Building Robust Machine Learning Models for Water Quality Prediction DOI

Lourdu Mahimai Doss P,

M. Gunasekaran

Published: Nov. 24, 2023

This research focuses on developing machine learning models that can accurately predict water quality while being robust against adversarial attacks. The chosen approach involves using a logistic regression classifier and training the dataset containing various characteristics. study investigates models' vulnerability to poisoning evasion attacks data injection iterative FGSM techniques, respectively. To enhance their resilience, feature selection algorithm is proposed. identifies removes malicious or vulnerable features from data. effectiveness of proposed defense mechanisms evaluated through experiments, demonstrating ability achieve accurate prediction mitigating impact Our result analysis revealed that, initially reduced accuracy, models, fortified by mechanisms, consistently maintained an accuracy rate 62.80%. Overall, this contributes improving security reliability assessment systems.

Language: Английский

Duck swarm algorithm: theory, numerical optimization, and applications DOI
Mengjian Zhang, Guihua Wen

Cluster Computing, Journal Year: 2024, Volume and Issue: 27(5), P. 6441 - 6469

Published: March 1, 2024

Language: Английский

Citations

10
HSDG: A dual-prior semantic driven entropy grouping snapshot medical hyperspectral tongue image reconstruction method DOI
Huiyuan Zhang, Zhaohua Yang, Yijing Chen

et al.

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 105, P. 107689 - 107689

Published: Feb. 14, 2025

Language: Английский

Citations

1
Tongue shape classification based on IF-RCNet DOI Creative Commons
Tiantian Liang, Haowei Wang, Wei Yao

et al.

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

Published: March 1, 2025

Language: Английский

Citations

1
Multi-label body constitution recognition via HWmixer-MLP for facial and tongue images DOI
Mengjian Zhang, Guihua Wen, Pei Yang

et al.

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 126383 - 126383

Published: Jan. 1, 2025

Language: Английский

Citations

0
Interpretive machine learning predicts short-term mortality risk in elderly sepsis patients DOI Creative Commons
Xingyu Zhu, Zefei Jiang, Li Xiao

et al.

Frontiers in Physiology, Journal Year: 2025, Volume and Issue: 16

Published: March 26, 2025

Backgrounds Sepsis is a leading cause of in-hospital mortality. However, its prevalence increasing among the elderly population. Therefore, early identification and prediction risk death in patients with sepsis crucial. The objective this study was to create machine learning model that can predict short-term mortality severe clear concise manner. Methods Data collected from MIMIC-IV (2.2). It randomly divided into training set validation using 7:3 ratio. Mortality predictors were determined through Recursive Feature Elimination (RFE). A for 28 days ICU stay built six machine-learning algorithms. To comprehensive nuanced resolution, Shapley Additive Explanations (SHAP) Local Interpretable Model-Agnostic (LIME) used systematically interpret models at both global detailed level. Results involved analysis 4,056 sepsis. feature recursive elimination algorithm utilized select eight variables out 49 development. Six assessed, Extreme Gradient Boosting (XGBoost) found perform best. achieved an AUC 0.88 (95% CI: 0.86–0.90) accuracy 0.84 0.81–0.86) model. examine roles key model, SHAP employed. ranking order made evident, use LIME analysis, weights each range determined. Conclusion study’s dependable tool forecasting prognosis

Language: Английский

Citations

0
TSRNet: Tongue image segmentation with global and local refinement DOI Open Access
Wenjun Cai, Mengjian Zhang, Guihua Wen

et al.

Displays, Journal Year: 2023, Volume and Issue: 81, P. 102601 - 102601

Published: Nov. 29, 2023

Language: Английский

Citations

8
Evaluating deep learning techniques for identifying tongue features in subthreshold depression: a prospective observational study DOI Creative Commons
Bo Han,

Yue Chang,

Rui-rui Tan

et al.

Frontiers in Psychiatry, Journal Year: 2024, Volume and Issue: 15

Published: Aug. 8, 2024

This study aims to evaluate the potential of using tongue image features as non-invasive biomarkers for diagnosing subthreshold depression and assess correlation between these acupuncture treatment outcomes advanced deep learning models.

Language: Английский

Citations

2
Investigating the impact of novel XRayGAN in feature extraction for thoracic disease detection in chest radiographs: lung cancer DOI
Tehreem Awan, Khan Bahadar Khan

Signal Image and Video Processing, Journal Year: 2024, Volume and Issue: 18(5), P. 3957 - 3972

Published: May 8, 2024

Language: Английский

Citations

2
Duck swarm algorithm: a novel swarm intelligence algorithm DOI Creative Commons
Mengjian Zhang, Guihua Wen, Jing Yang

et al.

arXiv (Cornell University), Journal Year: 2021, Volume and Issue: unknown

Published: Jan. 1, 2021

A swarm intelligence-based optimization algorithm, named Duck Swarm Algorithm (DSA), is proposed in this study, which inspired by the searching for food sources and foraging behaviors of duck swarm. Two rules are modeled from finding duck, corresponds to exploration exploitation phases DSA, respectively. The performance DSA verified using multiple CEC benchmark functions, where its statistical (best, mean, standard deviation, average running-time) results compared with seven well-known algorithms like Particle (PSO), Firefly algorithm (FA), Chicken (CSO), Grey wolf optimizer (GWO), Sine cosine (SCA), Marine-predators (MPA), Archimedes (AOA). Moreover, Wilcoxon rank-sum test, Friedman convergence curves comparison utilized prove superiority against other algorithms. demonstrate that a high-performance method terms speed exploration-exploitation balance solving numerical problems. Also, applied optimal design six engineering constrained problems node deployment task Wireless Sensor Network (WSN). Overall, revealed promising very competitive different

Language: Английский

Citations

5
Multi-modal ultrasound multistage classification of PTC cervical lymph node metastasis via DualSwinThyroid DOI Creative Commons
Qiong Liu, Yue Li,

Yanhong Hao

et al.

Frontiers in Oncology, Journal Year: 2024, Volume and Issue: 14

Published: Feb. 15, 2024

Objective This study aims to predict cervical lymph node metastasis in papillary thyroid carcinoma (PTC) patients with high accuracy. To achieve this, we introduce a novel deep learning model, DualSwinThyroid, leveraging multi-modal ultrasound imaging data for prediction. Materials and methods We assembled substantial dataset consisting of 3652 images from 299 PTC this retrospective study. The newly developed DualSwinThyroid model integrates various modalities clinical data. Following its creation, rigorously assessed the model’s performance against separate testing set, comparing it established machine models previous approaches. Results Demonstrating remarkable precision, achieved an AUC 0.924 96.3% accuracy on test set. efficiently processed data, pinpointing features indicative nodule images. It offers three-tier classification that aligns each level specific surgical strategy treatment. Conclusion designed radiomics, effectively estimates degree patients. In addition, also provides early, precise identification facilitation interventions high-risk groups, thereby enhancing strategic selection approaches managing

Language: Английский

Citations

0