The Crucial Role of Interdisciplinary Conferences in Advancing Explainable AI in Healthcare

1. Introduction

As new technologies and software aid in the push toward realization of personalized medicine, computational biology emerges as integral practice for revealing the complexity of biomedical interactions.[1,2] Large collections of patient data offer genomic, proteomic, metabolomic, and histopathological insight. When analyzed in concert, newly elucidated cellular interactions reveal novel treatment targets and disease mechanisms. In turn, a growing recognition of the need for transparency and interpretability in AI systems has spurred collective collaboration between clinicians, data scientists, and their associated stakeholders (i.e., software engineers, quality engineers, electrical engineers).[3]

As machine learning (ML) models became more complex and harder to interpret, the need for XAI, particularly when used in healthcare, has grown.[4,5] ML models are often seen as "black-boxes" due to the difficulty that arises in attempting to understand the pathway leading to a computational prediction. Though model-derived outputs may align with clinical guidelines, the algorithmic 'reasoning' behind ML predictions is often opaque and indecipherable.[6] To align with the evidence-based framework of healthcare, the decisions of computer aided diagnostic (CAD) tools must be seamlessly interrogable by the clinicians who utilize them in practice.[7,8]

XAI aims to overcome black-box opacity to reveal how ML predictions are made.[7,8,9] In the realm of healthcare, explainability and accountability are not only desirable but also legally required for AI systems that can have a significant impact on human lives.[8] The clinical impact of algorithmic bias for minority cohorts and others that are underrepresented in data used in algorithm training is also of growing concern.[8]

The majority of the state-of-the-art interpretable ML methods are domain-agnostic. Having evolved from fields such as computer vision, automated reasoning, or statistics, direct application to bioinformatics problems is challenging without customization and domain adaptation.[8] Recent years have seen rise in development of domain-specific XAI techniques tailored to healthcare and computational biology applications.[5,10] These techniques aim to address the unique challenges in these fields, such as the need for accurate and interpretable models, the integration of data from multiple sources, and the management and analysis of large datasets.[5] Despite the progress made in XAI research, there is still much work to be done in developing effective and transparent AI systems that are specifically tailored to healthcare and computational biology applications.

Challenges such as the development of domain-specific XAI techniques and the need for legal and ethical guidelines for the use of AI requires a multidisciplinary approach. As AI is increasingly utilized in patient care as co-pilot to physician-centric diagnostic, prognostic, and therapeutic applications, a host of medical, legal, ethical, and societal questions are invoked in consideration of the implications of XAI for the future of healthcare.[4,5,6,11] All stakeholders, including healthcare professionals, computational biologists, and policymakers, are increasingly compelled to understand each other's domain in order to innovate a solution.[12,13,14,15,16,17]

Specialized conferences play a crucial role in fostering interdisciplinary interactions, serving as platforms where experts from diverse fields can come together to share their research, exchange ideas, and form collaborative partnerships.[18,19,20] Through such collaborative ventures may transparency and interpretability of AI systems be advanced in healthcare and computational biology arenas.

Conferences are not limited to scientific presentations of research to the wider community, rather extend to offer an important venue for brainstorming, networking, and making vital connections that can lead to new initiatives, publications, and funding.[19] They facilitate exchanges across universities and institutions, allowing for the combination of resources and experiences that can further specific research efforts and science in general.[19]

However, the interdisciplinary nature of these conferences also presents certain challenges. For instance, individual expertise reflects what a specific field values, relevant problem spaces and worthwhile problems to pursue, the constitution of data within that field, and how that data is collected and disseminated. Therefore, although interdisciplinary collaborations are powerful because they unite disparate disciplines, the very nature of being interdisciplinary requires overcoming an array of barriers.[21]

These barriers can include difficulties in receiving funding for interdisciplinary research, the predominance of specialized journals and conferences that contribute to professional disciplinary siloing, and the need for researchers to step outside of their comfort zones and interact with researchers from other disciplines.[18,21] Despite these challenges, specialized conferences such as the ISCB GLBIO (Great Lakes Bioinformatics Conference, hosted by the International Society for Computational Biology) and the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) continue to play a vital role in fostering interdisciplinary collaboration and innovation in healthcare and computational biology.[18,19,20,21]

2. Current Challenges in Explainable AI and Computational Biology

While AI has the potential to transform healthcare, its application in healthcare is fraught with challenges. These challenges span various levels of AI adoption, including data collection, technological development, clinical application, ethical and societal concerns.[22,23,24,25,26]

