Neuroscience informatics最新文献_第4页

Decoding memory with explainable AI: A large-scale EEG-based machine learning study of encoding vs. retrieval 用可解释的人工智能解码记忆：基于脑电图的大规模机器学习研究编码与检索

Neuroscience informatics

Pub Date : 2025-08-14 DOI: 10.1016/j.neuri.2025.100227

Mohammed Tawshif Hossain , Adnan Sami Sarker , Arnab Chowdhury , Rajesh Mitra , Raiyan Rahman , M.R.C. Mahdy

Understanding the distinct neural signatures that differentiate memory encoding from retrieval remains a key challenge in cognitive neuroscience. This study applies machine learning to EEG data from the Penn Electrophysiology of Encoding and Retrieval Study (PEERS), involving 100 participants across over 400 sessions, to classify these cognitive states. We used Discrete Wavelet Transform (DWT) on EEG signals from six critical brain regions and evaluated seven machine learning models. Gradient Boosting emerged as the most effective classifier, achieving 81.97% accuracy and a 91.62% AUC. To interpret this performance, we applied Explainable AI (XAI) methods, specifically SHapley Additive exPlanations (SHAP). This analysis revealed that theta-band relative energy, especially in the Left and Right Anterior Superior (LAS/RAS) regions, was the most influential predictor. Low theta-band energy and RMS values were particularly indicative of encoding states. Topographic maps provided further validation, showing significant neural differences in anterior regions, notably within the theta range. However, the study is limited by the use of a fixed 2.5 s analysis window and demographic skew in the dataset, which may affect generalizability. Future work should address these issues through varied windowing strategies and more diverse populations. This study advances understanding of cognitive memory processes and supports the development of adaptive, memory-aware AI systems, contributing to both neuroscience and neurotechnology.

理解区分记忆编码和检索的不同神经特征仍然是认知神经科学的一个关键挑战。这项研究将机器学习应用于宾夕法尼亚大学编码和检索电生理学研究（PEERS）的脑电图数据，涉及100名参与者，跨越400多个会议，对这些认知状态进行分类。我们使用离散小波变换（DWT）对来自6个关键脑区的脑电图信号进行处理，并评估了7种机器学习模型。Gradient Boosting是最有效的分类器，准确率达到81.97%，AUC为91.62%。为了解释这种表现，我们应用了可解释人工智能（Explainable AI， XAI）方法，特别是SHapley加性解释（SHAP）。该分析显示，theta波段相对能量，特别是在左右前上（LAS/RAS）区域，是最具影响力的预测因子。低波段能量和RMS值特别表明编码状态。地形图提供了进一步的验证，显示了显著的神经差异在前部区域，特别是在θ波范围内。然而，该研究受到使用固定的2.5 s分析窗口和数据集中的人口统计偏差的限制，这可能会影响通用性。未来的工作应该通过不同的窗口策略和更多样化的人群来解决这些问题。这项研究促进了对认知记忆过程的理解，并支持了自适应、记忆感知的人工智能系统的发展，为神经科学和神经技术做出了贡献。

{"title":"Decoding memory with explainable AI: A large-scale EEG-based machine learning study of encoding vs. retrieval","authors":"Mohammed Tawshif Hossain , Adnan Sami Sarker , Arnab Chowdhury , Rajesh Mitra , Raiyan Rahman , M.R.C. Mahdy","doi":"10.1016/j.neuri.2025.100227","DOIUrl":"10.1016/j.neuri.2025.100227","url":null,"abstract":"<div><div>Understanding the distinct neural signatures that differentiate memory encoding from retrieval remains a key challenge in cognitive neuroscience. This study applies machine learning to EEG data from the Penn Electrophysiology of Encoding and Retrieval Study (PEERS), involving 100 participants across over 400 sessions, to classify these cognitive states. We used Discrete Wavelet Transform (DWT) on EEG signals from six critical brain regions and evaluated seven machine learning models. Gradient Boosting emerged as the most effective classifier, achieving 81.97% accuracy and a 91.62% AUC. To interpret this performance, we applied Explainable AI (XAI) methods, specifically SHapley Additive exPlanations (SHAP). This analysis revealed that theta-band relative energy, especially in the Left and Right Anterior Superior (LAS/RAS) regions, was the most influential predictor. Low theta-band energy and RMS values were particularly indicative of encoding states. Topographic maps provided further validation, showing significant neural differences in anterior regions, notably within the theta range. However, the study is limited by the use of a fixed 2.5 s analysis window and demographic skew in the dataset, which may affect generalizability. Future work should address these issues through varied windowing strategies and more diverse populations. This study advances understanding of cognitive memory processes and supports the development of adaptive, memory-aware AI systems, contributing to both neuroscience and neurotechnology.</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 4","pages":"Article 100227"},"PeriodicalIF":0.0,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144860746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Impact of physical activity and cardiorespiratory fitness on brain morphology among overweight and obese populations: A systematic review and meta-analysis of neuroimaging studies 身体活动和心肺健康对超重和肥胖人群脑形态的影响：神经影像学研究的系统回顾和荟萃分析

Neuroscience informatics

Pub Date : 2025-08-05 DOI: 10.1016/j.neuri.2025.100224

Dilip Shettigar , Suresh Sukumar , Rajagopal Kadavigere , K. Vaishali , Nitika C. Panakkal , Winniecia Dkhar , Abhimanyu Pradhan , Baskaran Chandrasekaran , Hari Prakash Palaniswamy , Poovitha Shruthi Paramashiva , Sneha Ravichandran , Sathya Sabina Muthu , Koustubh Kamath

