The DeepDream algorithm is utilized to visualize the learned features of different layers, and multiple configurations of the proposed CNN are tested, achieving the highest weighted F1-score of 75.00 % on the test set. LIME is subsequently applied to identify the most important features influencing the model's decision-making. The findings demonstrate that the number of colors in lesions is a significant feature for describing skin conditions. The proposed CNN, particularly with three skip connections, shows strong potential for clinical application in diagnosing melanoma, supporting its use alongside traditional diagnostic methods.

Development, Application and Utility of a Machine Learning Approach for Melanoma and Non-Melanoma Lesion Classification Using Counting Box Fractal Dimension

The diagnosis and identification of melanoma are not always accurate, even for experienced dermatologists. Histopathology continues to be the gold standard, assessing specific parameters such as the Breslow index. However, it remains invasive and may lack effectiveness. Therefore, leveraging mathematical modeling and informatics has been a pursuit of diagnostic methods favoring early detection. Fractality, a mathematical parameter quantifying complexity and irregularity, has proven useful in melanoma diagnosis. Nonetheless, no studies have implemented this metric to feed artificial intelligence algorithms for the automatic classification of dermatological lesions, including melanoma. Hence, this study aimed to determine the combined utility of fractal dimension and unsupervised low-computational-requirements machine learning models in classifying melanoma and non-melanoma lesions. We analyzed 39,270 dermatological lesions obtained from the International Skin Imaging Collaboration. Box-counting fractal dimensions were calculated for these lesions. Fractal values were used to implement classification methods by unsupervised machine learning based on principal component analysis and iterated K-means (100 iterations). A clear separation was observed, using only fractal dimension values, between benign or malignant lesions (sensibility 72.4% and specificity 50.1%) and melanoma or non-melanoma lesions (sensibility 72.8% and specificity 50%) and subsequently, the classification quality based on the machine learning model was ≈80% for both benign and malignant or melanoma and non-melanoma lesions. However, the grouping of metastatic melanoma versus non-metastatic melanoma was less effective, probably due to the small sample size included in MM lesions. Nevertheless, we could suggest a decision algorithm based on fractal dimension for dermatological lesion discrimination. On the other hand, it was also determined that the fractal dimension is sufficient to generate unsupervised artificial intelligence models that allow for a more efficient classification of dermatological lesions.

Alzheimer's disease: new insight in assessing of amyloid plaques morphologies using multifractal geometry based on Naive Bayes optimized by random forest algorithm

Alzheimer's disease (AD) is a physical illness, which damages a person's brain; it is the most common cause of dementia. AD can be characterized by the formation of amyloid-beta (Aβ) deposits. They exhibit diverse morphologies that range from diffuse to dense-core plaques. Most of the histological images cannot be described precisely by traditional geometry or methods. Therefore, this study aims to employ multifractal geometry in assessing and classifying amyloid plaque morphologies. The classification process is based on extracting the most descriptive features related to the amyloid-beta (Aβ) deposits using the Naive Bayes classifier. To eliminate the less important features, the Random Forest algorithm has been used. The proposed methodology has achieved an accuracy of 99%, sensitivity of 100%, and specificity of 98.5%. This study employed a new dataset that had not been widely used before.

Enhanced detonators detection in X-ray baggage inspection by image manipulation and deep convolutional neural networks

Detecting detonators is a challenging task because they can be easily mis-classified as being a harmless organic mass, especially in high baggage throughput scenarios. Of particular interest is the focus on automated security X-ray analysis for detonators detection. The complex security scenarios require increasingly advanced combinations of computer-assisted vision. We propose an extensive set of experiments to evaluate the ability of Convolutional Neural Network (CNN) models to detect detonators, when the quality of the input images has been altered through manipulation. We leverage recent advances in the field of wavelet transforms and established CNN architectures-as both of these can be used for object detection. Various methods of image manipulation are used and further, the performance of detection is evaluated. Both raw X-ray images and manipulated images with the Contrast Limited Adaptive Histogram Equalization (CLAHE), wavelet transform-based methods and the mixed CLAHE RGB-wavelet method were analyzed. The results showed that a significant number of operations, such as: edges enhancements, altered color information or different frequency components provided by wavelet transforms, can be used to differentiate between almost similar features. It was found that the wavelet-based CNN achieved the higher detection performance. Overall, this performance illustrates the potential for a combined use of the manipulation methods and deep CNNs for airport security applications.

