A smartphone-based application for an early skin disease prognosis: Towards a lean healthcare system via computer-based vision

Advanced Engineering Informatics, Год журнала: 2023, Номер 57, С. 102036 - 102036

Опубликована: Май 30, 2023

Язык: Английский

---

Analysis of Artificial Intelligence-Based Approaches Applied to Non-Invasive Imaging for Early Detection of Melanoma: A Systematic Review

Cancers, Год журнала: 2023, Номер 15(19), С. 4694 - 4694

Опубликована: Сен. 23, 2023

Background: Melanoma, the deadliest form of skin cancer, poses a significant public health challenge worldwide. Early detection is crucial for improved patient outcomes. Non-invasive imaging techniques allow diagnostic accuracy; however, their use often limited due to need skilled practitioners trained interpret images in standardized fashion. Recent innovations artificial intelligence (AI)-based lesion image interpretation show potential AI early melanoma. Objective: The aim this study was evaluate current state AI-based used combination with non-invasive modalities including reflectance confocal microscopy (RCM), optical coherence tomography (OCT), and dermoscopy. We also aimed determine whether application can lead accuracy Methods: A systematic search conducted via Medline/PubMed, Cochrane, Embase databases eligible publications between 2018 2022. Screening methods adhered 2020 version PRISMA (Preferred Reporting Items Systematic Reviews Meta-Analyses) guidelines. Included studies utilized algorithms melanoma directly addressed review objectives. Results: retrieved 40 papers amongst three databases. All comparing performance dermatologists reported superior or equivalent improving In algorithm on dermoscopy dermatologists, achieved higher ROC (>80%) these comparative using dermoscopic images, mean sensitivity 83.01% specificity 85.58%. Studies evaluating machine learning conjunction OCT boasted 95%, while RCM rate 82.72%. Conclusions: Our results demonstrate robust improve outcomes through identification Further are needed assess generalizability across different populations types, standardization processing, further compare board-certified clinical applicability.

Язык: Английский

---

Blockchain-Federated and Deep-Learning-Based Ensembling of Capsule Network with Incremental Extreme Learning Machines for Classification of COVID-19 Using CT Scans

Bioengineering, Год журнала: 2023, Номер 10(2), С. 203 - 203

Опубликована: Фев. 3, 2023

Due to the rapid rate of SARS-CoV-2 dissemination, a conversant and effective strategy must be employed isolate COVID-19. When it comes determining identity COVID-19, one most significant obstacles that researchers overcome is propagation virus, in addition dearth trustworthy testing models. This problem continues difficult for clinicians deal with. The use AI image processing has made formerly insurmountable challenge finding COVID-19 situations more manageable. In real world, there handled about difficulties sharing data between hospitals while still honoring privacy concerns organizations. training global deep learning (DL) model, crucial handle fundamental such as user collaborative model development. For this study, novel framework designed compiles information from five different databases (several hospitals) edifies using blockchain-based federated (FL). validated through blockchain technology (BCT), FL trains on scale maintaining secrecy proposed divided into three parts. First, we provide method normalization can diversity collected sources several computed tomography (CT) scanners. Second, categorize patients, ensemble capsule network (CapsNet) with incremental extreme machines (IELMs). Thirdly, interactively BCT anonymity. Extensive tests employing chest CT scans comparison classification performance DL algorithms predicting protecting variety users, were undertaken. Our findings indicate improved effectiveness identifying patients achieved an accuracy 98.99%. Thus, our provides substantial aid medical practitioners their diagnosis

Язык: Английский

---

Automated Skin Cancer Detection and Classification using Cat Swarm Optimization with a Deep Learning Model

Engineering Technology & Applied Science Research, Год журнала: 2024, Номер 14(1), С. 12734 - 12739