引用次数: 0

AI-enhanced diagnosis of very late-onset schizophrenia-like psychosis: A step toward preventing dementia in older adults 人工智能对极晚发性精神分裂症样精神病的增强诊断：预防老年人痴呆的一步

Neuroscience informatics

Pub Date : 2025-07-28 DOI: 10.1016/j.neuri.2025.100223

Ali Allahgholi , Ava Mazhari

The rapid aging of the global population, projected to reach 2.1 billion individuals aged 60 and older by 2050, is associated with an increased prevalence of mental health conditions, particularly dementia and psychosis. Among these, very late-onset schizophrenia-like psychosis (VLOSLP), defined as occurring after age 60, poses significant diagnostic challenges due to overlapping neurobiological changes and medical conditions common in older adults. Studies have indicated a higher risk of dementia in patients with VLOSLP, emphasizing the necessity for ongoing symptom monitoring. In recent years, artificial intelligence (AI), particularly deep learning (DL) and machine learning (ML), has shown promise in enhancing disease diagnosis through advanced medical imaging techniques. This study aims to classify VLOSLP using MRI images from patients aged 60 and older, obtained from the COBRE and MCICshare databases via the SchizoConnect platform. To address the challenge of limited data, synthetic images were generated using Generative Adversarial Networks (GAN) following preprocessing techniques. These images were then classified using a Support Vector Machine (SVM) classifier, with feature extraction performed through Zernike moments. The findings achieved an area under the curve (AUC) of 0.98, contributing to more accurate diagnoses of VLOSLP and facilitating better management and monitoring of this complex condition in the aging population.

全球人口迅速老龄化，预计到2050年60岁及以上人口将达到21亿，这与精神健康状况，特别是痴呆症和精神病的患病率上升有关。其中，非常晚发性精神分裂症样精神病（VLOSLP），定义为发生在60岁以后，由于重叠的神经生物学变化和老年人常见的医疗条件，给诊断带来了重大挑战。研究表明，VLOSLP患者痴呆的风险更高，强调了持续监测症状的必要性。近年来，人工智能（AI），特别是深度学习（DL）和机器学习（ML），在通过先进的医学成像技术增强疾病诊断方面显示出了希望。本研究旨在通过SchizoConnect平台从COBRE和MCICshare数据库获得60岁及以上患者的MRI图像，对VLOSLP进行分类。为了解决有限数据的挑战，在预处理技术之后使用生成对抗网络（GAN）生成合成图像。然后使用支持向量机（SVM）分类器对这些图像进行分类，并通过泽尼克矩进行特征提取。研究结果达到了0.98的曲线下面积（AUC），有助于更准确地诊断VLOSLP，并有助于更好地管理和监测老年人群中这一复杂疾病。

{"title":"AI-enhanced diagnosis of very late-onset schizophrenia-like psychosis: A step toward preventing dementia in older adults","authors":"Ali Allahgholi , Ava Mazhari","doi":"10.1016/j.neuri.2025.100223","DOIUrl":"10.1016/j.neuri.2025.100223","url":null,"abstract":"<div><div>The rapid aging of the global population, projected to reach 2.1 billion individuals aged 60 and older by 2050, is associated with an increased prevalence of mental health conditions, particularly dementia and psychosis. Among these, very late-onset schizophrenia-like psychosis (VLOSLP), defined as occurring after age 60, poses significant diagnostic challenges due to overlapping neurobiological changes and medical conditions common in older adults. Studies have indicated a higher risk of dementia in patients with VLOSLP, emphasizing the necessity for ongoing symptom monitoring. In recent years, artificial intelligence (AI), particularly deep learning (DL) and machine learning (ML), has shown promise in enhancing disease diagnosis through advanced medical imaging techniques. This study aims to classify VLOSLP using MRI images from patients aged 60 and older, obtained from the COBRE and MCICshare databases via the SchizoConnect platform. To address the challenge of limited data, synthetic images were generated using Generative Adversarial Networks (GAN) following preprocessing techniques. These images were then classified using a Support Vector Machine (SVM) classifier, with feature extraction performed through Zernike moments. The findings achieved an area under the curve (AUC) of 0.98, contributing to more accurate diagnoses of VLOSLP and facilitating better management and monitoring of this complex condition in the aging population.</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 3","pages":"Article 100223"},"PeriodicalIF":0.0,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144722969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep learning model for patient emotion recognition using EEG-tNIRS data 基于EEG-tNIRS数据的患者情绪识别深度学习模型

Neuroscience informatics

Pub Date : 2025-07-22 DOI: 10.1016/j.neuri.2025.100219

Mohan Raparthi , Nischay Reddy Mitta , Vinay Kumar Dunka , Sowmya Gudekota , Sandeep Pushyamitra Pattyam , Venkata Siva Prakash Nimmagadda

This study presents a novel approach that integrates electroencephalogram (EEG) and functional near-infrared spectroscopy (tNIRS) data to enhance emotion classification accuracy. A Modality-Attentive Multi-Channel Graph Convolution Model (MAMP-GF) is introduced, leveraging GraphSAGE-based representation learning to capture inter-channel relationships. Multi-level feature extraction techniques, including Channel Features (CF), Statistical Features (SF), and Graph Features (GF), are employed to maximize the discriminative power of EEG-tNIRS signals. To enhance modality fusion, we propose and evaluate three fusion strategies: MA-GF, MP-GF, and MA-MP-GF, which integrate graph convolutional networks with a modality attention mechanism. The model is trained and validated using EEG and tNIRS data collected from 30 subjects exposed to emotionally stimulating video clips. Experimental results demonstrate that the proposed MA-MP-GF fusion model achieves 98.77% accuracy in subject-dependent experiments, significantly outperforming traditional single-modal and other multimodal fusion methods. In cross-subject validation, the model attains a 55.53% accuracy, highlighting its robustness despite inter-subject variability. The findings illustrate that the proposed graph convolution fusion approach, combined with modality attention, effectively enhances emotion recognition accuracy and stability. This research underscores the potential of EEG-tNIRS fusion in real-time, non-invasive emotion monitoring, paving the way for advanced applications in personalized healthcare and affective computing.

