Enhancing Skin Cancer Diagnosis Through Fine‐Tuning of Pretrained Models: A Two‐Phase Transfer Learning Approach

International Journal of Breast Cancer, Год журнала: 2025, Номер 2025(1)

Опубликована: Янв. 1, 2025

Skin cancer is among the most prevalent types of worldwide, and early detection crucial for improving treatment outcomes patient survival rates. Traditional diagnostic methods, often reliant on visual examination manual evaluation, can be subjective time-consuming, leading to variability in accuracy. Recent developments machine learning, particularly using pretrained models fine-tuning techniques, offer promising advancements automating skin classification. This paper explores application a two-phase model HAM10000 dataset, which comprises wide range lesion images. The first phase employs transfer learning with frozen layers, followed by all layers second adapt more specifically dataset. I evaluate nine models, including VGG16, VGG19, InceptionV3, Xception (extreme inception), DenseNet121, assessing their performance based accuracy, precision, recall, F1 score metrics. VGG16 model, after fine-tuning, achieved highest test set accuracy 99.3%, highlighting its potential highly accurate study provides important insights clinicians researchers, demonstrating efficacy advanced enhancing supporting clinical decision-making dermatology.

Язык: Английский

---

An intelligent framework for skin cancer detection and classification using fusion of Squeeze-Excitation-DenseNet with Metaheuristic-driven ensemble deep learning models

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 3, 2025

Skin cancer is the most dominant and critical method of cancer, which arises all over world. Its damaging effects can range from disfigurement to major medical expenditures even death if not analyzed preserved timely. Conventional models skin recognition require a complete physical examination by specialist, time-wasting in few cases. Computer-aided medicinal analytical methods have gained massive popularity due their efficiency effectiveness. This model assist dermatologists initial significant for early diagnosis. An automatic classification utilizing deep learning (DL) help doctors perceive kind lesion improve patient's health. The one hot topics research field, along with development DL structure. manuscript designs develops Detection Cancer Using an Ensemble Deep Learning Model Gray Wolf Optimization (DSC-EDLMGWO) method. proposed DSC-EDLMGWO relies on biomedical imaging. presented initially involves image preprocessing stage at two levels: contract enhancement using CLAHE noise removal wiener filter (WF) model. Furthermore, utilizes SE-DenseNet method, fusion squeeze-and-excitation (SE) module DenseNet extract features. For process, ensemble models, namely long short-term memory (LSTM) technique, extreme machine (ELM) model, stacked sparse denoising autoencoder (SSDA) employed. Finally, gray wolf optimization (GWO) optimally adjusts models' hyperparameter values, resulting more excellent performance. effectiveness approach evaluated benchmark database, outcomes measured across various performance metrics. experimental validation portrayed superior accuracy value 98.38% 98.17% under HAM10000 ISIC datasets other techniques.

Язык: Английский

---

Improved performance on melanoma skin cancer classification using deep learning based ensemble technique

Intelligent Data Analysis, Год журнала: 2025, Номер unknown

Опубликована: Март 11, 2025

Skin cancer, particularly melanoma, arises from DNA damage that leads to abnormal cell growth in the epidermis. Early detection is crucial as melanoma can spread rapidly, but it highly curable if identified promptly. Detecting and diagnosing early are essential reduce mortality rates associated with this type of cancer. In literature, various ensemble techniques have been proposed improve performance. This paper introduces a deep learning based method aimed at enhancing accuracy skin cancer detection. Additionally, presents thorough performance evaluation five techniques. Initially, dataset underwent pre-processing, involving removal artifacts through hair removal, achieving balance distribution images for each class image augmentation Then, architecture 16 pre-trained models was modified by adding additional layers their The achieved highest were selected ensembling. Since VGG16, MobileNetV2, DenseNet169 accuracy, they chosen Five techniques, namely, weighted average, voting, bagging, boosting, stacking, applied architectures fine-tuned such classify images. experiments performed on combined HAM10000 ISIC2019, which contains seven lesion classes. results demonstrate average model achieves overall 81.99% classification 89.85%. positive outcomes affirm employing adjusted enhances performance, thereby demonstrating potential utility

