Search

Intracranial EEG (iEEG) recording, characterized by high spatial and temporal resolution and superior signal-to-noise ratio (SNR), enables the development of precise brain-computer interface (BCI) systems for neural decoding. However, the invasive nature of the procedure significantly limits the availability of iEEG datasets in terms of both the number of participants and the duration of recorded sessions. To address this limitation, we propose a single-participant machine learning model optimized for decoding iEEG signals. The model employs 18 key features and operates in two modes: best channel and combined channel. The combined channel mode integrates spatial information from multiple brain regions, leading to superior classification performance. Evaluations across three datasets -- Music Reconstruction, Audio Visual, and AJILE12 -- demonstrate that the combined channel mode consistently outperforms the best channel mode across all classifiers. In the best-performing cases, Random Forest achieved an F1 score of 0.81 +/- 0.05 in the Music Reconstruction dataset and 0.82 +/- 0.10 in the Audio Visual dataset, while XGBoost achieved an F1 score of 0.84 +/- 0.08 in the AJILE12 dataset. Furthermore, the analysis of brain region contributions in the combined channel mode revealed that the model identifies relevant brain regions aligned with physiological expectations for each task and effectively combines data from electrodes in these regions to achieve high performance. These findings highlight the potential of integrating spatial information across brain regions to improve task decoding, offering new avenues for advancing BCI systems and neurotechnological applications.

Utilization of intracranial electroencephalography (iEEG) is rapidly increasing for clinical and brain-computer interface applications. iEEG facilitates the recording of neural activity with high spatial and temporal resolution, making it a desirable neuroimaging modality for studying neural dynamics. Despite its benefits, iEEG faces challenges such as inter-subject variability in electrode implantation, which makes the development of unified neural decoder models across different patients difficult. In this research, we introduce a novel decoder model that is robust to inter-subject electrode implantation variability. We call this model RISE-iEEG, which stands for Robust Inter-Subject Electrode Implantation Variability iEEG Classifier. RISE-iEEG employs a deep neural network structure preceded by a patient-specific projection network. The projection network maps the neural data of individual patients onto a common low-dimensional space, compensating for the implantation variability. In other words, we developed an iEEG decoder model that can be applied across multiple patients' data without requiring the coordinates of electrode for each patient. The performance of RISE-iEEG across multiple datasets, including the Audio-Visual dataset, Music Reconstruction dataset, and Upper-Limb Movement dataset, surpasses that of state-of-the-art iEEG decoder models such as HTNet and EEGNet. Our analysis shows that the performance of RISE-iEEG is 10\% higher than that of HTNet and EEGNet in terms of F1 score, with an average F1 score of 83\%, which is the highest result among the evaluation methods defined. Furthermore, the analysis of projection network weights in the Music Reconstruction dataset across patients suggests that the Superior Temporal lobe serves as the primary encoding neural node. This finding aligns with the auditory processing physiology.

Evaluating large language models (LLM) in clinical scenarios is crucial to assessing their potential clinical utility. Existing benchmarks rely heavily on static question-answering, which does not accurately depict the complex, sequential nature of clinical decision-making. Here, we introduce AgentClinic, a multimodal agent benchmark for evaluating LLMs in simulated clinical environments that include patient interactions, multimodal data collection under incomplete information, and the usage of various tools, resulting in an in-depth evaluation across nine medical specialties and seven languages. We find that solving MedQA problems in the sequential decision-making format of AgentClinic is considerably more challenging, resulting in diagnostic accuracies that can drop to below a tenth of the original accuracy. Overall, we observe that agents sourced from Claude-3.5 outperform other LLM backbones in most settings. Nevertheless, we see stark differences in the LLMs' ability to make use of tools, such as experiential learning, adaptive retrieval, and reflection cycles. Strikingly, Llama-3 shows up to 92% relative improvements with the notebook tool that allows for writing and editing notes that persist across cases. To further scrutinize our clinical simulations, we leverage real-world electronic health records, perform a clinical reader study, perturb agents with biases, and explore novel patient-centric metrics that this interactive environment firstly enables.

