Comparative Analysis of Machine Learning Models for Automated Skin Cancer Detection: Advancements in Diagnostic Accuracy and AI Integration

Аннотация

Skin cancer detection remains a critical challenge in dermatology, with early diagnosis significantly improving patient outcomes. This study presents a comparative analysis of machine learning models for automated skin cancer detection, highlighting the superior performance of Convolutional Neural Networks (CNNs). The CNN model achieved the highest accuracy (92.5%), sensitivity (91.8%), and specificity (93.1%) compared to other algorithms such as Support Vector Machines (SVMs) and Random Forests. The use of advanced preprocessing techniques and diverse datasets ensured the model's robustness and generalizability. While the findings demonstrate the potential of deep learning in dermatological diagnostics, limitations such as model interpretability and dataset diversity were identified. This research underscores the transformative role of AI in improving diagnostic accuracy, enabling early detection, and addressing healthcare disparities, particularly in resource- constrained settings. Future work aims to enhance model explainability and expand its applicability across diverse populations. Keywords: skin cancer detection, machine learning, Convolutional Neural Networks, accuracy, sensitivity, specificity, preprocessing techniques, dataset diversity, deep learning, dermatological diagnostics, artificial intelligence. The American Journal of Medical Sciences and Pharmaceutical Research 15 https://www.theamericanjournals.com/index.php/tajmspr The American Journal of Medical Sciences and Pharmaceutical Research INTRODUCTION: nature of AI, has raised concerns about their interpretability and trustworthiness in clinical Skin cancer is one of the most prevalent and life- applications (Holzinger et al., 2019). Addressing these threatening conditions worldwide, with its incidence challenges requires the development of robust, increasing at an alarming rate over the past few explainable, and scalable AI systems that can seamlessly decades (American Cancer Society, 2024). According to integrate into existing healthcare workflows. recent statistics, over 5 million cases of skin cancer are diagnosed annually in the United States alone, making The current study aims to address these gaps by it the most diagnosed form of cancer. Among the proposing a novel deep learning-based approach for various types of skin cancer, melanoma is the skin cancer detection. Leveraging state-of-the-art CNN deadliest, responsible for most skin cancer-related architectures and advanced image preprocessing deaths, while non-melanoma cancers, such as basal techniques, the proposed system is designed to achieve cell carcinoma (BCC) and squamous cell carcinoma high diagnostic accuracy while maintaining robustness (SCC), are more common but less aggressive (Rogers et and scalability. Furthermore, this study conducts a al., 2021). Early detection and accurate diagnosis play comprehensive comparative analysis of the proposed a crucial role in the successful treatment and model's performance against existing methods, management of skin cancer, as survival rates highlighting its advantages and limitations. dramatically decrease when the disease progresses to By advancing the understanding of AI's role in skin advanced stages. cancer detection, this research contributes to the Traditional methods for diagnosing skin cancer rely on broader effort to improve cancer diagnostics and dermatologists' visual inspection of skin lesions, often outcomes. The findings have the potential to inform aided by dermoscopy. However, these methods are future developments in AI-driven healthcare, ultimately highly subjective and depend on the clinician's enhancing the accessibility and quality of expertise, which can lead to variability in diagnostic dermatological care worldwide. accuracy. Studies have shown that even experienced LITERATURE REVIEW dermatologists may misdiagnose skin cancer, Advancements in AI and machine learning have particularly in cases of atypical lesions or rare subtypes revolutionized medical diagnostics by providing (Esteva et al., 2017). This highlights the pressing need innovative tools for disease detection and monitoring. for more reliable, objective, and scalable diagnostic Several studies have explored the application of deep tools that can assist healthcare professionals in learning algorithms in the field of dermatology. Esteva identifying skin cancer at its earliest stages. et al. (2017) were among the pioneers in demonstrating In recent years, the integration of artificial intelligence that deep learning models could achieve dermatologist- (AI) and machine learning (ML) in medical diagnostics level accuracy in diagnosing skin cancer. Using a large has gained significant attention due to its potential to dataset of labeled dermoscopic images, the authors enhance diagnostic precision and efficiency. Deep trained a CNN that significantly outperformed learning, a subset of AI, has demonstrated remarkable traditional methods. success in analyzi