Язык: Английский

---

Enhancing skin lesion classification: a CNN approach with human baseline comparison

PeerJ Computer Science, Год журнала: 2025, Номер 11, С. e2795 - e2795

Опубликована: Апрель 15, 2025

This study presents an augmented hybrid approach for improving the diagnosis of malignant skin lesions by combining convolutional neural network (CNN) predictions with selective human interventions based on prediction confidence. The algorithm retains high-confidence CNN while replacing low-confidence outputs expert assessments to enhance diagnostic accuracy. A model utilizing EfficientNetB3 backbone is trained datasets from ISIC-2019 and ISIC-2020 SIIM-ISIC melanoma classification challenges evaluated a 150-image test set. model’s are compared against 69 experienced medical professionals. Performance assessed using receiver operating characteristic (ROC) curves area under curve (AUC) metrics, alongside analysis resource costs. baseline achieves AUC 0.822, slightly below performance experts. However, improves true positive rate 0.782 reduces false 0.182, delivering better minimal involvement. offers scalable, resource-efficient solution address variability in image analysis, effectively harnessing complementary strengths humans CNNs.

Язык: Английский

---

Explainable machine learning and feature interpretation to predict survival outcomes in the treatment of lung cancer

Seminars in Oncology, Год журнала: 2025, Номер 52(3), С. 152364 - 152364

Опубликована: Май 24, 2025

Язык: Английский

---

Diagnosing Skin Cancer Using Shearlet Transform Multiresolution Computation

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Авг. 26, 2024

Skin cancer diagnosis relies on the accurate analysis of medical images to identify malignant and benign lesions. The Shearlet transform, a powerful mathematical tool for multiresolution analysis, has shown promise in enhancing detection classification skin cancer. This study investigates application transform-based diagnosis. known its ability capture anisotropic features directional information, provides comprehensive representation lesion at multiple scales orientations. We integrate transform with advanced image processing techniques extract discriminative from dermoscopic images. These are then utilized train machine learning classifier, specifically support vector (SVM), distinguish between proposed methodology is evaluated publicly available dataset, results demonstrate significant improvements diagnostic accuracy compared traditional methods. Our approach enhances feature extraction capabilities, leading more reliable precise diagnosis, ultimately contributing better patient outcomes.

Язык: Английский

---

Developing an early detection model for skin diseases using a hybrid deep neural network to enhance health independence in coastal communities

Eastern-European Journal of Enterprise Technologies, Год журнала: 2024, Номер 6(9 (132)), С. 71 - 85

Опубликована: Дек. 30, 2024

The object of study is a solution for early skin disease diagnosis by integrating hybrid deep neural networks – EfficientNetB7 Classification and YOLOv8 detection. system designed to classify five conditions: Melanoma, Basal Cell Carcinoma (BCC), melanoma type cancer that originates from melanocytes, the cells produce pigment, Melanocytic Nevi (NV) nevus mole or dark spot on formed due accumulation Benign Keratosis-like Lesions (BKL) term group changes resemble keratosis but are non-cancerous, Seborrheic Keratoses other benign tumors enhance health diagnostics. problem be solved in this revolves around improving accurate diagnosis, particularly resource-limited underserved areas lack Accessible Diagnostic Tools Low Efficiency Current Methods. highlights EfficientNetB7's classification accuracy at 94 % YOLOv8's means average precision (mAP) 0.812 This processes images efficiently, providing detection outcomes with consistent performance multiple tests. results demonstrate model achieved an test data, while delivered mean 0.812. web-based efficiently processed provided outcomes. Furthermore, allowed different diseases assess malignancy risk. systems portable can used minimal setup, making them practical real-world diagnostic use. Scope Practical applications accessibility settings. website-based tool provides user-friendly platform accessible public healthcare providers, especially areas. Each application's high ease use make viable aids potentially access

Язык: Английский

---