Understanding Inner States of Humans using Measurements of “Invisibles”
Organizers: Bjoern M. Eskofier, Martin Vossiek, Anne D. Koelewijn
Affiliations: FAU
Abstract:
Every movement of the body of a living being is the result of mechanisms of action taking place inside of the body and interactions between them. The body’s motor functions are initiated and regulated by neuronal processes and, more complex movements, controlled by sensitive body perception and cognition. Depending on the physical constitution, state of health or stress, and movements, parameters of the (interaction) mechanisms change. The symposium will focus on new ideas, sensors, and technologies such as radar, radio, laser, and camera sensors (called “Invisibles”) to improve the diagnosis process and the daily lives of patients and physicians. It will bring together technical experts in biomedical signal processing and physicians specialising in patient care. The symposium will provide a forum for academia, clinicians, industry, health insurance, and governing bodies to exchange ideas and promote collaboration.
The Opportunities and Challenges of ChatGPT in Brain-Machine Interface
Organizers: Ker-Jiun Wang, Ramana Vinjamuri, Maryam Alimardani, Zhi-Hong Mao, Midori Sugaya
Affiliations: University of Pittsburg, UMBC, Tilberg University, University of Pittsburg, Shibaura Institute of Technology
Abstract:
The emergence of brain-machine interface (BMI) technology has revolutionized the way we interact with machines and computers, enabling direct communication between the human brain and external devices. With recent advancements in artificial intelligence, ChatGPT, a language model developed by OpenAI, has become a powerful tool for natural language processing and generation. This workshop explores the challenges and opportunities associated with integrating ChatGPT into BMI systems. One of the main challenges lies in developing a seamless and efficient connection between ChatGPT and the BMI, allowing for real-time bidirectional communication. This requires overcoming the limitations of current BMI technology, such as low signal resolution, signal latency, and the need for invasive neural implants. Additionally, ChatGPT needs to be adapted to process and generate text in a way that aligns with the user’s intentions, while also respecting ethical and privacy considerations.
However, the integration of ChatGPT in BMI systems presents significant opportunities. Firstly, it can enhance the communication capabilities of BMIs by enabling users to generate and receive text-based information through their thoughts. This can greatly improve the speed and efficiency of human-machine interaction. Moreover, ChatGPT’s ability to understand and generate natural language can facilitate more intuitive and user-friendly interfaces for BMIs, enabling individuals with limited mobility to operate devices and communicate with others more effectively. Furthermore, ChatGPT can act as a mediator between users and external systems, providing a bridge for individuals with disabilities to access and control various technologies. It can assist in tasks such as composing emails, browsing the internet, or controlling smart home devices, opening up new possibilities for individuals with motor impairments. Additionally, the continuous interaction with users through ChatGPT can enable personalized and adaptive BMI systems, improving user satisfaction and system performance.
While integrating ChatGPT into BMI systems poses technical and ethical challenges, it also offers immense potential for advancing human-machine interaction. By addressing these challenges and leveraging the opportunities, the synergy between ChatGPT and BMIs can pave the way for more natural, efficient, and inclusive communication between humans and machines, empowering individuals with disabilities and enhancing overall quality of life.
Unraveling Challenges in Time Series Analysis with Open Source Tools for Digital Health Applications
Organizers: Md. Mobashir Hasan Sandhi, Hayoung Jeong, Ke Wang, Bill Chen, Bobak Jack Mortazavi, Jessilyn Dunn
Affiliations: Duke University, Duke University, Duke University, Duke University, Texas A&M University, Duke University
Abstract:
Time series analysis is becoming increasingly important in digital health and healthcare predictive analytics because it allows us to analyze data collected over time from medical devices and wearables. By analyzing time series data, clinicians can better understand the trajectory of a patient’s condition, monitor disease progression, and provide longitudinal care and management. However, challenges still exist in identifying proper preprocessing and feature engineering methods, incorporating multimodal signals, imputing missing values, and cleaning noisy data while retaining meaningful information about one’s physiological and behavioral changes. From processing the multimodal wearable sensor data to the development of machine learning models, we will introduce open-source initiatives and resources that can be readily used by researchers in digital health. In addition, we will emphasize the importance of open science and invite researchers to contribute their work to open source initiatives to enhance the reproducibility and accessibility of their research. In this workshop we will introduce digital health, digital biomarkers, and time-series physiological signals and data collected through various wearable sensors. We will discuss the current trends (in academia and industry) & challenges in preprocessing and performing data analytics/predictive modeling using multi-modal, time-series biomedical signal/data. We will further introduce a open source initiative, i.e., the Digital Biomarker Discovery Pipeline (DBDP) that host pre-built modular methods to process time-series data and demonstrate how currently available open source tools can be utilized by researchers, clinicians, and bioinformaticians. Finally, we will summarize the lessons learned in the workshop and share how the community can contribute to these efforts.