Опубликована: Фев. 8, 2024

The application of Computer Vision (CV) and image processing in the medical sector is great significance, especially recognition skin cancer using dermoscopic images. Dermoscopy denotes a non-invasive imaging system that offers clear visuals cancers, allowing dermatologists to analyze identify various features crucial for lesion assessment. Over past few years, there has been an increasing fascination with Deep Learning (DL) applications recognition, particular focus on impressive results achieved by Neural Networks (DNNs). DL approaches, predominantly CNNs, have exhibited immense potential automating classification detection cancers. This study presents Automated Skin Cancer Detection Classification method Cat Swarm Optimization (ASCDC-CSODL). main objective ASCDC-CSODL enforce model recognize classify tumors In ASCDC-CSODL, Bilateral Filtering (BF) applied noise elimination U-Net employed segmentation process. Moreover, exploits MobileNet feature extraction Gated Recurrent Unit (GRU) approach used cancer. Finally, CSO algorithm alters hyperparameter values GRU. A wide-ranging simulation was performed evaluate performance model, demonstrating significantly improved over other approaches.

Язык: Английский

---

An effective multiclass skin cancer classification approach based on deep convolutional neural network

Cluster Computing, Год журнала: 2024, Номер unknown

Опубликована: Июнь 17, 2024

Abstract Skin cancer is one of the most dangerous types due to its immediate appearance and possibility rapid spread. It arises from uncontrollably growing cells, rapidly dividing cells in area body, invading other bodily tissues, spreading throughout body. Early detection helps prevent progress reaching critical levels, reducing risk complications need for more aggressive treatment options. Convolutional neural networks (CNNs) revolutionize skin diagnosis by extracting intricate features images, enabling an accurate classification lesions. Their role extends early detection, providing a powerful tool dermatologists identify abnormalities their nascent stages, ultimately improving patient outcomes. This study proposes novel deep convolutional network (DCNN) approach classifying The proposed DCNN model evaluated using two unbalanced datasets, namely HAM10000 ISIC-2019. compared with transfer learning models, including VGG16, VGG19, DenseNet121, DenseNet201, MobileNetV2. Its performance assessed four widely used evaluation metrics: accuracy, recall, precision, F1-score, specificity, AUC. experimental results demonstrate that outperforms (DL) models utilized these datasets. achieved highest accuracy ISIC-2019 $$98.5\%$$ 98.5 % $$97.1\%$$ 97.1 , respectively. These show how competitive successful overcoming problems caused class imbalance raising accuracy. Furthermore, demonstrates superior performance, particularly excelling terms recent studies utilize same which highlights robustness effectiveness DCNN.

Язык: Английский

---

Skin Cancer Detection and Classification Using Neural Network Algorithms: A Systematic Review

Diagnostics, Год журнала: 2024, Номер 14(4), С. 454 - 454

Опубликована: Фев. 19, 2024

In recent years, there has been growing interest in the use of computer-assisted technology for early detection skin cancer through analysis dermatoscopic images. However, accuracy illustrated behind state-of-the-art approaches depends on several factors, such as quality images and interpretation results by medical experts. This systematic review aims to critically assess efficacy challenges this research field order explain usability limitations highlight potential future lines work scientific clinical community. study, was carried out over 45 contemporary studies extracted from databases Web Science Scopus. Several computer vision techniques related image video processing diagnosis were identified. context, focus process included algorithms employed, result accuracy, validation metrics. Thus, yielded significant advancements using deep learning machine algorithms. Lastly, establishes a foundation research, highlighting contributions opportunities improve effectiveness learning.