Non-parametric models, such as Gaussian Processes (GP), show promising results in the analysis of complex data. Their applications in neuroscience data have recently gained traction. In this research, we introduce a novel neural decoder model built upon GP models. The core idea is that two GPs generate neural data and their associated labels using a set of low-dimensional latent variables. Under this modeling assumption, the latent variables represent the underlying manifold or essential features present in the neural data. When GPs are trained, the latent variable can be inferred from neural data to decode the labels with a high accuracy. We demonstrate an application of this decoder model in a verbal memory experiment dataset and show that the decoder accuracy in predicting stimulus significantly surpasses the state-of-the-art decoder models. The preceding performance of this model highlights the importance of utilizing non-parametric models in the analysis of neuroscience data.

There is increasing interest in the application large language models (LLMs) to the medical field, in part because of their impressive performance on medical exam questions. While promising, exam questions do not reflect the complexity of real patient-doctor interactions. In reality, physicians' decisions are shaped by many complex factors, such as patient compliance, personal experience, ethical beliefs, and cognitive bias. Taking a step toward understanding this, our hypothesis posits that when LLMs are confronted with clinical questions containing cognitive biases, they will yield significantly less accurate responses compared to the same questions presented without such biases. In this study, we developed BiasMedQA, a benchmark for evaluating cognitive biases in LLMs applied to medical tasks. Using BiasMedQA we evaluated six LLMs, namely GPT-4, Mixtral-8x70B, GPT-3.5, PaLM-2, Llama 2 70B-chat, and the medically specialized PMC Llama 13B. We tested these models on 1,273 questions from the US Medical Licensing Exam (USMLE) Steps 1, 2, and 3, modified to replicate common clinically-relevant cognitive biases. Our analysis revealed varying effects for biases on these LLMs, with GPT-4 standing out for its resilience to bias, in contrast to Llama 2 70B-chat and PMC Llama 13B, which were disproportionately affected by cognitive bias. Our findings highlight the critical need for bias mitigation in the development of medical LLMs, pointing towards safer and more reliable applications in healthcare.

Extracting meaningful information from high-dimensional data poses a formidable modeling challenge, particularly when the data is obscured by noise or represented through different modalities. This research proposes a novel non-parametric modeling approach, leveraging the Gaussian process (GP), to characterize high-dimensional data by mapping it to a latent low-dimensional manifold. This model, named the latent discriminative generative decoder (LDGD), employs both the data and associated labels in the manifold discovery process. We derive a Bayesian solution to infer the latent variables, allowing LDGD to effectively capture inherent stochasticity in the data. We demonstrate applications of LDGD on both synthetic and benchmark datasets. Not only does LDGD infer the manifold accurately, but its accuracy in predicting data points' labels surpasses state-of-the-art approaches. In the development of LDGD, we have incorporated inducing points to reduce the computational complexity of Gaussian processes for large datasets, enabling batch training for enhanced efficient processing and scalability. Additionally, we show that LDGD can robustly infer manifold and precisely predict labels for scenarios in that data size is limited, demonstrating its capability to efficiently characterize high-dimensional data with limited samples. These collective attributes highlight the importance of developing non-parametric modeling approaches to analyze high-dimensional data., Comment: 40 pages, 6 figures

Large language models (LLMs) are becoming increasingly relevant as a potential tool for healthcare, aiding communication between clinicians, researchers, and patients. However, traditional evaluations of LLMs on medical exam questions do not reflect the complexity of real patient-doctor interactions. An example of this complexity is the introduction of patient self-diagnosis, where a patient attempts to diagnose their own medical conditions from various sources. While the patient sometimes arrives at an accurate conclusion, they more often are led toward misdiagnosis due to the patient's over-emphasis on bias validating information. In this work we present a variety of LLMs with multiple-choice questions from United States medical board exams which are modified to include self-diagnostic reports from patients. Our findings highlight that when a patient proposes incorrect bias-validating information, the diagnostic accuracy of LLMs drop dramatically, revealing a high susceptibility to errors in self-diagnosis., Comment: 4 pages, Deep Generative Models for Health NeurIPS 2023