本研究提出了一种结合脑电图（EEG）和功能近红外光谱（tNIRS）数据的新方法，以提高情绪分类的准确性。引入了一种模态关注的多通道图卷积模型（MAMP-GF），利用基于graphsage的表示学习来捕获通道间关系。采用通道特征（CF）、统计特征（SF）和图特征（GF）等多层次特征提取技术，最大限度地提高了EEG-tNIRS信号的判别能力。为了增强模态融合，我们提出并评估了三种融合策略：MA-GF、MP-GF和MA-MP-GF，它们将图卷积网络与模态注意机制相结合。该模型是通过从30名观看情绪刺激视频片段的受试者中收集的EEG和tnir数据进行训练和验证的。实验结果表明，MA-MP-GF融合模型在主体相关实验中准确率达到98.77%，显著优于传统的单模态和其他多模态融合方法。在跨主题验证中，该模型达到55.53%的准确率，突出了其鲁棒性，尽管存在不同主题的差异。研究结果表明，本文提出的图卷积融合方法与模态关注相结合，有效地提高了情感识别的准确性和稳定性。这项研究强调了EEG-tNIRS融合在实时、非侵入性情绪监测中的潜力，为个性化医疗保健和情感计算的高级应用铺平了道路。

{"title":"Deep learning model for patient emotion recognition using EEG-tNIRS data","authors":"Mohan Raparthi , Nischay Reddy Mitta , Vinay Kumar Dunka , Sowmya Gudekota , Sandeep Pushyamitra Pattyam , Venkata Siva Prakash Nimmagadda","doi":"10.1016/j.neuri.2025.100219","DOIUrl":"10.1016/j.neuri.2025.100219","url":null,"abstract":"<div><div>This study presents a novel approach that integrates electroencephalogram (EEG) and functional near-infrared spectroscopy (tNIRS) data to enhance emotion classification accuracy. A Modality-Attentive Multi-Channel Graph Convolution Model (MAMP-GF) is introduced, leveraging GraphSAGE-based representation learning to capture inter-channel relationships. Multi-level feature extraction techniques, including Channel Features (CF), Statistical Features (SF), and Graph Features (GF), are employed to maximize the discriminative power of EEG-tNIRS signals. To enhance modality fusion, we propose and evaluate three fusion strategies: MA-GF, MP-GF, and MA-MP-GF, which integrate graph convolutional networks with a modality attention mechanism. The model is trained and validated using EEG and tNIRS data collected from 30 subjects exposed to emotionally stimulating video clips. Experimental results demonstrate that the proposed MA-MP-GF fusion model achieves 98.77% accuracy in subject-dependent experiments, significantly outperforming traditional single-modal and other multimodal fusion methods. In cross-subject validation, the model attains a 55.53% accuracy, highlighting its robustness despite inter-subject variability. The findings illustrate that the proposed graph convolution fusion approach, combined with modality attention, effectively enhances emotion recognition accuracy and stability. This research underscores the potential of EEG-tNIRS fusion in real-time, non-invasive emotion monitoring, paving the way for advanced applications in personalized healthcare and affective computing.</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 3","pages":"Article 100219"},"PeriodicalIF":0.0,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144696612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A systematic review and meta-analysis on the diagnostic accuracy of artificial intelligence and computer-aided diagnosis of lumbar prolapsed intervertebral disc 人工智能和计算机辅助诊断腰椎间盘突出症诊断准确性的系统回顾和荟萃分析

Neuroscience informatics

Pub Date : 2025-07-22 DOI: 10.1016/j.neuri.2025.100221

Sandeep Pattnaik , Manu Goyal , Rajneesh Kumar Gujral , Amit Mittal

Introduction

Lumbar prolapsed intervertebral disc (PIVD) is a debilitating lower back condition, whose accurate and timely diagnosis is crucial for its effective management. Artificial intelligence (AI) and computer-aided diagnosis (CAD) techniques have the potential to revolutionise diagnosis by improving accuracy, efficiency, and objectivity. This systematic review with meta-analysis thus aims to thoroughly assess the available knowledge on the usability of different AI and CAD-based tools in lumbar PIVD diagnosis.

Methods

A systematic search of electronic databases, between June and August 2024 for relevant full-text studies. The primary outcomes for review included the diagnostic accuracy (of each AI and CAD system. Subsequently, a meta-analysis was conducted to synthesise the results of the included studies.

Result

A total of eight studies were identified, evaluating thirteen CAD or AI systems. The meta-analysis involved three of the studies, and it demonstrated a high pooled sensitivity (0.901, 95% CI: 0.871–0.924) and specificity (0.919, 95% CI: 0.898–0.936) for lumbar PIVD diagnosis.