Refining skin lesions classification performance using geometric features of superpixels

This paper introduces superpixels to enhance the detection of skin lesions and to discriminate between melanoma and nevi without false negatives, in dermoscopy images. An improved Simple Linear Iterative Clustering (iSLIC) superpixels algorithm for image segmentation in digital image processing is proposed. The local graph cut method to identify the region of interest (i.e., either the nevi or melanoma lesions) has been adopted. The iSLIC algorithm is then exploited to segment sSPs. iSLIC discards all the SPs belonging to image background based on assigned labels and preserves the segmented skin lesions. A shape and geometric feature extraction task is performed for each segmented SP. The extracted features are fed into six machine learning algorithms such as: random forest, support vector machines, AdaBoost, k-nearest neighbor, decision trees (DT), Gaussian Naïve Bayes and three neural networks. These include Pattern recognition neural network, Feed forward neural network, and 1D Convolutional Neural Network for classification. The method is evaluated on the 7-Point MED-NODE and PAD-UFES-20 datasets and the results have been compared to the state-of-art findings. Extensive experiments show that the proposed method outperforms the compared existing methods in terms of accuracy.

A survey, review, and future trends of skin lesion segmentation and classification

The Computer-aided Diagnosis or Detection (CAD) approach for skin lesion analysis is an emerging field of research that has the potential to alleviate the burden and cost of skin cancer screening. Researchers have recently indicated increasing interest in developing such CAD systems, with the intention of providing a user-friendly tool to dermatologists to reduce the challenges encountered or associated with manual inspection. This article aims to provide a comprehensive literature survey and review of a total of 594 publications (356 for skin lesion segmentation and 238 for skin lesion classification) published between 2011 and 2022. These articles are analyzed and summarized in a number of different ways to contribute vital information regarding the methods for the development of CAD systems. These ways include: relevant and essential definitions and theories, input data (dataset utilization, preprocessing, augmentations, and fixing imbalance problems), method configuration (techniques, architectures, module frameworks, and losses), training tactics (hyperparameter settings), and evaluation criteria. We intend to investigate a variety of performance-enhancing approaches, including ensemble and post-processing. We also discuss these dimensions to reveal their current trends based on utilization frequencies. In addition, we highlight the primary difficulties associated with evaluating skin lesion segmentation and classification systems using minimal datasets, as well as the potential solutions to these difficulties. Findings, recommendations, and trends are disclosed to inform future research on developing an automated and robust CAD system for skin lesion analysis.

Concatenated Xception-ResNet50 - A novel hybrid approach for accurate skin cancer prediction

Skin cancer is a malignant disease that affects millions of people around the world every year. It is an invasive disease characterised by an abnormal proliferation of skin cells in the body that multiply and spread through the lymph nodes, killing the surrounding tissue. The number of skin cancer cases is on the rise due to lifestyle changes and sun-seeking behaviour. As skin cancer is a deadly disease, early diagnosis and grading are crucial to save lives. In this work, state-of-the-art AI approaches are applied to develop a unique deep learning model that integrates Xception and ResNet50. This network achieves maximum accuracy by combining the properties of two robust networks. The proposed concatenated Xception-ResNet50 (X-R50) model can classify skin tumours as basal cell carcinoma, melanoma, melanocytic nevi, dermatofibroma, actinic keratoses and intraepithelial carcinoma, vascular and non-cancerous benign keratosis-like lesions. The performance of the proposed method is compared with a DeepCNN and other state-of-the-art transfer learning models. The Human Against Machine (HAM10000) dataset assesses the suggested method's performance. For this study, 10,500 skin images were used. The model is trained and tested with the sliding window technique. The proposed concatenated X-R50 model is cutting-edge, with a 97.8% prediction accuracy. The performance of the model is also validated by a statistical hypothesis test using analysis of variance (ANOVA). The reported approach is both accurate and efficient and can help dermatologists and clinicians detect skin cancer at an early stage of the clinical process.

Automatic Modulation Recognition Based on the Optimized Linear Combination of Higher-Order Cumulants

Automatic modulation recognition (AMR) is used in various domains-from general-purpose communication to many military applications-thanks to the growing popularity of the Internet of Things (IoT) and related communication technologies. In this research article, we propose an innovative idea of combining the classical mathematical technique of computing linear combinations (LCs) of cumulants with a genetic algorithm (GA) to create super-cumulants. These super-cumulants are further used to classify five digital modulation schemes on fading channels using the K-nearest neighbor (KNN). Our proposed classifier significantly improves the percentage recognition accuracy at lower SNRs when using smaller sample sizes. A comparison with existing techniques manifests the supremacy of our proposed classifier.

