An enhanced technique of skin cancer classification using deep convolutional neural network with transfer learning models

Machine Learning with Applications, Journal Year: 2021, Volume and Issue: 5, P. 100036 - 100036

Published: April 29, 2021

Skin cancer is one of the top three perilous types caused by damaged DNA that can cause death. This begins cells to grow uncontrollably and nowadays it getting increased speedily. There exist some researches for computerized analysis malignancy in skin lesion images. However, these images very challenging having troublesome factors like light reflections from surface, variations color illumination, different shapes, sizes lesions. As a result, evidential automatic recognition valuable build up accuracy proficiency pathologists early stages. In this paper, we propose deep convolutional neural network (DCNN) model based on learning approach accurate classification between benign malignant preprocessing firstly, apply filter or kernel remove noise artifacts; secondly, normalize input extract features help classification; finally, data augmentation increases number improves rate. To evaluate performance our proposed, DCNN compared with transfer models such as AlexNet, ResNet, VGG-16, DenseNet, MobileNet, etc. The evaluated HAM10000 dataset ultimately obtained highest 93.16% training 91.93% testing respectively. final outcomes proposed define more reliable robust when existing models.

Language: Английский

---

IoT data analytics in dynamic environments: From an automated machine learning perspective

Engineering Applications of Artificial Intelligence, Journal Year: 2022, Volume and Issue: 116, P. 105366 - 105366

Published: Sept. 16, 2022

Language: Английский

---

Data-driven early diagnosis of Chronic Kidney Disease: development and evaluation of an explainable AI model

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 38359 - 38369

Published: Jan. 1, 2023

Chronic Kidney Disease (CKD), where delayed recognition implies premature mortality, is currently experiencing a globally increasing incidence and high cost to health systems. Data mining allows discovering subtle patterns in CKD indicators contribute an early diagnosis. This work presents the development evaluation of explainable prediction model that would support clinicians diagnosis patients. The based on data management pipeline detects best combination ensemble trees algorithms features selected concerning classification performance. Furthermore, main contribution paper involves explainability-driven approach selecting predictive maintaining balance between accuracy explainability. Therefore, most balanced implements extreme gradient boosting classifier over 3 (packed cell value, specific gravity, hypertension), achieving 99.2% 97.5% with cross-validation technique new unseen respectively. In addition, analysis model's explainability shows packed value relevant feature influences results model, followed by gravity hypertension. small number reduced implying promising solution for developing countries.

Language: Английский

---

Detection of renal cell hydronephrosis in ultrasound kidney images: a study on the efficacy of deep convolutional neural networks

PeerJ Computer Science, Journal Year: 2024, Volume and Issue: 10, P. e1797 - e1797

Published: Jan. 23, 2024

In the realm of medical imaging, early detection kidney issues, particularly renal cell hydronephrosis, holds immense importance. Traditionally, identification such conditions within ultrasound images has relied on manual analysis, a labor-intensive and error-prone process. However, in recent years, emergence deep learning-based algorithms paved way for automation this domain. This study aims to harness power learning models autonomously detect hydronephrosis taken close proximity kidneys. State-of-the-art architectures, including VGG16, ResNet50, InceptionV3, innovative Novel DCNN, were put test subjected rigorous comparisons. The performance each model was meticulously evaluated, employing metrics as F1 score, accuracy, precision, recall. results paint compelling picture. DCNN outshines its peers, boasting an impressive accuracy rate 99.8%. same arena, InceptionV3 achieved notable 90% ResNet50 secured 89%, VGG16 reached 85%. These outcomes underscore DCNN's prowess images. Moreover, offers detailed view model's through confusion matrices, shedding light their abilities categorize true positives, negatives, false negatives. regard, exhibits remarkable proficiency, minimizing both positives conclusion, research underscores supremacy automating With exceptional minimal error rates, stands promising tool healthcare professionals, facilitating early-stage diagnosis treatment. Furthermore, convergence hold potential enhancement further exploration, testing larger more diverse datasets investigating optimization strategies.

Language: Английский

---

FSPBO-DQN: SeGAN based segmentation and Fractional Student Psychology Optimization enabled Deep Q Network for skin cancer detection in IoT applications

Artificial Intelligence in Medicine, Journal Year: 2022, Volume and Issue: 129, P. 102299 - 102299

Published: April 8, 2022

Language: Английский

---

Ensemble approach of transfer learning and vision transformer leveraging explainable AI for disease diagnosis: An advancement towards smart healthcare 5.0

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 179, P. 108874 - 108874

Published: July 15, 2024

Language: Английский

---

An Application of Data Envelopment Analysis and Machine Learning Approach to Risk Management

IEEE Access, Journal Year: 2021, Volume and Issue: 9, P. 85978 - 85994

Published: Jan. 1, 2021

An integrated method comprising DEA and machine learning for risk management is proposed in this paper. Initially, the process of assessment, cross-efficiency used to evaluate a set factors obtained from FMEA. This FMEA-DEA not only overcomes some drawbacks FMEA, but also eliminates several limitations offer high discrimination capability decision units. For treatment monitoring processes, an ML mechanism utilized predict degree remaining depending on simulated data corresponding scenario. Prediction using more accurate since predictive power model better than that which potentially contains errors. The motivation study combination approaches gives flexible realistic choice management. Based case logistics business, results ascertain short-term urgent solutions service cost performance are necessary sustainable operations under COVID-19 pandemic. prediction findings show skilled personnel next concern once strategies have been prioritised. approach allow decision-makers assess level handling forthcoming events unusual conditions. It serves as useful knowledge repository such appropriate mitigation can be planned monitored. outcome our empirical evaluation indicates contributes towards robustness business operations.

Language: Английский

---

A novel enhanced decision tree model for detecting chronic kidney disease

Network Modeling Analysis in Health Informatics and Bioinformatics, Journal Year: 2021, Volume and Issue: 10(1)

Published: April 11, 2021

Language: Английский

---

An Application of Machine Learning to Logistics Performance Prediction: An Economics Attribute-Based of Collective Instance

Computational Economics, Journal Year: 2023, Volume and Issue: 63(2), P. 741 - 792

Published: Feb. 1, 2023

Language: Английский

---

Integrating Homomorphic Encryption with Blockchain Technology for Machine Learning Applications

Journal of Machine and Computing, Journal Year: 2025, Volume and Issue: unknown, P. 395 - 408

Published: Jan. 3, 2025

Leveraging cutting-edge technology like blockchain and machine intelligence, smart healthcare systems have emerged as a potential strategy for enhancing services. In order to secure health data, this study offers unique design analysis of system that applies technique the paillier homomorphic encryption algorithm in addition learning detect cardiological disease. The suggested method seeks solve problems with predictive analytics safe data exchange medical field. Sensitive information is encrypted during transmission storage using Paillier Homomorphic Encryption technique, guaranteeing its confidentiality. By providing traceability accountability access sharing, used construct transparent record transactions. addition, forecast cardiac illness based on giving practitioners insightful help them make judgments. integration these technologies their advantages improving services are highlighted discussion proposed scheme's constructional operational specification section. Simulation experiments assess method’s efficiency reflect efficacy terms security, detection accurateness, computing proficiency. Comparing integrated approach conventional approaches, results demonstrate considerable improvement prediction accuracy security data. To sum up, provides thorough patient Machine learning, technology, all into it, which shows promise developing field systems.

Language: Английский

---