A systematic review of trustworthy and explainable artificial intelligence in healthcare: Assessment of quality, bias risk, and data fusion

Albahri, A. S.; Duhaim, Ali M.; Fadhel, Mohammed A.; Alnoor, Alhamzah; Baqer, Noor S.; Alzubaidi, Laith; Albahri, O. S.; Alamoodi, A. H.; Bai, Jinshuai; Salhi, Asma; Santamaría, José; Ouyang, Chun; Gupta, Ashish; Gu, YuanTong; Deveci, Muhammet

doi:10.1016/j.inffus.2023.03.008

Cited by 257 publications

(60 citation statements)

References 99 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These systems can greatly enhance our daily lives, but they also have the potential to cause harm to users or society, either directly or indirectly. Therefore, ensuring that these systems are safe, reliable, and trustworthy is essential [ 75 , 76 ]. Trustworthy deep learning technology in these applications can be achieved through several measures, including thorough testing and evaluation, transparent design and decision-making processes, and robust security and privacy measures.…”

Section: Current Challenges: Trustworthy Deep Learning In Ssvep-based...mentioning

confidence: 99%

A Systematic Review of Using Deep Learning Technology in the Steady-State Visually Evoked Potential-Based Brain-Computer Interface Applications: Current Trends and Future Trust Methodology

Albahri

Al-qaysi

Alzubaidi

et al. 2023

International Journal of Telemedicine and Applications

Self Cite

View full text Add to dashboard Cite

The significance of deep learning techniques in relation to steady-state visually evoked potential- (SSVEP-) based brain-computer interface (BCI) applications is assessed through a systematic review. Three reliable databases, PubMed, ScienceDirect, and IEEE, were considered to gather relevant scientific and theoretical articles. Initially, 125 papers were found between 2010 and 2021 related to this integrated research field. After the filtering process, only 30 articles were identified and classified into five categories based on their type of deep learning methods. The first category, convolutional neural network (CNN), accounts for 70% ( n = 21 / 30 ). The second category, recurrent neural network (RNN), accounts for 10% ( n = 3 / 30 ). The third and fourth categories, deep neural network (DNN) and long short-term memory (LSTM), account for 6% ( n = 30 ). The fifth category, restricted Boltzmann machine (RBM), accounts for 3% ( n = 1 / 30 ). The literature’s findings in terms of the main aspects identified in existing applications of deep learning pattern recognition techniques in SSVEP-based BCI, such as feature extraction, classification, activation functions, validation methods, and achieved classification accuracies, are examined. A comprehensive mapping analysis was also conducted, which identified six categories. Current challenges of ensuring trustworthy deep learning in SSVEP-based BCI applications were discussed, and recommendations were provided to researchers and developers. The study critically reviews the current unsolved issues of SSVEP-based BCI applications in terms of development challenges based on deep learning techniques and selection challenges based on multicriteria decision-making (MCDM). A trust proposal solution is presented with three methodology phases for evaluating and benchmarking SSVEP-based BCI applications using fuzzy decision-making techniques. Valuable insights and recommendations for researchers and developers in the SSVEP-based BCI and deep learning are provided.

show abstract

Section: Current Challenges: Trustworthy Deep Learning In Ssvep-based...mentioning

confidence: 99%

A Systematic Review of Using Deep Learning Technology in the Steady-State Visually Evoked Potential-Based Brain-Computer Interface Applications: Current Trends and Future Trust Methodology

Albahri

Al-qaysi

Alzubaidi

et al. 2023

International Journal of Telemedicine and Applications

Self Cite

View full text Add to dashboard Cite

show abstract

“…the realm of early-stage diabetes detection, the Decision Tree Classi er emerges as a promising tool for developing predictive models. Leveraging a collected dataset comprising crucial health parameters and clinical measurements [35], this algorithm enables the construction of a decision tree model. Through the utilization of various input features, the model learns the underlying patterns and relationships within the data, thereby facilitating precise predictions concerning the probability of an individual having early-stage diabetes.…”