Fractal Dimension Analysis of Melanocytic Nevi and Melanomas in Normal and Polarized Light-A Preliminary Report

Clinical diagnosis of pigmented lesions can be a challenge in everyday practice. Benign and dysplastic nevi and melanomas may have similar clinical presentations, but completely different prognoses. Fractal dimensions of shape and texture can describe the complexity of the pigmented lesion structure. This study aims to apply fractal dimension analysis to differentiate melanomas, dysplastic nevi, and benign nevi in polarized and non-polarized light. A total of 87 Eighty-four patients with 97 lesions were included in this study. All examined lesions were photographed under polarized and non-polarized light, surgically removed, and examined by a histopathologist to establish the correct diagnosis. The obtained images were then processed and analyzed. Area, perimeter, and fractal dimensions of shape and texture were calculated for all the lesions under polarized and non-polarized light. The fractal dimension of shape in polarized light enables differentiating melanomas, dysplastic nevi, and benign nevi. It also makes it possible to distinguish melanomas from benign and dysplastic nevi under non-polarized light. The fractal dimension of texture allows distinguishing melanomas from benign and dysplastic nevi under polarized light. All examined parameters of shape and texture can be used for developing an automatic computer-aided diagnosis system. Polarized light is superior to non-polarized light for imaging texture details.

A Workflow for Computer-Aided Evaluation of Keloid Based on Laser Speckle Contrast Imaging and Deep Learning

A keloid results from abnormal wound healing, which has different blood perfusion and growth states among patients. Active monitoring and treatment of actively growing keloids at the initial stage can effectively inhibit keloid enlargement and has important medical and aesthetic implications. LSCI (laser speckle contrast imaging) has been developed to obtain the blood perfusion of the keloid and shows a high relationship with the severity and prognosis. However, the LSCI-based method requires manual annotation and evaluation of the keloid, which is time consuming. Although many studies have designed deep-learning networks for the detection and classification of skin lesions, there are still challenges to the assessment of keloid growth status, especially based on small samples. This retrospective study included 150 untreated keloid patients, intensity images, and blood perfusion images obtained from LSCI. A newly proposed workflow based on cascaded vision transformer architecture was proposed, reaching a dice coefficient value of 0.895 for keloid segmentation by 2% improvement, an error of 8.6 ± 5.4 perfusion units, and a relative error of 7.8% ± 6.6% for blood calculation, and an accuracy of 0.927 for growth state prediction by 1.4% improvement than baseline.

Image Moment-Based Features for Mass Detection in Breast US Images via Machine Learning and Neural Network Classification Models

Differentiating between malignant and benign masses using machine learning in the recognition of breast ultrasound (BUS) images is a technique with good accuracy and precision, which helps doctors make a correct diagnosis. The method proposed in this paper integrates Hu’s moments in the analysis of the breast tumor. The extracted features feed a k-nearest neighbor (k-NN) classifier and a radial basis function neural network (RBFNN) to classify breast tumors into benign and malignant. The raw images and the tumor masks provided as ground-truth images belong to the public digital BUS images database. Certain metrics such as accuracy, sensitivity, precision, and F1-score were used to evaluate the segmentation results and to select Hu’s moments showing the best capacity to discriminate between malignant and benign breast tissues in BUS images. Regarding the selection of Hu’s moments, the k-NN classifier reached 85% accuracy for moment M1 and 80% for moment M5 whilst RBFNN reached an accuracy of 76% for M1. The proposed method might be used to assist the clinical diagnosis of breast cancer identification by providing a good combination between segmentation and Hu’s moments.

Detection and Classification of Knee Injuries from MR Images Using the MRNet Dataset with Progressively Operating Deep Learning Methods

This study aimed to build progressively operating deep learning models that could detect meniscus injuries, anterior cruciate ligament (ACL) tears and knee abnormalities in magnetic resonance imaging (MRI). The Stanford Machine Learning Group MRNet dataset was employed in the study, which included MRI image indexes in the coronal, sagittal, and axial axes, each having 1130 trains and 120 validation items. The study is divided into three sections. In the first section, suitable images are selected to determine the disease in the image index based on the disturbance under examination. It is also used to identify images that have been misclassified or are noisy and/or damaged to the degree that they cannot be utilised for diagnosis in the first section. The study employed the 50-layer residual networks (ResNet50) model in this section. The second part of the study involves locating the region to be focused on based on the disturbance that is targeted to be diagnosed in the image under examination. A novel model was built by integrating the convolutional neural networks (CNN) and the denoising autoencoder models in the second section. The third section is dedicated to making a diagnosis of the disease. In this section, a novel ResNet50 model is trained to identify disease diagnoses or abnormalities, independent of the ResNet50 model used in the first section. The images that each model selects as output after training are referred to as progressively operating deep learning methods since they are supplied as an input to the following model.