Язык: Английский

---

Towards unbiased skin cancer classification using deep feature fusion

BMC Medical Informatics and Decision Making, Год журнала: 2025, Номер 25(1)

Опубликована: Янв. 31, 2025

Abstract This paper introduces SkinWiseNet (SWNet), a deep convolutional neural network designed for the detection and automatic classification of potentially malignant skin cancer conditions. SWNet optimizes feature extraction through multiple pathways, emphasizing width augmentation to enhance efficiency. The proposed model addresses potential biases associated with conditions, particularly in individuals darker tones or excessive hair, by incorporating fusion assimilate insights from diverse datasets. Extensive experiments were conducted using publicly accessible datasets evaluate SWNet’s effectiveness.This study utilized four datasets-Mnist-HAM10000, ISIC2019, ISIC2020, Melanoma Skin Cancer-comprising images categorized into benign classes. Explainable Artificial Intelligence (XAI) techniques, specifically Grad-CAM, employed interpretability model’s decisions. Comparative analysis was performed three pre-existing learning networks-EfficientNet, MobileNet, Darknet. results demonstrate superiority, achieving an accuracy 99.86% F1 score 99.95%, underscoring its efficacy gradient propagation capture across various levels. research highlights significant advancing classification, providing robust tool accurate early diagnosis. integration enhances mitigates hair tones. outcomes this contribute improved patient healthcare practices, showcasing exceptional capabilities classification.

Язык: Английский

---

A multi-stage melanoma recognition framework with deep residual neural network and hyperparameter optimization-based decision support in dermoscopy images

Expert Systems with Applications, Год журнала: 2022, Номер 215, С. 119352 - 119352

Опубликована: Ноя. 28, 2022

Язык: Английский

---

Federated and Transfer Learning Methods for the Classification of Melanoma and Nonmelanoma Skin Cancers: A Prospective Study

Sensors, Год журнала: 2023, Номер 23(20), С. 8457 - 8457

Опубликована: Окт. 13, 2023

Skin cancer is considered a dangerous type of with high global mortality rate. Manual skin diagnosis challenging and time-consuming method due to the complexity disease. Recently, deep learning transfer have been most effective methods for diagnosing this deadly cancer. To aid dermatologists other healthcare professionals in classifying images into melanoma nonmelanoma enabling treatment patients at an early stage, systematic literature review (SLR) presents various federated (FL) (TL) techniques that widely applied. This study explores FL TL classifiers by evaluating them terms performance metrics reported research studies, which include true positive rate (TPR), negative (TNR), area under curve (AUC), accuracy (ACC). was assembled systemized reviewing well-reputed studies published eminent fora between January 2018 July 2023. The existing compiled through search seven databases. A total 86 articles were included SLR. SLR contains recent on algorithms malignant In addition, taxonomy presented summarizes many non-malignant classes. results highlight limitations challenges research. Consequently, future direction work opportunities interested researchers are established help automated classification cancers.

Язык: Английский

---

Assist-Dermo: A Lightweight Separable Vision Transformer Model for Multiclass Skin Lesion Classification

Diagnostics, Год журнала: 2023, Номер 13(15), С. 2531 - 2531

Опубликована: Июль 29, 2023

A dermatologist-like automatic classification system is developed in this paper to recognize nine different classes of pigmented skin lesions (PSLs), using a separable vision transformer (SVT) technique assist clinical experts early cancer detection. In the past, researchers have few systems PSLs. However, they often require enormous computations achieve high performance, which burdensome deploy on resource-constrained devices. paper, new approach designing SVT architecture based SqueezeNet and depthwise CNN models. The primary goal find deep learning with parameters that has comparable accuracy state-of-the-art (SOTA) architectures. This modifies design for improved runtime performance by utilizing convolutions rather than simple conventional units. To develop Assist-Dermo system, data augmentation applied control PSL imbalance problem. Next, pre-processing step integrated select most dominant region then enhance lesion patterns perceptual-oriented color space. Afterwards, designed improve efficacy several layers multiple filter sizes but fewer filters parameters. For training evaluation models, set images collected from online sources such as Ph2, ISBI-2017, HAM10000, ISIC On chosen dataset, it achieves an (ACC) 95.6%, sensitivity (SE) 96.7%, specificity (SP) 95%, area under curve (AUC) 0.95. experimental results show suggested outperformed SOTA algorithms when recognizing performed better other competitive can support dermatologists diagnosis wide variety PSLs through dermoscopy. model code freely available GitHub scientific community.

Язык: Английский

---