Section: Decision Tree Classimentioning

confidence: 99%

Early Stage Diabetes Prediction by Approach Using Machine Learning Techniques

Zarar

Wang

2023

Preprint

View full text Add to dashboard Cite

Diabetes is the most viral and chronic disease throughout the world. A large number of people are affected by this chronic disease. Early detection of diabetes in a patient is crucial for ensuring a good quality of life. Machine learning techniques or Data Mining Techniques are playing a significant role in today’s life to detect diabetes and improve performance to make further accurate predictions. The aim of this research is diabetes prediction with the approach of machine learning techniques. In this technical approach, we have taken two data sets Pi-ma Indian diabetes data set and the Kaggle diabetes data set, and proposed a model for diabetes prediction. We have used four different machine learning algorithms such as Support Vector Machine, Decision Forest, Linear Regression, and Artificial Neural Network. In these machine learning algorithms, ANN gives the best prediction performance where the highest accuracy is 98.8% so, it could be used as an alternative method to support predict diabetes complication diseases at an initial stage. Further, this work can be extended to find how likely non-diabetic people can have diabetes in the next few years and also, this predicted model can be used for imaging processing in the future to find diabetes for the prediction of diabetic and non-diabetic.

show abstract

“…Unfortunately, a significant portion of the existing literature has failed to incorporate virtualisation techniques to elucidate the decision-making mechanisms employed by DL systems. Consequently, this lack of transparency can undermine trust in these systems' final decisions [24]. In light of this concern, the present study employs the Grad-CAM method to address these limitations and provide a more comprehensive understanding of how DL arrives at its decisions.…”

Section: Introductionmentioning

confidence: 99%

Novel Deep Feature Fusion Framework for Multi-Scenario Violence Detection

Jebur,

Hussein,

Hoomod

et al. 2023

Computers

Self Cite

View full text Add to dashboard Cite

Detecting violence in various scenarios is a difficult task that requires a high degree of generalisation. This includes fights in different environments such as schools, streets, and football stadiums. However, most current research on violence detection focuses on a single scenario, limiting its ability to generalise across multiple scenarios. To tackle this issue, this paper offers a new multi-scenario violence detection framework that operates in two environments: fighting in various locations and rugby stadiums. This framework has three main steps. Firstly, it uses transfer learning by employing three pre-trained models from the ImageNet dataset: Xception, Inception, and InceptionResNet. This approach enhances generalisation and prevents overfitting, as these models have already learned valuable features from a large and diverse dataset. Secondly, the framework combines features extracted from the three models through feature fusion, which improves feature representation and enhances performance. Lastly, the concatenation step combines the features of the first violence scenario with the second scenario to train a machine learning classifier, enabling the classifier to generalise across both scenarios. This concatenation framework is highly flexible, as it can incorporate multiple violence scenarios without requiring training from scratch with additional scenarios. The Fusion model, which incorporates feature fusion from multiple models, obtained an accuracy of 97.66% on the RLVS dataset and 92.89% on the Hockey dataset. The Concatenation model accomplished an accuracy of 97.64% on the RLVS and 92.41% on the Hockey datasets with just a single classifier. This is the first framework that allows for the classification of multiple violent scenarios within a single classifier. Furthermore, this framework is not limited to violence detection and can be adapted to different tasks.

show abstract

A systematic review of trustworthy and explainable artificial intelligence in healthcare: Assessment of quality, bias risk, and data fusion

Cited by 257 publications

References 99 publications

A Systematic Review of Using Deep Learning Technology in the Steady-State Visually Evoked Potential-Based Brain-Computer Interface Applications: Current Trends and Future Trust Methodology

A Systematic Review of Using Deep Learning Technology in the Steady-State Visually Evoked Potential-Based Brain-Computer Interface Applications: Current Trends and Future Trust Methodology

Early Stage Diabetes Prediction by Approach Using Machine Learning Techniques

Novel Deep Feature Fusion Framework for Multi-Scenario Violence Detection

Contact Info

Product

Resources

About