Conclusion

To conclude, these findings strongly support the potential of AI/CAD systems to improve the accuracy and efficiency of lumbar PIVD diagnosis.

Prospero ID

CRD42023444785

腰椎间盘突出症（PIVD）是一种使腰背部衰弱的疾病，准确及时的诊断对其有效治疗至关重要。人工智能（AI）和计算机辅助诊断（CAD）技术有可能通过提高准确性、效率和客观性来彻底改变诊断。因此，本系统综述和荟萃分析旨在全面评估不同人工智能和基于cad的工具在腰椎PIVD诊断中的可用性。方法系统检索电子数据库，于2024年6 - 8月间进行相关全文研究。评估的主要结果包括每个AI和CAD系统的诊断准确性。随后，进行荟萃分析以综合纳入研究的结果。结果共确定了8项研究，评估了13个CAD或AI系统。荟萃分析涉及三项研究，结果显示腰椎PIVD诊断具有较高的综合敏感性（0.901,95% CI: 0.871-0.924）和特异性（0.919,95% CI: 0.898-0.936）。综上所述，这些发现有力地支持了AI/CAD系统在提高腰椎PIVD诊断的准确性和效率方面的潜力。普洛斯彼罗IDCRD42023444785

{"title":"A systematic review and meta-analysis on the diagnostic accuracy of artificial intelligence and computer-aided diagnosis of lumbar prolapsed intervertebral disc","authors":"Sandeep Pattnaik , Manu Goyal , Rajneesh Kumar Gujral , Amit Mittal","doi":"10.1016/j.neuri.2025.100221","DOIUrl":"10.1016/j.neuri.2025.100221","url":null,"abstract":"<div><h3>Introduction</h3><div>Lumbar prolapsed intervertebral disc (PIVD) is a debilitating lower back condition, whose accurate and timely diagnosis is crucial for its effective management. Artificial intelligence (AI) and computer-aided diagnosis (CAD) techniques have the potential to revolutionise diagnosis by improving accuracy, efficiency, and objectivity. This systematic review with meta-analysis thus aims to thoroughly assess the available knowledge on the usability of different AI and CAD-based tools in lumbar PIVD diagnosis.</div></div><div><h3>Methods</h3><div>A systematic search of electronic databases, between June and August 2024 for relevant full-text studies. The primary outcomes for review included the diagnostic accuracy (of each AI and CAD system. Subsequently, a meta-analysis was conducted to synthesise the results of the included studies.</div></div><div><h3>Result</h3><div>A total of eight studies were identified, evaluating thirteen CAD or AI systems. The meta-analysis involved three of the studies, and it demonstrated a high pooled sensitivity (0.901, 95% CI: 0.871–0.924) and specificity (0.919, 95% CI: 0.898–0.936) for lumbar PIVD diagnosis.</div></div><div><h3>Conclusion</h3><div>To conclude, these findings strongly support the potential of AI/CAD systems to improve the accuracy and efficiency of lumbar PIVD diagnosis.</div></div><div><h3>Prospero ID</h3><div>CRD42023444785</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 3","pages":"Article 100221"},"PeriodicalIF":0.0,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144722968","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Reinforcement learning in artificial intelligence and neurobiology 人工智能和神经生物学中的强化学习

Neuroscience informatics

Pub Date : 2025-07-22 DOI: 10.1016/j.neuri.2025.100220

Tursun Alkam, Andrew H Van Benschoten, Ebrahim Tarshizi

Reinforcement learning (RL), a computational framework rooted in behavioral psychology, enables agents to learn optimal actions through trial and error. It now powers intelligent systems across domains such as autonomous driving, robotics, and logistics, solving tasks once thought to require human cognition. As RL reshapes artificial intelligence (AI), it raises a critical question in neuroscience: does the brain learn through similar mechanisms? Growing evidence suggests it does.

To bridge this interdisciplinary gap, this review introduces core RL concepts to neuroscientists and clinicians with limited AI exposure. We outline the agent–environment interaction loop and describe key architectures including model-free, model-based, and meta-RL. We then examine how advances in deep RL have generated testable hypotheses about neural computation and behavior. In parallel, we discuss how neurobiological findings, especially the role of dopamine in encoding reward prediction errors, have inspired biologically grounded RL models. Empirical studies reveal neural correlates of RL algorithms in the basal ganglia, prefrontal cortex, and hippocampus, supporting their roles in planning, memory, and decision-making. We also highlight clinical applications, including how RL frameworks are used to model cognitive decline and psychiatric disorders, while acknowledging limitations in scaling RL to biological complexity.

Looking ahead, RL offers powerful tools for understanding brain function, guiding brain–machine interfaces, and personalizing psychiatric treatment. The convergence of RL and neuroscience offers a promising interdisciplinary lens for advancing our understanding of learning and decision-making in both artificial agents and the human brain.

