SkinNet-INIO: Multiclass Skin Lesion Localization and Classification Using Fusion-Assisted Deep Neural Networks and Improved Nature-Inspired Optimization Algorithm

Diagnostics, Journal Year: 2023, Volume and Issue: 13(18), P. 2869 - 2869

Published: Sept. 6, 2023

Background: Using artificial intelligence (AI) with the concept of a deep learning-based automated computer-aided diagnosis (CAD) system has shown improved performance for skin lesion classification. Although convolutional neural networks (DCNNs) have significantly many image classification tasks, it is still difficult to accurately classify lesions because lack training data, inter-class similarity, intra-class variation, and inability concentrate on semantically significant parts. Innovations: To address these issues, we proposed an learning best feature selection framework multiclass in dermoscopy images. The performs preprocessing step at initial contrast enhancement using new technique that based dark channel haze top–bottom filtering. Three pre-trained models are fine-tuned next trained transfer concept. In fine-tuning process, added removed few additional layers lessen parameters later selected hyperparameters genetic algorithm (GA) instead manual assignment. purpose hyperparameter GA improve performance. After that, deeper layer each network features extracted. extracted fused novel serial correlation-based approach. This reduces vector length serial-based approach, but there little redundant information. We anti-Lion optimization this issue. finally classified machine algorithms. Main Results: experimental process was conducted two publicly available datasets, ISIC2018 ISIC2019. Employing obtained accuracy 96.1 99.9%, respectively. Comparison also state-of-the-art techniques shows accuracy. Conclusions: successfully enhances cancer region. Moreover, framework. fusion version maintains shorten computational time.

Language: Английский

---

Fuzzy Logic with Deep Learning for Detection of Skin Cancer

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(15), P. 8927 - 8927

Published: Aug. 3, 2023

Melanoma is the deadliest type of cancerous cell, which developed when melanocytes, melanin producing starts its uncontrolled growth. If not detected and cured in situ, it might decrease chances survival patients. The diagnosis a melanoma lesion still challenging task due to visual similarities with benign lesions. In this paper, fuzzy logic-based image segmentation along modified deep learning model proposed for skin cancer detection. highlight paper dermoscopic enhancement using pre-processing techniques, infusion mathematical logics, standard deviation methods, L-R defuzzification method enhance results segmentation. These steps are improve visibility by removing artefacts such as hair follicles, scales, etc. Thereafter, enhanced histogram equalization method, segmented prior performing detection phase. employs neural network algorithm, You Look Only Once (YOLO), established on application Deep convolutional (DCNN) from digital images. YOLO composed series DCNN layers we have added more depth adding layer residual connections. Moreover, introduced feature concatenation at different combines multi-scale features. Our experimental confirm that provides better accuracy score faster than most pre-existing classifiers. classifier trained 2000 8695 images ISIC 2017 2018 datasets, whereas PH2 datasets both previously mentioned used testing algorithm.

Language: Английский

---

Optimization Convolutional Neural Network for Automatic Skin Lesion Diagnosis Using a Genetic Algorithm

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(5), P. 3248 - 3248

Published: March 3, 2023

Examining and predicting skin cancer from lesion images is challenging due to the complexity of images. Early detection treatment disease can prevent mortality as it be curable. Computer-aided diagnosis (CAD) provides a second opinion for dermatologists they classify type with high accuracy their ability show various clinical identification features locally globally. Convolutional neural networks (CNNs) have significantly improved performance CAD systems medical image segmentation classifications. However, tuning CNNs are since search space all possible hyperparameter configurations substantially vast. In this paper, we adopt genetic algorithm automatically configure CNN model an accurate, reliable, robust automated classification early diagnosis. The optimized uses four public datasets train able detect abnormalities based on in different orientations. achieves best scores each DICE coefficients, precision measure, F-score. These compare better than other existing methods. Considering success model, could valuable method implement settings.

Language: Английский

---

FDUM-Net: An enhanced FPN and U-Net architecture for skin lesion segmentation

Biomedical Signal Processing and Control, Journal Year: 2024, Volume and Issue: 91, P. 106037 - 106037

Published: Feb. 7, 2024

Language: Английский

---

SL-R-CNN-HHO: multi-class skin lesion classification using region-based convolutional neural networks and harris hawk optimization on the HAM dataset

International Journal of Information Technology, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 8, 2025

Language: Английский

---

Precision and efficiency in skin cancer segmentation through a dual encoder deep learning model

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 9, 2025

Skin cancer is a prevalent health concern, and accurate segmentation of skin lesions crucial for early diagnosis. Existing methods lesion often face trade-offs between efficiency feature extraction capabilities. This paper proposes Dual Segmentation (DuaSkinSeg), deep-learning model, to address this gap by utilizing dual encoders improved performance. DuaSkinSeg leverages pre-trained MobileNetV2 efficient local extraction. Subsequently, Vision Transformer-Convolutional Neural Network (ViT-CNN) encoder-decoder architecture extracts higher-level features focusing on long-range dependencies. approach aims combine the with capabilities ViT encoder To evaluate DuaSkinSeg's effectiveness, we conducted experiments three publicly available benchmark datasets: ISIC 2016, 2017, 2018. The results demonstrate that achieves competitive performance compared existing methods, highlighting potential segmentation.

Language: Английский

---

Systematic review of approaches to detection and classification of skin cancer using artificial intelligence: Development and prospects

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 178, P. 108742 - 108742

Published: June 14, 2024

Language: Английский

---

Pixel-guided pattern alignment based Hopfield Neural Networks for generalize cancer diagnosis

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 103, P. 107397 - 107397

Published: Jan. 7, 2025

Language: Английский

---

Classification of skin cancer from dermatological image by multimodal method

Published: Jan. 15, 2025

Skin cancer remains a critical health issue, with over 1.2 million new cases diagnosed annually. Early detection is crucial in reducing mortality rates, yet challenges diagnosis persist due to variability dermoscopic image quality. Traditional methods for skin lesion classification are cumbersome and involve significant manual preprocessing. This study introduces an innovative approach using deep learning automate feature extraction enhance diagnostic accuracy. We investigate ensemble of advanced neural networks (VGG, ResNet, GoogleNet, Visiontransformer) combined multimodal method that integrates patient metadata features. Our dataset includes 9,013 training images 1,002 testing across seven categories pigmented lesions. The approach, mainly the DenseNet121-Mul model, demonstrated superior performance precision, recall, F1-score, achieving F1-score 0.91. findings highlight potentiality multiple deeplearning models combining diverse data types advance accuracy efficiency computer-aided diagnostics dermatology, paving way systems could match dermatologist-level capabilities.

Language: Английский

---

Melanoma lesion localization using UNet and explainable AI

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: March 13, 2025

Language: Английский

---