AI models that lack transparency are often developed by non-medical professionals. A lack of control may result over the derivation of model results by end users, such as healthcare providers and patients.[24]

AI systems require large amounts of data, and the quality and availability of this data is crucial for the performance of these systems. However, datasets used to develop AI systems often include unforeseen gaps, despite intensive attempts to clean and analyze the data. Issues with regulation and compatibility across institutions also constrain the amount of data that can be utilized to develop efficient algorithms.[23]

The implementation of AI in clinical settings is also stymied by a lack of empirical data validating the effectiveness of AI-based interventions in planned clinical trials. Most research on AI's application has been conducted in non-clinical settings, making it challenging to generalize research results,[23] also raising questions about safety and efficacy.[26] Some argue that the opaque nature of many AI systems implies that physicians and patients cannot and should not rely on the results of such systems. In contrast, others oppose the central role of explainability in AI. However, sufficient explanations of AI models allow medical doctors to comprehend and trust AI-based clinical decision support systems.[27,28,29]

Logistical difficulties in implementing AI systems in healthcare include barriers to adoption and the need for sociocultural or pathway changes.[22] There are also challenges related to the internal capacity for strategic change management and the transformation of healthcare professions and healthcare practice.[25]

Ethical and societal concerns, including issues related to privacy, consent, and the patient right to choose their treatment and its delivery, pose additional challenges.

2.1. Legal Implications

There is a current paucity of well-defined regulations that specifically address issues which may arise due to the use of AI in healthcare settings.[30] This includes concerns about safety and effectiveness, liability, data protection and privacy, cybersecurity, and intellectual property law.[31] For instance, the sharing of responsibility and accountability when the implementation of an AI-based recommendation causes clinical problems is not clear.[32]

2.2. Ethical Implications

Ethical dilemmas in the application of AI in healthcare encompass a broad range of issues, including privacy and data protection, which are paramount due to the sensitive nature of personal health information. Informed consent is also critical, as patients must understand and agree to the use of AI in their healthcare procedures, recognizing the implications and outcomes associated with these technologies. Social gaps, characterized by disparities and inequalities within healthcare systems and broader society, may be intensified by the introduction of AI. For example, healthcare access disparities can result in unequal medical AI distribution across different socioeconomic populations and healthcare systems, potentially widening existing health inequities.

Furthermore, the incorporation of AI in healthcare raises questions about the preservation of medical consultation, which traditionally involves nuanced human interaction and shared decision-making between a patient and a healthcare provider. The personal touch of empathy and sympathy, which are foundational to patient care, may not be fully replicable by AI systems. While AI can enhance diagnostic and treatment processes, it is crucial to maintain the human elements of understanding and compassion that are integral to the healing process. It is also important to ensure that AI supports, rather than replaces, the human-centric aspects of medical consultation, allowing for the continuation of personalized care that addresses individual patient needs and concerns.[33] There are also concerns about intrinsic biases in the data used in AI system tests, which can lead to poor or negative outcomes.[34] The principles of medical ethics, including autonomy, beneficence, nonmaleficence, and justice, should be emphasized before integrating AI into healthcare systems.[11,33]

2.3. Societal Implications

AI can lead to healthcare inequities through biased data collection, algorithm development, and a lack of diversity in training data.[35] Such inequities may lead to automation bias, which can lead to discrimination and inequity at great scale.[36] Furthermore, the rapid and commercial development of AI could challenge known methods, protocols, standards and regulatory measures that govern the development, deployment, and management of technology in healthcare settings. This could necessitate new national and international regulations to ensure that AI is developed and used ethically, safely, and equitably in healthcare.[11] Ensuring ethical and legal implementation of AI, with consideration to societal implications, requires continuous attention and thoughtful policy.[37]

To ensure that AI-powered decision aids uphold the principles of patient-centered care, they must be developed with a focus on patient engagement and autonomy. Patient-centered care treats individuals as active participants in their health management, where their preferences and values guide clinical decisions. Trust in AI systems, then, hinges on their ability to operate transparently and provide explanations that patients can understand. Only with such clarity can patients confidently and independently choose to accept the recommendations provided by AI.[6]

2.4. Bridging the Gap with Computational Biology

The convergence of computational biology with clinical practice is essential for translating vast data science insights into medical advancements. With the emergence of high-throughput technologies which generate massive amounts of data requiring advanced data analysis techniques, the lines between bioinformatics and data science have become increasingly indistinct. Both fields now share common methodologies and tools to manage, analyze, and interpret large datasets (Table 1).