强化学习（RL）是一种植根于行为心理学的计算框架，它使代理能够通过试错来学习最佳行为。它现在为自动驾驶、机器人和物流等领域的智能系统提供动力，解决了曾经被认为需要人类认知的任务。随着强化学习重塑人工智能（AI），它提出了神经科学中的一个关键问题：大脑是否通过类似的机制进行学习？越来越多的证据表明确实如此。为了弥合这一跨学科的差距，本综述向人工智能接触有限的神经科学家和临床医生介绍了核心RL概念。我们概述了代理-环境交互循环，并描述了包括无模型、基于模型和元强化学习在内的关键体系结构。然后，我们研究了深度强化学习的进展如何产生关于神经计算和行为的可测试假设。同时，我们讨论了神经生物学的发现，特别是多巴胺在编码奖励预测错误中的作用，如何启发了基于生物学的RL模型。实证研究揭示了RL算法在基底神经节、前额叶皮层和海马体中的神经关联，支持它们在计划、记忆和决策中的作用。我们还强调了临床应用，包括RL框架如何用于模拟认知能力下降和精神疾病，同时承认将RL扩展到生物复杂性的局限性。展望未来，强化学习为理解大脑功能、指导脑机接口和个性化精神治疗提供了强大的工具。强化学习和神经科学的融合提供了一个很有前途的跨学科视角，可以促进我们对人工智能体和人脑中学习和决策的理解。

{"title":"Reinforcement learning in artificial intelligence and neurobiology","authors":"Tursun Alkam, Andrew H Van Benschoten, Ebrahim Tarshizi","doi":"10.1016/j.neuri.2025.100220","DOIUrl":"10.1016/j.neuri.2025.100220","url":null,"abstract":"<div><div>Reinforcement learning (RL), a computational framework rooted in behavioral psychology, enables agents to learn optimal actions through trial and error. It now powers intelligent systems across domains such as autonomous driving, robotics, and logistics, solving tasks once thought to require human cognition. As RL reshapes artificial intelligence (AI), it raises a critical question in neuroscience: does the brain learn through similar mechanisms? Growing evidence suggests it does.</div><div>To bridge this interdisciplinary gap, this review introduces core RL concepts to neuroscientists and clinicians with limited AI exposure. We outline the agent–environment interaction loop and describe key architectures including model-free, model-based, and meta-RL. We then examine how advances in deep RL have generated testable hypotheses about neural computation and behavior. In parallel, we discuss how neurobiological findings, especially the role of dopamine in encoding reward prediction errors, have inspired biologically grounded RL models. Empirical studies reveal neural correlates of RL algorithms in the basal ganglia, prefrontal cortex, and hippocampus, supporting their roles in planning, memory, and decision-making. We also highlight clinical applications, including how RL frameworks are used to model cognitive decline and psychiatric disorders, while acknowledging limitations in scaling RL to biological complexity.</div><div>Looking ahead, RL offers powerful tools for understanding brain function, guiding brain–machine interfaces, and personalizing psychiatric treatment. The convergence of RL and neuroscience offers a promising interdisciplinary lens for advancing our understanding of learning and decision-making in both artificial agents and the human brain.</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 3","pages":"Article 100220"},"PeriodicalIF":0.0,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144713172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Short-window EEG-based auditory attention decoding for neuroadaptive hearing support for smart healthcare 基于短窗口脑电图的听觉注意解码用于智能医疗的神经适应性听力支持

Neuroscience informatics

Pub Date : 2025-07-22 DOI: 10.1016/j.neuri.2025.100222

Ihtiram Raza Khan , Sheng-Lung Peng , Rupali Mahajan , Rajesh Dey

Background

Selective auditory attention the brain's ability to focus on a specific speaker in multi-talker environments is often compromised in individuals with auditory or neurological disorders. While Auditory Attention Decoding (AAD) using EEG has shown promise in detecting attentional focus, existing models primarily utilize temporal or spectral features, often neglecting the synergistic relationships across time, space, and frequency. This limitation significantly reduces decoding accuracy, particularly in short decision windows, which are crucial for real-time applications like neuro-steered hearing aids. This study is to enhance short-window AAD performance by fully leveraging multi-dimensional EEG characteristics.

Methods

To address this, we propose TSF-AADNet, a novel neural framework that integrates temporal–spatial and frequency–spatial features using dual-branch architectures and advanced attention-based fusion.

Results

Tested on KULeuven and DTU datasets, TSF-AADNet achieves 91.8% and 81.1% accuracy at 0.1-second windows—outperforming the state-of-the-art by up to 7.99%.

Conclusions

These results demonstrate the model's potential in enabling precise, real-time attention tracking for hearing impairment diagnostics and next-generation neuroadaptive auditory prosthetics.

选择性听觉注意在多说话的环境中，大脑专注于特定说话者的能力在听觉或神经障碍的个体中经常受到损害。虽然利用脑电图进行听觉注意解码（AAD）在检测注意焦点方面显示出前景，但现有模型主要利用时间或频谱特征，往往忽略了时间、空间和频率之间的协同关系。这种限制大大降低了解码的准确性，特别是在短决策窗口中，这对于神经导向助听器等实时应用至关重要。本研究旨在充分利用脑电图的多维特征，提高短窗口AAD的性能。为了解决这个问题，我们提出了一种新的神经框架TSF-AADNet，它使用双分支架构和先进的基于注意力的融合技术集成了时空和频率空间特征。结果在KULeuven和DTU数据集上测试，TSF-AADNet在0.1秒窗口下的准确率分别达到91.8%和81.1%，比目前最先进的准确率高出7.99%。这些结果证明了该模型在精确、实时的注意力跟踪听力障碍诊断和下一代神经适应性听觉假肢方面的潜力。