Computational Medicine Leveraging High Performance Computing and Artificial Intelligence
Organizers: Khan Iftekharuddin, Georgia Tourassi
Affiliations: Old Dominion University, ORNL
Abstract:
The convergence of personalized digital health technologies, high performance computing, and artificial intelligence have ushered a new era in health informatics and healthcare delivery. Computational medicine holds immense promise for personalized disease management, from diagnosis, to treatment, to prognosis. This session aims to highlight successful examples of multi-scale, multi-modal disease modeling and applications from the bench to the bedside, to fully realize the potential of computational modeling and personalized data in clinical practice. The workshop will include 5 invited speakers covering a broad spectrum of computational medicine applications highlighting the importance of high-performance computing and AI. The session will conclude with a panel discussion.
We will send announcements to relevant computational, machine learning, and biomedical mailing lists. As soon as our proposal is accepted, we will create a webpage and call for participation immediately. We will continue to send the call for participation to relevant mailing lists. We will also advertise the workshop in all appropriate presentations that we give, and we will ask our speakers to do the same.
Addressing real-word needs for digital health care: current and future challenges of the “human centered” design
Organizers: Federica Masci, Rosella Onofrio
Affiliations: KU Leuven, Politecnico di Milano
Abstract:
The Workshop the topic is “Addressing real-word needs for digital health care: current and future challenges of the “human centered” design “. This Workshop aims at discussing real-word needs for digital health care, focusing on current and future challenges of the “human centered” design. It is in fact well known that mobile and web technologies and digital solutions are tools for strengthening health systems, providing new opportunities for coordination, decentralization, and expansion of care quality and equity. Focusing on the design process has some appeal because it acknowledges the potential for tech-enabled innovation without limiting effects to specific technologies or technical solutions. In this context, we emphasized that human-centered design is not only about creating technologies or addressing purely technical issues, but also about accompanying people and understanding the complex challenge of strengthening health systems in the digital age. It is key to driving truly transformational change in any organization. The end goal of a Human-Centered design approach is to truly connect with your user as you are planning for your digital transformation. And if all humanity could benefit from the human-centered approach? Which are the real needs to be addressed? We will explore with our speakers current and future challenges of the “human centered” design in addressing the needs of the real-world stakeholders and point out possible strategies.
Transformative Pathways: Advancing Global Collaboration for Gender Equality and Inclusivity in Research and Innovation
Organizers: Maria Fernanda Cabrera-Umpierrez, Yolanda Ursa
Affiliations: Life Supporting Technologies, INMARK
Abstract:
The workshop aims to promote a collective comprehension of the fundamental principles and values that should drive international cooperation in science, technology, and innovation (STI), specifically focusing on fostering gender equality and inclusivity in research. Despite advancements in policy adoption at both international and national levels, the implementation of gender equality policies continues to pose challenges at the institutional level. Moreover, it is crucial to broaden the scope of gender equality policies to encompass inclusivity and intersectionality, considering other dimensions of diversity and potential grounds for discrimination, such as ethnicity, disability, or sexual orientation.
By exchanging best practices and embracing a multilateral perspective, participants will have the opportunity to share policies, initiatives, and activities, thereby cultivating a trusted environment for collaborative research and innovation with specific attention to gender equality and inclusion. The workshop aims to create a platform for open discussions, enabling the exchange of ideas and experiences among diverse stakeholders. Through interactive sessions, participants will explore effective strategies to address the existing gaps and barriers in achieving gender equality in R&I.
The workshop will be divided into two engaging parts, fostering collaboration and enabling participants to contribute their insights and ideas towards driving change in research and innovation.