Trustworthy Explainability Acceptance: A New Metric to Measure the Trustworthiness of Interpretable AI Medical Diagnostic Systems

Kaur, Davinder; Uslu, Süleyman; Durresi, Arjan; Badve, Sunil; Dündar, Murat

doi:10.1007/978-3-030-79725-6_4

Cited by 26 publications

(16 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A two-sided t-test of the TCAV scores should be used to evaluate if the null hypothesis is rejected [63]. Trustworthy Explainability Acceptance metric quantitatively measures the distance between explanations produced by the AI system and the explanations provided by medical experts [61]. Singla et al [129] utilized three metrics for the evaluation of the counterfactual explanations for chest X-ray classification : Frechet Inception Distance (FID), Counterfactual Validity (CV) and Foreign Object Preservation (FOP).…”

Section: Other Methodsmentioning

confidence: 99%

Transparency of Deep Neural Networks for Medical Image Analysis: A Review of Interpretability Methods

Salahuddin¹,

Woodruff²,

Chatterjee³

et al. 2021

Preprint

View full text Add to dashboard Cite

Artificial Intelligence (AI) has emerged as a useful aid in numerous clinical applications for diagnosis and treatment decisions. Deep neural networks have shown same or better performance than clinicians in many tasks owing to the rapid increase in the available data and computational power. In order to conform to the principles of trustworthy AI, it is essential that the AI system be transparent, robust, fair and ensure accountability. Current deep neural solutions are referred to as black-boxes due to a lack of understanding of the specifics concerning the decision making process. Therefore, there is a need to ensure interpretability of deep neural networks before they can be incorporated in the routine clinical workflow. In this narrative review, we utilized systematic keyword searches and domain expertise to identify nine different types of interpretability methods that have been used for understanding deep learning models for medical image analysis applications based on the type of generated explanations and technical similarities. Furthermore, we report the progress made towards evaluating the explanations produced by various interpretability methods. Finally we discuss limitations, provide guidelines for using interpretability methods and future directions concerning the interpretability of deep neural networks for medical imaging analysis.

show abstract

Section: Other Methodsmentioning

confidence: 99%

Transparency of Deep Neural Networks for Medical Image Analysis: A Review of Interpretability Methods

Salahuddin¹,

Woodruff²,

Chatterjee³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Numerous studies have explored trust modeling and its application in diverse scenarios. For instance, a trust framework proposed by Reference 11, grounded in measurement theory, finds applicability across various domains, including crime detection, 12 social networks, 13,14 the food-energy sector, [15][16][17][18][19][20] healthcare, 21,22 edge computing, [23][24][25][26][27] quantum computing, 28 and beyond. Recent work 21,[29][30][31] has highlighted trust's utility as an acceptance criterion for artificial intelligence algorithms.…”

Section: Role Of Trustmentioning

confidence: 99%

Artificial intelligence control for trust‐based detection of attackers in 5G social networks

Kaur,

Uslu,

Durresi

et al. 2023

Computational Intelligence

Self Cite

View full text Add to dashboard Cite

This study introduces a comprehensive framework designed for detecting and mitigating fake and potentially threatening user communities within 5G social networks. Leveraging geo‐location data, community trust dynamics, and AI‐driven community detection algorithms, this framework aims to pinpoint users posing potential harm. Including an artificial control model facilitates the selection of suitable community detection algorithms, coupled with a trust‐based strategy to effectively identify and filter potential attackers. A distinctive feature of this framework lies in its ability to consider attributes that prove challenging for malicious users to emulate, such as the established trust within the community, geographical location, and adaptability to diverse attack scenarios. To validate its efficacy, we illustrate the framework using synthetic social network data, demonstrating its ability to distinguish potential malicious users from trustworthy ones.

show abstract

“…In [24], a measurement theory-based trust management framework was proposed for online social communities. This framework has since been proven to facilitate decisionmaking in multiple areas such as online social networks [25], the food-energy-water nexus [44], crime detection [29], and cancer diagnosis [53]. It is a very flexible yet robust framework that can be adapted to different scenarios to capture trust.…”

Section: Trust Management Frameworkmentioning

confidence: 99%

“…To address the concerns about measuring the different aspects of trustworthiness, the metrics of acceptance [51] and fairness [52] were proposed to facilitate environmental decision making, an explainability metric [53] was proposed to interpret AI medical diagnosis systems, and a trustability metric [54] to assess trust in cloud computing. This paper presents an extended version of the trust management framework that includes the trustability metric, which helps to take action when an external attack or an internal event occurs in an autonomous device equipped with sensors or in a service running on the cloud.…”

Section: Introductionmentioning

confidence: 99%

Trustability for Resilient Internet of Things Services on 5G Multiple Access Edge Cloud Computing

Uslu

Kaur

Durrësi

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

Billions of Internet of Things (IoT) devices and sensors are expected to be supported by fifth-generation (5G) wireless cellular networks. This highly connected structure is predicted to attract different and unseen types of attacks on devices, sensors, and networks that require advanced mitigation strategies and the active monitoring of the system components. Therefore, a paradigm shift is needed, from traditional prevention and detection approaches toward resilience. This study proposes a trust-based defense framework to ensure resilient IoT services on 5G multi-access edge computing (MEC) systems. This defense framework is based on the trustability metric, which is an extension of the concept of reliability and measures how much a system can be trusted to keep a given level of performance under a specific successful attack vector. Furthermore, trustability is used as a trade-off with system cost to measure the net utility of the system. Systems using multiple sensors with different levels of redundancy were tested, and the framework was shown to measure the trustability of the entire system. Furthermore, different types of attacks were simulated on an edge cloud with multiple nodes, and the trustability was compared to the capabilities of dynamic node addition for the redundancy and removal of untrusted nodes. Finally, the defense framework measured the net utility of the service, comparing the two types of edge clouds with and without the node deactivation capability. Overall, the proposed defense framework based on trustability ensures a satisfactory level of resilience for IoT on 5G MEC systems, which serves as a trade-off with an accepted cost of redundant resources under various attacks.

show abstract

Trustworthy Explainability Acceptance: A New Metric to Measure the Trustworthiness of Interpretable AI Medical Diagnostic Systems

Cited by 26 publications

References 30 publications

Transparency of Deep Neural Networks for Medical Image Analysis: A Review of Interpretability Methods

Transparency of Deep Neural Networks for Medical Image Analysis: A Review of Interpretability Methods

Artificial intelligence control for trust‐based detection of attackers in 5G social networks

Trustability for Resilient Internet of Things Services on 5G Multiple Access Edge Cloud Computing

Contact Info

Product

Resources

About