{"title":"Short-window EEG-based auditory attention decoding for neuroadaptive hearing support for smart healthcare","authors":"Ihtiram Raza Khan , Sheng-Lung Peng , Rupali Mahajan , Rajesh Dey","doi":"10.1016/j.neuri.2025.100222","DOIUrl":"10.1016/j.neuri.2025.100222","url":null,"abstract":"<div><h3>Background</h3><div>Selective auditory attention the brain's ability to focus on a specific speaker in multi-talker environments is often compromised in individuals with auditory or neurological disorders. While Auditory Attention Decoding (AAD) using EEG has shown promise in detecting attentional focus, existing models primarily utilize temporal or spectral features, often neglecting the synergistic relationships across time, space, and frequency. This limitation significantly reduces decoding accuracy, particularly in short decision windows, which are crucial for real-time applications like neuro-steered hearing aids. This study is to enhance short-window AAD performance by fully leveraging multi-dimensional EEG characteristics.</div></div><div><h3>Methods</h3><div>To address this, we propose TSF-AADNet, a novel neural framework that integrates temporal–spatial and frequency–spatial features using dual-branch architectures and advanced attention-based fusion.</div></div><div><h3>Results</h3><div>Tested on KULeuven and DTU datasets, TSF-AADNet achieves 91.8% and 81.1% accuracy at 0.1-second windows—outperforming the state-of-the-art by up to 7.99%.</div></div><div><h3>Conclusions</h3><div>These results demonstrate the model's potential in enabling precise, real-time attention tracking for hearing impairment diagnostics and next-generation neuroadaptive auditory prosthetics.</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 3","pages":"Article 100222"},"PeriodicalIF":0.0,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144696611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multimodal lightweight neural network for Alzheimer's disease diagnosis integrating neuroimaging and cognitive scores 综合神经影像学和认知评分的阿尔茨海默病多模态轻量级神经网络诊断

Neuroscience informatics

Pub Date : 2025-07-10 DOI: 10.1016/j.neuri.2025.100218

Bhoomi Gupta , Ganesh Kanna Jegannathan , Mohammad Shabbir Alam , Kottala Sri Yogi , Janjhyam Venkata Naga Ramesh , Vemula Jasmine Sowmya , Isa Bayhan

Conventional single-modal approaches for auxiliary diagnosis of Alzheimer's disease (AD) face several limitations, including insufficient availability of expertly annotated imaging datasets, unstable feature extraction, and high computational demands. To address these challenges, we propose Light-Mo-DAD, a lightweight multimodal diagnostic neural network designed to integrate MRI, PET imaging, and neuropsychological assessment scores for enhanced AD detection. In the neuroimaging feature extraction module, redundancy-reduced convolutional operations are employed to capture fine-grained local features, while a global filtering mechanism enables the extraction of holistic spatial patterns. Multimodal feature fusion is achieved through spatial image registration and summation, allowing for effective integration of structural and functional imaging modalities. The neurocognitive feature extraction module utilizes depthwise separable convolutions to process cognitive assessment data, which are then fused with multimodal imaging features. To further enhance the model's discriminative capacity, transfer learning techniques are applied. A multilayer perceptron (MLP) classifier is incorporated to capture complex feature interactions and improve diagnostic precision. Evaluation on the ADNI dataset demonstrates that Light-Mo-DAD achieves 98.0% accuracy, 98.5% sensitivity, and 97.5% specificity, highlighting its robustness in early AD detection. These results suggest that the proposed architecture not only enhances diagnostic accuracy but also offers strong potential for real-time, mobile deployment in clinical settings, supporting neurologists in efficient and reliable Alzheimer's diagnosis.

传统的用于阿尔茨海默病（AD）辅助诊断的单模态方法面临一些限制，包括专业注释的成像数据集可用性不足，特征提取不稳定以及计算需求高。为了解决这些挑战，我们提出了Light-Mo-DAD，这是一个轻量级的多模态诊断神经网络，旨在整合MRI， PET成像和神经心理学评估评分，以增强AD的检测。在神经成像特征提取模块中，采用了减少冗余的卷积运算来捕获细粒度的局部特征，同时采用了全局过滤机制来提取整体空间模式。通过空间图像配准和求和实现多模态特征融合，从而实现结构和功能成像模式的有效整合。神经认知特征提取模块利用深度可分离卷积来处理认知评估数据，然后将其与多模态成像特征融合。为了进一步提高模型的判别能力，本文采用了迁移学习技术。采用多层感知器（MLP）分类器捕获复杂的特征交互，提高诊断精度。对ADNI数据集的评估表明，Light-Mo-DAD的准确率为98.0%，灵敏度为98.5%，特异性为97.5%，突出了其在早期AD检测中的稳健性。这些结果表明，所提出的架构不仅提高了诊断的准确性，而且为临床环境中的实时、移动部署提供了强大的潜力，支持神经科医生高效、可靠地诊断阿尔茨海默病。