As the field of life sciences shifts towards a more data-centric, integrative, and computational approach, it's becoming imperative for biomedical researchers to develop proficiency in bioinformatics to keep pace with this evolution.[38]

Skill gaps in this proficiency now threaten the progress of modern research and fuel a global need for bioinformatics education and training. Bridging this gap is critical to the advancement of research and the pharmaceutical and biopharmaceutical industries.[38]

Policy makers and research funders should acknowledge the existing gap between the 'two cultures' of clinical informatics and data science. The full social and economic benefits of digital health and data science can only be realized by accepting the interdisciplinary nature of biomedical informatics and supporting a significant expansion of clinical informatics capacity and capability.[39]

3. The Importance of Interdisciplinary Collaboration

Interdisciplinary collaboration is crucial for the development and implementation of robust and effective solutions in healthcare AI. This approach brings together diverse expertise, including physicians, researchers, technologists, and policymakers, to refine AI algorithms, validate their clinical utility, and address the ethical and regulatory challenges associated with their implementation.[40]

3.1. Case Studies and Examples

3.1.1. Human-in the loop (HITL) Approach

• Sezgin, E. emphasizes the importance of a HITL approach in healthcare AI, where AI systems are guided, communicated, and supervised by human expertise. This approach ensures safety and quality in healthcare services. Therefore, there is a need for multidisciplinary teams to explore and evaluate cost-effective and impactful collaborative AI solutions and establish HITL protocols.[41]

3.1.2. Interdisciplinary Research in Digital Health

• Krause-Jüttler, G. et. al produced a case study on two interdisciplinary research projects involving 20 researchers from medicine and engineering sciences working jointly on digital health solutions. The study identified factors at the individual, team, and organizational levels that influence the implementation of interdisciplinary research projects.[42]

3.1.3. Intelligent Tutoring System for Medical Students

• Bilgic, E. and Harley, JM. offer an example of successful interdisciplinary collaboration: an intelligent tutoring system designed to help medical students with their diagnostic reasoning skills through virtual patient cases. This project brought together individuals from different disciplines, demonstrating the potential of interdisciplinary teams to produce and deliver AI-enhanced education effectively.[43]

3.1.4. Quality Management Systems (QMS) in Healthcare AI

• The integration of QMS principles into the life cycle of AI technologies within healthcare settings can close the AI translation gap by establishing a robust framework that accelerates the safe, ethical, and effective delivery of AI in patient care. Implementing a QMS requires adaptability, customization, and interdisciplinary collaboration, fostering awareness, education, and organizational change.[44,45]

Interdisciplinary collaboration is key to addressing the challenges and providing constructive solutions for the successful implementation of AI in healthcare. It ensures appropriate cooperation between computer scientists and healthcare providers, and fosters collaboration among multiple healthcare settings to share data, ensure its quality, and verify analyzed outcomes.[46]

Moreover, Interdisciplinary collaboration can help manage and optimize the utilization of equipment, beds, and staff by providing predictive insights, which is particularly valuable in healthcare settings where resources are often strained.[47]

Cross-disciplinary collaboration can be highly rewarding, yet challenging. We propose a series of best practices for successful cross-disciplinary collaboration based on research findings (Table 2)).

Successful cross-disciplinary collaboration requires open-mindedness, effective communication, clear roles and responsibilities, good leadership, and a commitment to reproducible science. By following these best practices, researchers can overcome the challenges associated with cross-disciplinary research and reap the benefits of this collaborative approach.[55,67,68]

3.2. The Role of Academia, Industry, and Healthcare Professionals

Academia, industry, and healthcare professionals all play essential roles in fostering interdisciplinary collaboration in healthcare AI.

