Journal of Medical Internet Research

View query in Help articles search

Field

e.g. mhealth

Operator

Field

Criteria

e.g. mhealth

Search Results (1 to 9 of 9 Results)

Download search results: CSV END BibTex RIS

Sort by:

An Interpretable Model With Probabilistic Integrated Scoring for Mental Health Treatment Prediction: Design Study

After initial interest in the explainability of AI models in the 1980s and 1990s, recent advances in AI and ML, and the increased use of such technologies in safety-critical, socioeconomically and medically impactful applications have driven a renewed interest in XAI [20]. The explainability of AI models is generally understood as the ability for a wider range of users to understand the outputs of a given AI or ML model [21].

Anthony Kelly, Esben Kjems Jensen, Eoin Martino Grua, Kim Mathiasen, Pepijn Van de Ven

JMIR Med Inform 2025;13:e64617

Download Citation: END BibTex RIS

Download PDF Download XML

Enhancing Interpretable, Transparent, and Unobtrusive Detection of Acute Marijuana Intoxication in Natural Environments: Harnessing Smart Devices and Explainable AI to Empower Just-In-Time Adaptive Interventions: Longitudinal Observational Study

To enhance the interpretability of our algorithms and provide insights for just-in-time adaptive interventions, we incorporated explainable artificial intelligence (XAI) into our machine-learning pipeline. XAI helps clarify the role of digital biomarkers associated with self-reported marijuana intoxication in natural environments.

Sang Won Bae, Tammy Chung, Tongze Zhang, Anind K Dey, Rahul Islam

JMIR AI 2025;4:e52270

Download Citation: END BibTex RIS

Download PDF Download XML

Exploring the Applications of Explainability in Wearable Data Analytics: Systematic Literature Review

Figure 1 [13] depicts the relationship among AI, ML, DL, and XAI. Relationship among artificial intelligence (AI), machine learning (ML), deep learning (DL), and explainable AI (XAI) [13]. XAI enriches AI models with information comprehensible to the end users. While AI algorithms enable users to make informed business decisions, the opacity of these algorithms often leaves users uninformed about the decision-making processes [14]. This lack of transparency is where XAI comes into play.

Yasmin Abdelaal, Michaël Aupetit, Abdelkader Baggag, Dena Al-Thani

J Med Internet Res 2024;26:e53863

Download Citation: END BibTex RIS

Download PDF Download XML

The Depth Estimation and Visualization of Dermatological Lesions: Development and Usability Study

However, for fields like health care, where context plays a vital role, recent research has been explored to develop XAI. XAI methods help explain the decisions and predictions made by the model. This helps us improve our systems and fix our hyperparameters while implementing the models [4]. In the next section, we shall review some XAI methods and use them to detect skin melanoma. The second step is reconstructing the detected melanoma lesion as a 3 D holographic projection.

Pranav Parekh, Richard Oyeleke, Tejas Vishwanath

JMIR Dermatol 2024;7:e59839

Download Citation: END BibTex RIS

Download PDF Download XML

How Explainable Artificial Intelligence Can Increase or Decrease Clinicians’ Trust in AI Applications in Health Care: Systematic Review

A series of recent reviews have examined XAI from a trusted perspective. However, partly reflecting the speed of development of the field, these do not include the most recent empirical evidence from clinical settings, although they did consistently speculate that XAI could increase users’ trust and thus the intention to use AI tools [20,21], as well as enhance confidence in decisions and thus, the trust of clinicians [22,23].

Rikard Rosenbacke, Åsa Melhus, Martin McKee, David Stuckler

JMIR AI 2024;3:e53207

Download Citation: END BibTex RIS

Download PDF Download XML

Impact of Responsible AI on the Occurrence and Resolution of Ethical Issues: Protocol for a Scoping Review

xai

Selina Boege, Madison Milne-Ives, Edward Meinert

JMIR Res Protoc 2024;13:e52349

Download Citation: END BibTex RIS

Download PDF Download XML

Framework for Classifying Explainable Artificial Intelligence (XAI) Algorithms in Clinical Medicine

Furthermore, the clinical decision points supported by XAI as well as the manner in which explanations are provided to the user may differ greatly between algorithms and influence their efficacy. Here, we propose a framework for classifying XAI algorithms in clinical medicine in order to simplify this additional complexity and allow for performance evaluation of XAI in clinical practice. The ultimate scope of clinical medicine is to prolong and improve the quality of human life.

Thomas Gniadek, Jason Kang, Talent Theparee, Jacob Krive

Online J Public Health Inform 2023;15:e50934

Download Citation: END BibTex RIS

Download PDF Download XML

Leveraging Mobile Phone Sensors, Machine Learning, and Explainable Artificial Intelligence to Predict Imminent Same-Day Binge-drinking Events to Support Just-in-time Adaptive Interventions: Algorithm Development and Validation Study

Recent work has explored how explainable artificial intelligence (XAI) can provide greater transparency in how artificial intelligence generates model output. There are 2 core benefits of XAI methods. First, XAI provides transparency into factors contributing to the prediction of the outcome [25], which can increase trust in the model results [26]. Second, XAI enables hypothesis testing and algorithm adaptation through methods such as identifying feature importance and rule extraction [27].

Sang Won Bae, Brian Suffoletto, Tongze Zhang, Tammy Chung, Melik Ozolcer, Mohammad Rahul Islam, Anind K Dey