Trustworthy AI for Healthcare and Medicine

Dear Colleagues,

The high availability of various health big data such as electronic health/medical records, medical images or biomedical sensors time series has directly facilitated Artificial Intelligence (AI) research in diverse healthcare application scenarios from disease diagnosis, prognosis, monitoring and prevention to treatment evaluation. AI driven solutions for smarter healthcare have the potential to improve healthcare outcomes and patient care, as well as to reduce and handle the health, social and economic burden of an aging population, the ongoing global epidemic of non-communicable diseases or COVID-19 pandemic crises.

AI IA. Currently, when we refer to AI driven solutions, we usually mean systems developed by using machine/deep learning (ML/DL) methodologies, emerging from the efforts of  AI Intelligence Augmentation (AI IA) research activities. These solutions promote human-centered cooperation between human and machine, by having the aim to improve efficiency of the human decision-making process using automation process learned from the available data, which is conventionally complemented with human reasoning in order to manage the potential risks of fully automated machine decisions in problems of high societal importance. Consequently, exemplifying this human-centered cooperation between human and machine, AI driven solutions for smarter healthcare are implemented from the basis of supporting and augmenting the decision-making process of clinicians and medical professionals in disease diagnosis, clinical treatment and prediction. For example, medical image analysis or in general AI assisted radiology are widely held as the most recognized canonical examples of this human machine collaboration within healthcare applications assisted by medical imaging devices and AI.

AI II. Furthermore, convergence of AI technology with newly emerging pervasive wearable/wireless sensing technologies, distributed ledger technologies (such as blockchain), scalable distributed computing technologies (such as Cloud computing and post-Cloud paradigms, namely Fog, Edge, and Dew Computing) and Internet of Things (IoT), even further enhance and promote ubiquitous entanglement between humans and machines forming Human-Machine Networks (HMN), which present natural step towards ubiquitous intelligent infrastructures and collaborative human and machine healthcare environments for supporting a better quality of life related to the the clinical epidemiology and global health. Examples include  smart home health monitoring for elderly care, smart hospital or even whole smart city infrastructures.

However, the adaptation of AI models within healthcare systems cannot rely only on their good prediction performance and efficiency but needs to provide interpretable and robust explanations for their decisions by following principles of Veridical Data Science and Trustworthy AI. These principals aim to ensure that AI development, deployment and use are aligned with fundamental human principles and legal system, and that AI systems, like any other critical infrastructures, are human-centric, explainable, robust to different technical and social disruptions, safe, secure, privacy preserving and used in service of society and the common good. Therefore, the development of trustworthy artificial intelligence driven solutions for smarter  healthcare and (deeper) medicine requires involvement of the healthcare professionals for whom these systems are intended for use.

This Special Issue aims at collecting original and high-quality research articles, as well as surveys, reviews or short communications, presenting both applicational and theoretical research in Trustworthy aspects of AI for various healthcare applications and challenges with respect to the both AI perspectives:

(AI IA) Intelligence Augmentation for disease diagnosis, clinical treatment and prediction.

(AI II) Intelligent infrastructures for smarter healthcare and medicine

The main topics of this Special Issue include but are not limited to the following:

• Reliable AI driven smart sensors for managing intelligent healthcare environments
• Trustworthy AI in medical imaging and analysis
• Radiomics enriched with robust and explainable AI methodology for guiding personalized cancer diagnosis and treatment
• Data augmentation and harmonization methods for medical imaging data
• Interpretable and robust ML for biomedical images and biomedical sensors time series
• Trustworthy AI tools to assess and predict the risk of diseases and their progression
• Development of AI driven digital solutions for smart healthcare with sensor technologies
• Human-centric Explainable AI driven smart healthcare applications
• Doctor-, caregiver-, patient- centric multimodal AI systems for diagnosis, treatment, and/or prevention.
• Interlacing Human-Centric AI, smart sensors, actuators (robotics) and distributed computing technologies for remote delivery of healthcare and assisted living, and integration of mobile health (mHealth) into existing eHealth or telemedicine services
• Smart sensors  for digital phenotyping (or personal sensing) in healthcare applications
• Interpretable and robust Machine-learning in health applications
• Privacy preserving AI technologies for healthcare data
• Sensitive Data Detection methods in large structured and unstructured healthcare data
• Security and privacy in health applications driven by wearables
• Distributed collaborative training, learning, and inference over biomedical and healthcare data 
• Federated learning for smart healthcare applications and data for solving patient data privacy ethical and legal issues
• Trustworthy usage of real world health data within and (especially) beyond healthcare

Dr. Tomislav Lipic
Prof. Dr. Maja Prutki
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• human-centric and explainable artificial intelligence
• deep learning
• complex networks
• agent based modeling
• trustworthy AI
• veridical data science
• human-machine networks
• social good applications
• smart healthcare and medicine, biomedical imaging