Two-Stages Input Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer Diagnosis

Опубликована: Ноя. 2, 2023

This research paper presents a deep learning approach to early detection of skin cancer using image augmentation techniques. The authors propose two-stage technique that involves the use geometric and generative adversarial network (GAN) classify lesions as either benign or malignant. utilized public HAM10000 dataset test proposed model. Several pre-trained models CNN were employed, namely Xception, Inceptionv3, Resnet152v2, EfficientnetB7, InceptionresnetV2, VGG19. Our achieved accuracy, precision, recall, F1-score 96.90%, 97.07%, 96.87%, 96.97%, respectively, which is higher than performance by other state-of-the-art methods. also discusses SHapley Additive exPlanations (SHAP), an interpretable for diagnosis, can help clinicians understand reasoning behind diagnosis improve trust in system. Overall, method promising automated could patient outcomes reduce healthcare costs.

Язык: Английский

---

Optimized clustering-based fusion for skin lesion image classification: Leveraging marine predators algorithm

Intelligent Decision Technologies, Год журнала: 2024, Номер 18(3), С. 2511 - 2536

Опубликована: Май 31, 2024

This manuscript presents a comprehensive approach to enhance the accuracy of skin lesion image classification based on HAM10000 and BCN20000 datasets. Building prior feature fusion models, this research introduces an optimized cluster-based address limitations observed in our previous methods. The study proposes two novel strategies, KFS-MPA (using K-means) DFS-MPA DBSCAN), for classification. These approaches leverage clustering-based deep marine predator algorithm (MPA). Ten fused sets are evaluated using three classifiers both datasets, their performance is compared terms dimensionality reduction improvement. results consistently demonstrate that outperforms other methods, achieving notable highest levels. ROC-AUC curves further support superiority DFS-MPA, highlighting its exceptional discriminative capabilities. Five-fold cross-validation tests comparison with previously proposed method (FOWFS-AJS) performed, confirming effectiveness enhancing performance. statistical validation Friedman test Bonferroni-Dunn also supports as promising among findings emphasize significance establish preferred choice study.

Язык: Английский

---

Improved FCM based Segmentation and Self Improved Tuna Swarm Optimized Hybrid Classifier for Skin Cancer Detection

Опубликована: Май 3, 2024

Язык: Английский

---

A Review on Utilizing Machine Learning Classification Algorithms for Skin Cancer

Journal of Applied Science and Technology Trends, Год журнала: 2024, Номер 5(2), С. 60 - 71

Опубликована: Авг. 22, 2024

Skin cancer is one of the most prevalent forms globally, with rising incidence rates posing significant challenges to healthcare systems. Early detection and accurate diagnosis are critical for effective treatment patient outcomes. In recent years, machine learning (ML) algorithms have emerged as powerful tools analyzing medical imaging data assisting clinicians in diagnosing skin cancer. This review paper provides a comprehensive overview ML classification context diagnosis. We discuss various types cancer, including melanoma, basal cell carcinoma, squamous along their characteristics diagnostic challenges. Furthermore, we current state-of-the-art techniques, such support vector machines (SVM), K-Nearest Neighbor (KNN), convolutional neural network (CNN), highlighting strengths limitations classification. A systematic search academic databases, Scopus, ResearchGate, Google Scholar, IEEE Xplore, Wiley Online Library, Elsevier, ScienceDirect, Springer, was conducted. Continued evolution promises enhanced accuracy personalized strategies.

Язык: Английский

---

ADVANCED SKIN CANCER DETECTION USING CONVOLUTIONAL NEURAL NETWORKS AND TRANSFER LEARNING

Middle East Journal of Science, Год журнала: 2024, Номер unknown

Опубликована: Дек. 18, 2024

This study investigates the effectiveness of MobileNetV2 transfer learning method and a deep based Convolutional Neural Network (CNN) model in categorization malignant benign skin lesions cancer diagnosis. Since is disease that can be cured with early detection but fatal if delayed, accurate diagnosis great importance. The was trained architecture performed classification task high accuracy on images lesions. Metrics such as accuracy, recall, precision F1 score obtained during training validation processes support performance model. 92.97%, Recall 92.71%, Precision 94.70% 93.47%. results show CNN-based reliable effective tool for diagnosis, small fluctuations phase require further data hyperparameter optimization to improve generalization ability demonstrates models enhanced offer powerful solution medical image problems have potential contribute development systems healthcare field.

Язык: Английский

---