AI in neuroscience最新文献

{"title":"Single cell transcriptomic analysis by deep learning identifies novel microglial transcription regulations in Alzheimer's Disease.","authors":"Maitry Ronakbhai Trivedi, Jay Shah, Yi Su, Eric M Reiman, Teresa Wu, Qi Wang","doi":"10.1177/2997979x261472562","DOIUrl":"10.1177/2997979x261472562","url":null,"abstract":"<p><p>Artificial Intelligence (AI) has been increasingly applied to investigate genetic irregularities associated with Alzheimer's Disease (AD). However, its potential to uncover deeper, more detailed molecular and cellular mechanisms remains underexplored, primarily due to limitations in integrating large-scale data and capturing the complex, cell-type-specific dynamics involved in AD pathology. Single-cell RNA sequencing (scRNA-seq) has emerged as a powerful tool in transcriptomics, offering high-resolution, cell-specific insights into complex biological systems. Despite this advancement, a significant gap remains in identifying both common and cell-type-specific transcriptomic signatures that define AD-related cellular and molecular processes. To address this, we propose a deep learning framework leveraging a Multi-Layer Perceptron (MLP) to classify AD versus control nuclei using scRNA-seq data from the Religious Orders Study/Memory and Aging Project (ROSMAP). We focus on microglial subclusters, particularly those representing homeostatic and activated states, to train the MLP model for optimal classification performance. We utilize the predicted embeddings from the MLP to model a disease progression trajectory for each of the datasets. Our model demonstrates strong performance in both classification and disease trajectory inference. To enhance interpretability, SHapley Additive exPlanations (SHAP) are applied to identify key AD-associated genes. Based on the most salient genes implicated in AD, we built transcription gene regulatory networks, revealing novel transcription factors and regulons for AD pathogenesis. These regulons highlight profound impacts of dysregulations of proteostasis, endoplasmic reticulum (ER) stress responses, and circadian rhythm on synaptic plasticity and neuronal survival in AD, offering a more holistic approach to drug target discovery compared to conventional single-target strategies, potentially leading to greater efficacy in slowing or reversing disease progression. This work demonstrates the transformative potential of AI in elucidating the molecular mechanisms of AD, offering improvements over traditional methods and uncovering novel insights into disease pathogenesis and potential therapeutic targets.</p>","PeriodicalId":520989,"journal":{"name":"AI in neuroscience","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13544018/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148898840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PIGSKIN: PIG SKull stripping with synthetic Images and Neural networks.","authors":"Javid Dadashkarimi, Drew Parker, Nima Broomand Lomer, Ramon Diaz-Arrastia, Douglas H Smith, Victoria Johnson, John Wolf, Ragini Verma","doi":"10.1177/2997979x261429297","DOIUrl":"10.1177/2997979x261429297","url":null,"abstract":"<p><p>Robust brain extraction is an essential first step for quantitative neuroimaging, yet no reliable method exists for porcine MRI despite its importance as a translational neuroscientific animal model, particularly gyrencephalic traumatic brain injury (TBI). Porcine anatomy presents unique challenges, with extensive extracranial fat and complex textures around the olfactory bulbs that confound existing human-based tools. We present PIGSKIN, a deep learning framework trained entirely on synthetic data generated from a small set of expert annotations. Unlike conventional approaches, PIGSKIN models brain and non-brain regions separately, applying distinct clustering and transformation parameters to each. This strategy constrains variability in brain anatomy while allowing greater diversity in extracranial tissues, ensuring anatomically consistent label maps in which the brain remains embedded within its surroundings. Additional transformations introduce further spatial and intensity variability, producing a rich stream of synthetic training pairs. During inference, PIGSKIN operates in a single step at native resolution within a standardized cube, preserving fine-scale anatomical detail and enabling generalization across cohorts differing in breed, acquisition protocol, and injury model. The model achieved comparable to expert consistency accuracy (Dice ≈ 0.97), approaching the range of reported inter-rater reliability. Finally, we show that incorporating co-registered T1- and T2-weighted inputs significantly outperforms single-modality training, underscoring the value of multimodal integration for synthetic data generation. Together, these results establish PIGSKIN as the first systematically validated solution for porcine brain extraction and a framework adaptable to other large animal brain extraction tasks.