3.2.1. Academia

Academic institutions are responsible for providing education and training in AI and related fields, as well as conducting research to advance the understanding and application of AI in healthcare.[69] They can facilitate interdisciplinary collaboration by offering joint programs, courses, and research projects that bring together students and faculty from diverse disciplines including medicine, computer science, engineering, and ethics.[70,71] By promoting interdisciplinary education and research, academia can help bridge the gap between data science and clinical applications, ensuring that AI technologies are developed and implemented effectively in healthcare settings.[71]

3.2.2. Industry

Industry partners, including technology companies and healthcare organizations, play a crucial role in the development, implementation, and evaluation of AI solutions in healthcare.[72] They can collaborate with academia and healthcare professionals to identify real-world problems, provide resources and expertise, and support the translation of research findings into practical applications.[72,73] Industry partners can also contribute to interdisciplinary collaboration by participating in conferences, workshops, and other events that bring together experts from different fields to share knowledge and ideas.[73]

3.2.3. Healthcare Professionals

Healthcare professionals, such as physicians, nurses, and allied health professionals (diverse group of health care experts distinct from nursing, medicine, and pharmacy, who provide a range of diagnostic, technical, therapeutic, and support services in patient care), are essential for the successful implementation of AI in healthcare. They can provide valuable contextual insights to identify relevant clinical problems, and contribute to the development and evaluation of AI solutions.[74] Healthcare professionals can also play a crucial role in promoting interdisciplinary collaboration by participating in research projects, sharing their expertise with academic and industry partners, and advocating for the integration of AI in healthcare practice.[74,75]

4. The Role of Conferences in Fostering Collaboration and Innovation

Conferences are vital for interdisciplinary knowledge exchange, fostering global connections, and forming professional networks.[76,77,78] Researchers present studies, gaining diverse perspectives and critiques.[77] Interactive sessions, incorporating technology and sustainability, offer practical insights (Table 3).[79]

Academic conferences, as platforms of situated learning, facilitate research sharing and agenda setting.[80] Networks formed impact access to resources.[78] They enable relationship building for researchers at different career stages, enhancing social trust.[78] Virtual conferences increase accessibility and inclusivity, extending reach and promoting diverse participation.[77] Speakers play a key role in audience engagement and innovation, adapting content for interactivity.[81] ISCB's GLBIO conference series has evolved to meet field demands, offering fellowship awards for diversity and managing a 'signature style' to handle organizational complexities.[79] The shift to virtual formats has broadened accessibility and representation, enriching discussions with diverse research and viewpoints[76].

These sessions provide opportunities for more in-depth discussions and hands-on experiences, fostering innovation and collaboration among participants.

In response to the global pandemic, GLBIO 2021 was held as a virtual event, demonstrating the conference's adaptability to changing circumstances.[82,83] This shift to a virtual format allowed for broader participation, especially from individuals who might not have been able to attend in person.

GLBIO has also incorporated novel approaches to encourage communication and foster collaborations. For example, a "matchmaking" session was introduced during GLBIO 2017, which was well-received and led to plans for similar sessions at future conferences.[84]

In addition, GLBIO has made efforts to promote diversity and inclusivity. Special consideration is given to first-time attendees, women, undergraduate or high school students, underrepresented minorities, and students with disabilities.[85]

5. Looking Forward: Future Directions and Innovations

Explainable AI (XAI) and computational biology are two rapidly evolving fields that are expected to bring significant advancements in the future.

5.1. Future Trends in Explainable AI

Explainable AI aims to make the decision-making process of AI models transparent and understandable. This is particularly important in healthcare, where the decisions made by AI can have significant impacts on patient outcomes.[6,11,27,86]

One of the key future trends in XAI is the development of models that can provide sufficient explanations to allow medical professionals to trust and comprehend AI-based clinical decision support systems.[27] This is crucial as the lack of explainability in AI systems has been identified as a potential threat to core ethical values in medicine.[6]

Moreover, the future of XAI in healthcare is expected to focus on addressing the interdisciplinary nature of explainability, which involves medical, legal, ethical, and societal considerations.[6] This will require fostering multidisciplinary collaboration and sensitizing developers, healthcare professionals, and legislators to the challenges and limitations of opaque algorithms in medical AI.[6]

5.2. Future Innovations in Computational Biology

The field is expected to continue to broaden its reach into new areas of research and development.[2] As we enter a period of unparalleled data accumulation and analysis, computational biology will undoubtedly continue to contribute to important advances in our understanding of molecular systems.[1] The success and wide acceptance of open data projects will impact how the general public sees computational biology as a field.[1]

5.3. Healthcare 5.0 and Explainable AI

Healthcare 5.0 is a vision for the future of healthcare that focuses on real-time patient monitoring, ambient control and wellness, and privacy compliance through assisted technologies.[87] Explainable AI (XAI) has emerged as a critical component in the evolution of healthcare, particularly in the context of Healthcare 5.0, where it plays a pivotal role in unlocking opportunities and addressing complex challenges.[88] XAI aims to produce a human-interpretable justification for each output, increasing confidence if the results appear plausible and match the clinicians' expectations.[87]