{"title":"Multimodal lightweight neural network for Alzheimer's disease diagnosis integrating neuroimaging and cognitive scores","authors":"Bhoomi Gupta , Ganesh Kanna Jegannathan , Mohammad Shabbir Alam , Kottala Sri Yogi , Janjhyam Venkata Naga Ramesh , Vemula Jasmine Sowmya , Isa Bayhan","doi":"10.1016/j.neuri.2025.100218","DOIUrl":"10.1016/j.neuri.2025.100218","url":null,"abstract":"<div><div>Conventional single-modal approaches for auxiliary diagnosis of Alzheimer's disease (AD) face several limitations, including insufficient availability of expertly annotated imaging datasets, unstable feature extraction, and high computational demands. To address these challenges, we propose Light-Mo-DAD, a lightweight multimodal diagnostic neural network designed to integrate MRI, PET imaging, and neuropsychological assessment scores for enhanced AD detection. In the neuroimaging feature extraction module, redundancy-reduced convolutional operations are employed to capture fine-grained local features, while a global filtering mechanism enables the extraction of holistic spatial patterns. Multimodal feature fusion is achieved through spatial image registration and summation, allowing for effective integration of structural and functional imaging modalities. The neurocognitive feature extraction module utilizes depthwise separable convolutions to process cognitive assessment data, which are then fused with multimodal imaging features. To further enhance the model's discriminative capacity, transfer learning techniques are applied. A multilayer perceptron (MLP) classifier is incorporated to capture complex feature interactions and improve diagnostic precision. Evaluation on the ADNI dataset demonstrates that Light-Mo-DAD achieves 98.0% accuracy, 98.5% sensitivity, and 97.5% specificity, highlighting its robustness in early AD detection. These results suggest that the proposed architecture not only enhances diagnostic accuracy but also offers strong potential for real-time, mobile deployment in clinical settings, supporting neurologists in efficient and reliable Alzheimer's diagnosis.</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 3","pages":"Article 100218"},"PeriodicalIF":0.0,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144634328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An automated measurement of head circumference using CT scans: An application in children with head abnormalities 使用CT扫描自动测量头围：在头部异常儿童中的应用

Neuroscience informatics

Pub Date : 2025-06-27 DOI: 10.1016/j.neuri.2025.100217

Priscila Satomi Acamine , Rafael Maffei Loureiro , Lucas dos Anjos Longas , Fabio Augusto Ribeiro Dalpra , Luigi Villanova Machado de Barros Lago , Larissa Vasconcellos de Moraes , Paulo Cesar Filho Estevam , Luiz Otávio Vittorelli , Lucas Silva Kallás , Ana Paula Antunes Pascalicchio Bertozzi , Maria Isabel Barros Guinle , Gilberto Szarf , Saulo Duarte Passos , Birajara Soares Machado , Joselisa Péres Queiroz De Paiva

Manual measurement of head circumference has been a widely adopted method of neurodevelopmental evaluation in both clinical and research settings. Here, we propose a method that uses axial slices of computerized tomography (CT) scans to detect the largest outer margin for measurement. Our method can both complement conventional tape measurements or be applied as a standalone tool, especially in the context of retrospective big data analysis. We applied our algorithm in a set of 74 head CT scans obtained from individual children (8,5 ± 14,1 months old). The method proved to be concordant

(ICC [2, k] = 0.99)

, consistent (

ICC [3, k]

= 1), and showed a correlation of 0.988 compared to obtaining manual head circumferences by specialists. Our method is a reliable alternative to conventional manual measurements of head circumference. It can be readily applied in macrocephaly and microcephaly screening studies and in growth reference charts for syndromes related to head alterations.

人工测量头围已成为临床和研究中广泛采用的神经发育评估方法。在这里，我们提出了一种使用计算机断层扫描（CT）轴向切片来检测测量的最大外缘的方法。我们的方法既可以补充传统的卷尺测量，也可以作为一个独立的工具应用，特别是在回顾性大数据分析的背景下。我们将我们的算法应用于74份来自个体儿童（8.5±14.1个月）的头部CT扫描。该方法被证明是一致的（ICC[2,k]=0.99），一致的（ICC[3,k] = 1），与专家手工获得的头部周长相比，显示出0.988的相关性。我们的方法是一个可靠的替代传统的手工测量头围。它可以很容易地应用于大头畸形和小头畸形的筛查研究以及与头部改变相关的综合征的生长参考图表。

{"title":"An automated measurement of head circumference using CT scans: An application in children with head abnormalities","authors":"Priscila Satomi Acamine , Rafael Maffei Loureiro , Lucas dos Anjos Longas , Fabio Augusto Ribeiro Dalpra , Luigi Villanova Machado de Barros Lago , Larissa Vasconcellos de Moraes , Paulo Cesar Filho Estevam , Luiz Otávio Vittorelli , Lucas Silva Kallás , Ana Paula Antunes Pascalicchio Bertozzi , Maria Isabel Barros Guinle , Gilberto Szarf , Saulo Duarte Passos , Birajara Soares Machado , Joselisa Péres Queiroz De Paiva","doi":"10.1016/j.neuri.2025.100217","DOIUrl":"10.1016/j.neuri.2025.100217","url":null,"abstract":"<div><div>Manual measurement of head circumference has been a widely adopted method of neurodevelopmental evaluation in both clinical and research settings. Here, we propose a method that uses axial slices of computerized tomography (CT) scans to detect the largest outer margin for measurement. Our method can both complement conventional tape measurements or be applied as a standalone tool, especially in the context of retrospective big data analysis. We applied our algorithm in a set of 74 head CT scans obtained from individual children (8,5 ± 14,1 months old). The method proved to be concordant <span><math><mo>(</mo><mrow><mi>ICC</mi></mrow><mo>[</mo><mn>2</mn><mo>,</mo><mi>k</mi><mo>]</mo><mo>=</mo><mn>0.99</mn><mo>)</mo></math></span>, consistent (<span><math><mrow><mi>ICC</mi></mrow><mo>[</mo><mn>3</mn><mo>,</mo><mi>k</mi><mo>]</mo></math></span> = 1), and showed a correlation of 0.988 compared to obtaining manual head circumferences by specialists. Our method is a reliable alternative to conventional manual measurements of head circumference. It can be readily applied in macrocephaly and microcephaly screening studies and in growth reference charts for syndromes related to head alterations.</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 3","pages":"Article 100217"},"PeriodicalIF":0.0,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144535145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Beyond the numbers: App-enabled stroke prediction system for high-risk individuals in imbalanced datasets 数字之外：应用程序支持的中风预测系统，用于不平衡数据集中的高风险人群

Neuroscience informatics

Pub Date : 2025-06-18 DOI: 10.1016/j.neuri.2025.100215

Abrar Faiaz Eram , Aliva Sadnim Mahmud , Marwan Mostafa Khadem , Md Amimul Ihsan

Background:

Brain stroke, characterized by interrupted blood flow to the brain, poses significant mortality risks and quality-of-life impacts. While machine learning approaches show promise in stroke prediction, current research often relies on synthetic data to address dataset imbalance, potentially compromising real-world model performance in clinical settings.