Understanding how external stimuli are encoded in distributed neural activity is of significant interest in clinical and basic neuroscience. To address this need, it is essential to develop analytical tools capable of handling limited data and the intrinsic stochasticity present in neural data. In this study, we propose a straightforward Bayesian time series classifier (BTsC) model that tackles these challenges whilst maintaining a high level of interpretability. We demonstrate the classification capabilities of this approach by utilizing neural data to decode colors in a visual task. The model exhibits consistent and reliable average performance of 75.55% on 4 patients' dataset, improving upon state-of-the-art machine learning techniques by about 3.0 percent. In addition to its high classification accuracy, the proposed BTsC model provides interpretable results, making the technique a valuable tool to study neural activity in various tasks and categories. The proposed solution can be applied to neural data recorded in various tasks, where there is a need for interpretable results and accurate classification accuracy.

Artificial neural networks (ANNs) have emerged as an essential tool in machine learning, achieving remarkable success across diverse domains, including image and speech generation, game playing, and robotics. However, there exist fundamental differences between ANNs' operating mechanisms and those of the biological brain, particularly concerning learning processes. This paper presents a comprehensive review of current brain-inspired learning representations in artificial neural networks. We investigate the integration of more biologically plausible mechanisms, such as synaptic plasticity, to enhance these networks' capabilities. Moreover, we delve into the potential advantages and challenges accompanying this approach. Ultimately, we pinpoint promising avenues for future research in this rapidly advancing field, which could bring us closer to understanding the essence of intelligence.

Nuclear detection, segmentation and morphometric profiling are essential in helping us further understand the relationship between histology and patient outcome. To drive innovation in this area, we setup a community-wide challenge using the largest available dataset of its kind to assess nuclear segmentation and cellular composition. Our challenge, named CoNIC, stimulated the development of reproducible algorithms for cellular recognition with real-time result inspection on public leaderboards. We conducted an extensive post-challenge analysis based on the top-performing models using 1,658 whole-slide images of colon tissue. With around 700 million detected nuclei per model, associated features were used for dysplasia grading and survival analysis, where we demonstrated that the challenge's improvement over the previous state-of-the-art led to significant boosts in downstream performance. Our findings also suggest that eosinophils and neutrophils play an important role in the tumour microevironment. We release challenge models and WSI-level results to foster the development of further methods for biomarker discovery.

Behavioral studies using personal digital devices typically produce rich longitudinal datasets of mixed data types. These data provide information about the behavior of users of these devices in real-time and in the users' natural environments. Analyzing the data requires multidisciplinary expertise and dedicated software. Currently, no generalizable, device-agnostic, freely available software exists within Python scientific computing ecosystem to preprocess and analyze such data. This paper introduces a Python package, Niimpy, for analyzing digital behavioral data. The Niimpy toolbox is a user-friendly open-source package that can quickly be expanded and adapted to specific research requirements. The toolbox facilitates the analysis phase by offering tools for preprocessing, extracting features, and exploring the data. It also aims to educate the user on behavioral data analysis and promotes open science practices. Over time, Niimpy will expand with extra data analysis features developed by the core group, new users, and developers. Niimpy can help the fast-growing number of researchers with diverse backgrounds who collect data from personal and consumer digital devices to systematically and efficiently analyze the data and extract useful information. This novel information is vital for answering research questions in various fields, from medicine to psychology, sociology, and others.

The aim of survival analysis in healthcare is to estimate the probability of occurrence of an event, such as a patient's death in an intensive care unit (ICU). Recent developments in deep neural networks (DNNs) for survival analysis show the superiority of these models in comparison with other well-known models in survival analysis applications. Ensuring the reliability and explainability of deep survival models deployed in healthcare is a necessity. Since DNN models often behave like a black box, their predictions might not be easily trusted by clinicians, especially when predictions are contrary to a physician's opinion. A deep survival model that explains and justifies its decision-making process could potentially gain the trust of clinicians. In this research, we propose the reverse survival model (RSM) framework that provides detailed insights into the decision-making process of survival models. For each patient of interest, RSM can extract similar patients from a dataset and rank them based on the most relevant features that deep survival models rely on for their predictions.