In the context of Healthcare 5.0, XAI can assist in finding suitable libraries that support visual explainability and interpretability of the output of AI models.[87] For instance, in medical imaging applications, end-to-end explainability can be provided through AI and federated transfer learning.[87]

5.4. Balancing Explainability and Accuracy/Performance in Future AI Models

The balance between explainability and accuracy/performance in future AI models is a critical consideration. More complex systems are capable of modeling intricate relationships in data, leading to higher accuracy, but their complexity often makes them less interpretable.[89] This trade-off between accuracy and explainability is a significant concern, especially for complex machine learning techniques like neural networks and deep learning.[90]

Research has shown that while professionals and the public value the explainability of AI systems, they may value it less in healthcare domains when weighed against system accuracy.[89] Van der Veer SN et al found that 88% of responding physicians preferred explainable over non-explainable AI, but without asking respondents to make the trade-off between explainability and accuracy.[89]

However, the absence of a plausible explanation does not imply an inaccurate model.[87] Therefore, instead of setting categorical rules around AI explainability, policymakers should consider the context and the specific needs of the application.[89]

The future of healthcare and research is poised to be significantly influenced by advancements in AI, computational biology, and their integration into various fields. These advancements are expected to revolutionize drug discovery, disease diagnosis, treatment recommendations, and patient engagement.[10,46,91,92,93,94]

One of the most promising areas of exploration is the application of AI and computational biology in drug discovery. AI can transform large amounts of aggregated data into usable knowledge, which can expedite the process of drug discovery and optimization.[10,91,94,95] For instance, the system known as Reinforcement Learning for Structural Evolution (ReLeaSE), implemented at the University of North Carolina, has shown potential in this area.[95] Furthermore, AI can handle the complex relationship between input and output variables for high-dimensional data, which is crucial in drug discovery.[10]

In addition to drug discovery, AI and computational biology are also being applied in the analysis of Whole Slide Images (WSIs) for tumor detection and segmentation, among other applications.[96,97]

Technological advancements in AI and data science are expected to continue at a rapid pace, with the AI-associated healthcare market projected to grow significantly.[98,99] These advancements are not only revolutionizing healthcare but also transforming the practice of medicine.[46,74,100,101]

Future research topics and areas of exploration are likely to focus on the ethical, legal, and societal challenges posed by the rapid advancements in AI. Addressing these challenges will require a multidisciplinary approach and the development of more rigorous AI techniques and models.[46,102,103]

Conferences and special issues of journals will play a crucial role in shaping these future directions. For instance, the Journal of Biomedical and Health Informatics is currently accepting submissions for special issues on topics such as "Ethical AI for Biomedical and Health Informatics in the Generative Era" and "Advancing Personalized Healthcare: Integrating AI and Health Informatics".[103] These platforms provide opportunities for interdisciplinary collaboration between AI experts, computer scientists, healthcare professionals, and informatics specialists, vital for the development of robust AI systems and ethical guidelines.[46,103]

6. Conclusions

AI's application in healthcare demands transparency and explainability to address legal, ethical, and societal concerns, especially in clinical decision-making and patient care.[71,72,73,74] The integration of AI and computational biology into healthcare and research holds immense potential for accelerating discoveries, improving diagnostics, and enhancing patient care. However, it also poses significant challenges that need to be addressed through continued research, innovation, and interdisciplinary collaboration.[46,102,103] Computational biology is pivotal in linking data science to clinical applications, necessitating interdisciplinary collaboration and training.[16] Conferences are key for collaboration and innovation, uniting experts across fields, enabling knowledge exchange, and stimulating idea sharing.[46,72,73,74] They are vital for academic progress, research refinement, and fostering real-world impacts. The role of academia, industry, and healthcare professionals in fostering interdisciplinary collaboration is crucial for effective AI implementation in healthcare. The future of explainable AI and computational biology is promising, with significant advancements expected in the development of transparent AI models and new computational methods for biological research. The future of XAI and computational biology in the context of Healthcare 5.0 will likely involve a careful balance between explainability and accuracy, with the ultimate goal of enhancing patient outcomes and upholding the essential elements of compassion, empathy, and ethical considerations that define the core of healthcare.[88] These advancements will likely have a profound impact on healthcare and biological research, leading to a deeper understanding of biological systems and improved patient outcomes.[71,72,73,74,75]