Method:

This research proposes an alternative approach focusing on recall as the primary evaluation metric for stroke prediction, specifically targeting the reduction of false negatives. In the context of stroke diagnosis, where missed detection can lead to severe consequences or fatality, recall is crucial for directly measuring the model's ability to identify actual stroke cases.

Results:

Three superior models were identified: Logistic Regression, an Ensemble using Soft Voting (combining Gaussian Naive Bayes and Logistic Regression), and customized Support Vector Machine. Exceptional stroke prediction was achieved with recall values of 92%, 92%, and 94%, respectively. Interpretability is enhanced through SHAP applied to the best one. While previous methods showed recall values between 5.6% and 40%, this approach outperformed these benchmarks (94%). Current research emphasizes accuracy metrics, relying on oversampling, being inappropriate for sensitive medical datasets. The pitfall is a slight increase in false positives, which is tolerable because the cost of misdiagnosing a stroke patient far outweighs the reverse scenario.

Conclusions:

The research demonstrates the effectiveness of focusing on recall as an evaluation metric for stroke prediction, minimizing false negative predictions. To facilitate practical implementation, a mobile application incorporating the best-performing model was included. A primary screening which can supplement doctors in stroke diagnosis and prediction was proposed.

背景：脑中风以脑部血流中断为特征，具有显著的死亡风险和生活质量影响。虽然机器学习方法在中风预测方面显示出前景，但目前的研究往往依赖于合成数据来解决数据集失衡问题，这可能会影响临床环境中真实世界模型的性能。方法：本研究提出了一种替代方法，将召回率作为中风预测的主要评估指标，特别是针对减少假阴性。在中风诊断的背景下，遗漏的检测可能导致严重的后果或死亡，召回对于直接测量模型识别实际中风病例的能力至关重要。结果：确定了三种优越的模型：逻辑回归、软投票集成（结合高斯朴素贝叶斯和逻辑回归）和定制支持向量机。异常脑卒中预测的召回率分别为92%、92%和94%。可解释性通过将SHAP应用于最好的代码而得到增强。虽然以前的方法显示召回值在5.6%到40%之间，但这种方法优于这些基准（94%）。目前的研究强调准确性指标，依赖于过采样，不适合敏感的医疗数据集。陷阱是假阳性的轻微增加，这是可以容忍的，因为误诊中风患者的成本远远超过相反的情况。结论：本研究证明了将回忆作为卒中预测的评估指标的有效性，最大限度地减少了错误的负面预测。为了便于实际实施，我们还提供了一个包含最佳性能模型的移动应用程序。提出了一种辅助医生进行脑卒中诊断和预测的初步筛查方法。

{"title":"Beyond the numbers: App-enabled stroke prediction system for high-risk individuals in imbalanced datasets","authors":"Abrar Faiaz Eram , Aliva Sadnim Mahmud , Marwan Mostafa Khadem , Md Amimul Ihsan","doi":"10.1016/j.neuri.2025.100215","DOIUrl":"10.1016/j.neuri.2025.100215","url":null,"abstract":"<div><h3>Background:</h3><div>Brain stroke, characterized by interrupted blood flow to the brain, poses significant mortality risks and quality-of-life impacts. While machine learning approaches show promise in stroke prediction, current research often relies on synthetic data to address dataset imbalance, potentially compromising real-world model performance in clinical settings.</div></div><div><h3>Method:</h3><div>This research proposes an alternative approach focusing on recall as the primary evaluation metric for stroke prediction, specifically targeting the reduction of false negatives. In the context of stroke diagnosis, where missed detection can lead to severe consequences or fatality, recall is crucial for directly measuring the model's ability to identify actual stroke cases.</div></div><div><h3>Results:</h3><div>Three superior models were identified: Logistic Regression, an Ensemble using Soft Voting (combining Gaussian Naive Bayes and Logistic Regression), and customized Support Vector Machine. Exceptional stroke prediction was achieved with recall values of 92%, 92%, and 94%, respectively. Interpretability is enhanced through SHAP applied to the best one. While previous methods showed recall values between 5.6% and 40%, this approach outperformed these benchmarks (94%). Current research emphasizes accuracy metrics, relying on oversampling, being inappropriate for sensitive medical datasets. The pitfall is a slight increase in false positives, which is tolerable because the cost of misdiagnosing a stroke patient far outweighs the reverse scenario.</div></div><div><h3>Conclusions:</h3><div>The research demonstrates the effectiveness of focusing on recall as an evaluation metric for stroke prediction, minimizing false negative predictions. To facilitate practical implementation, a mobile application incorporating the best-performing model was included. A primary screening which can supplement doctors in stroke diagnosis and prediction was proposed.</div></div>","PeriodicalId":74295,"journal":{"name":"Neuroscience informatics","volume":"5 3","pages":"Article 100215"},"PeriodicalIF":0.0,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144338309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0