Mini-Symposia 1

Paul Sajda, Columbia University

Abstract: Simultaneous EEG-fMRI is a multi-modal neuroimaging technique that combines the advantages of both modalities, offering insights into the spatial and temporal dynamics of neural activity. In this presentation, we address the inference problem inherent in this technique by employing a  transcoding framework. Transcoding refers to mapping from a specific encoding (modality) to decoding (the latent source space) and subsequently encoding the latent source space back to the original modality. Our proposed method focuses on developing a symmetric approach involving a cyclic
convolutional transcoder capable of transcoding EEG to fMRI and vice versa. Importantly, our method does not rely on prior knowledge of the hemodynamic response function or lead field matrix. Instead, it leverages the temporal and spatial relationships between the modalities and latent source spaces to learn these mappings. By applying our method to real EEG-fMRI data, we demonstrate its efficacy in accurately transcoding the modalities from one to another and recovering the underlying source spaces. It is worth noting that these results are obtained on previously unseen data, further emphasizing the robustness and generalizability of our approach. Furthermore, apart from its ability to enable symmetric inference of a latent source space, our method can also be viewed as low-cost computational neuroimaging. Specifically, it allows for generating an ‘expensive fMRI BOLD image using low-cost EEG data. This aspect highlights our approach’s potential practical significance and affordability for research and clinical applications.

May Wang, Georgia Tech/Emory

Abstract: Precision medicine is to use big data capturing “individual differences in patients’ genes, environments, and lifestyles focuses” to make prevention strategies, screening, diagnosis of diseases and treatment therapies. In this panel talk, I will briefly introduce Natural Language Processing (NLP) for Precision Medicine. Not only NLP can decipher and sequentially report death events for individual patients, but also the pretrained transformer BERT can precisely discern long COVID-19 patients from the rest. Also, transformer architecture with Long Short-Term Memory and convolutional neural networks can extract spatial-temporal features for accurate seizure detection. Then I will briefly introduce AI Implementation Science for real world impact. The American Medical Informatics Association has led such effort. The goal is to find the best practice of AI lifecycle to accelerate safe and effective use of AI for healthcare.

Natalie Mrachacz-Kersting, Freiburg University

Abstract: Brain-Computer Interfaces (BCIs) have emerged as promising tools for neurorehabilitation, enabling direct communication between the brain and external devices. Efficient data handling is crucial for the success of BCIs in neurorehabilitation applications. This talk provides a concise overview of data handling techniques in BCIs specifically designed for neurorehabilitation purposes. It discusses various aspects such as data acquisition, preprocessing, feature extraction, and classification methods tailored to neurorehabilitation tasks. Additionally, I will explore the integration of machine learning algorithms and real-time feedback mechanisms to enhance the adaptability and effectiveness of BCIs in supporting motor recovery and cognitive rehabilitation in patients with neurological disorders. By addressing the challenges related to data processing and interpretation, this research contributes to the development of more sophisticated and user-friendly BCIs, fostering advancements in neurorehabilitation therapies.

Dimitris Fotiadis, Ioannina University

Healthcare systems generally constitute a combination of healthcare facilities and providers that collaborate in order to offer a complete range of healthcare services. As the primary providers of healthcare, hospitals hold an essential place in the healthcare system. This study introduces three novel automated Artificial Intelligence (AI) systems by using the robotic platform of the TIAGo robot: (i) an image-based system that monitors the food consumption of the patients and provides information regarding their energy intake to prevent malnutrition; (ii) a rehabilitation system that monitors patients as they perform prescribed exercises to prevent loss of mobility, and (iii) a monitoring system for the correctness of the oxygen therapy in terms of the proper positioning of the oxygen mask to prevent hypoxia complications. A well-designed observational clinical study is planned to evaluate and validate the utility and effectiveness of the proposed AI systems in improving patient stay in the hospital. For the abovementioned systems, three novel datasets were created.

Nan Liu, National University of Singapore

Abstract: In the ever-evolving landscape of healthcare, the integration of Artificial Intelligence (AI) has paved the way for revolutionary advancements. We will explore the remarkable transformation of medical care through the lens of aiTriage, an AI-based medical device. aiTriage harnesses AI to assess the risk of adverse outcomes by analyzing short ECG recording and basic patient information. This talk will provide an insightful narrative of the journey undertaken in developing diverse algorithms for managing two critical emergency scenarios: patients with chest pain and those grappling with sepsis. We will delve into the inception of aiTriage, its development process, and the intricacies of translating this technology into clinical practice. The discussion will encompass the challenges faced, the clinical benefits realized, and the potential future implications of this AI tool.