</p>","PeriodicalId":520989,"journal":{"name":"AI in neuroscience","volume":"2 1","pages":"5-16"},"PeriodicalIF":0.0,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13484249/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148805493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and Validation of a Machine Learning-Derived Kinematic Deviation Index for the Evaluation of Forelimb in Rodents.","authors":"Abel Torres-Espin, Amanda Bernstein, Marwa Soliman, Juan Sebastián Jara, Yunuen Moreno-López, Edmund Hollis","doi":"10.1177/2997979x251370362","DOIUrl":"10.1177/2997979x251370362","url":null,"abstract":"<p><strong>Background: </strong>Rodent models are widely used to study neurological conditions and assess forelimb movement to measure functional performance, deficit, recovery, and treatment effectiveness. Traditional assessment methods based on endpoints such as whether the task is successfully accomplished, while easy to implement, provide limited information on movement patterns critical for assessing different functional strategies. On the other side, detailed kinematic analysis provides granular information on the movement patterns but is difficult to compare across laboratories and may not translate to clinical metrics of upper limb function.</p><p><strong>Methods: </strong>To address these limitations, we developed and validated a machine learning derived kinematic deviation index (KDI) for rodents that mimics current trends in clinical neurological research. The KDI is a unitless summary score that quantifies the difference between an animal movement during a task and its optimal performance derived from spatiotemporal marker sequences without pre-specifying movements.</p><p><strong>Results: </strong>We demonstrate the utility of KDI in assessing reaching and grasping in mice and validate its discrimination between trial endpoints in healthy animals. Furthermore, we show KDI sensitivity to interventions disrupting neurological function, including acute and chronic spinal cord injury and optogenetic disruption of sensorimotor circuits.</p><p><strong>Conclusion: </strong>The KDI provides a comprehensive measure of motor function that bridges the gap between detailed kinematic analysis and simple success/failure metrics, offering a valuable tool for assessing recovery and compensation in rodent models of neurological disorders.</p>","PeriodicalId":520989,"journal":{"name":"AI in neuroscience","volume":"1 3","pages":"124-135"},"PeriodicalIF":0.0,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12889874/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146169248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Instantaneous Frequency: A New Functional Biomarker for Dynamic Brain Causal Networks.","authors":"Haoteng Tang, Siyuan Dai, Lei Guo, Pengfei Gu, Guodong Liu, Alex D Leow, Paul M Thompson, Heng Huang, Liang Zhan","doi":"10.1089/ains.2025.0005","DOIUrl":"10.1089/ains.2025.0005","url":null,"abstract":"<p><strong>Background: </strong>This study introduces instantaneous frequency (IF) analysis as a novel method for characterizing dynamic brain causal networks from functional magnetic resonance imaging blood-oxygen-level-dependent signals.</p><p><strong>Methods: </strong>Effective connectivity, estimated using dynamic causal modeling, is analyzed to derive IF sequences, with the average IF across brain regions serving as a potential biomarker for global network oscillatory behavior.</p><p><strong>Results: </strong>Analysis of data from the Alzheimer's Disease (AD) Neuroimaging Initiative, Open Access Series of Imaging Studies, and Human Connectome Project demonstrates the method's efficacy in distinguishing between clinical and demographic groups, such as cognitive decline stages (e.g., normal control, early mild cognitive impairment [MCI], late MCI, and AD), sex differences, and sleep quality levels.</p><p><strong>Conclusion: </strong>Statistical analyses reveal significant group differences in IF metrics, highlighting its potential as a sensitive indicator for early diagnosis and monitoring of neurodegenerative and cognitive conditions.</p>","PeriodicalId":520989,"journal":{"name":"AI in neuroscience","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12412747/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145014288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DeepLabCut to Automate Behavioral Analysis of Parkinsonism.","authors":"Nabeel Rangoonwala, Khoi Le, Vaibhavi Peshattiwar, Caroline Swain, Dipesh Pokharel, Tatiana White, Thyagarajan Subramanian, Kala Venkiteswaran","doi":"10.1089/ains.2024.0008","DOIUrl":"10.1089/ains.2024.0008","url":null,"abstract":"<p><strong>Background: </strong>Behavioral assessment of parkinsonism often relies on human rater evaluation. However, human biases and variability necessitate larger sample sizes to maintain validity, leading to extensive video analysis and limiting researchers' time. Recent artificial intelligence (AI) and machine learning (ML) advancements enable efficient data analysis, offering unbiased decision-making and consistency across scenarios, bridging inter-rater differences. While not fully automating jobs, AI/ML boosts productivity when properly trained with diverse data. This study aims to show that AI/ML can assist in the analysis of rat parkinsonian behavioral studies to reduce labor dependence while still maintaining accuracy.</p><p><strong>Methods: </strong>DeepLabCut (DLC), an animal pose estimation software, was used to analyze motor behavior in video recordings of parkinsonian Sprague Dawley rats while they performed the stepping test (<i>n</i> = 24). The stepping test involves observing the animal's locomotor function and motor coordination while it is guided across a flat surface. The amount of adjusting steps was counted over the 1-meter distance. Twenty-eight videos (<i>n</i> = 24 + 4 training videos) were fed into DLC, which then selected 20 frames per video using a k-nearest neighbors' algorithm and subsequently labeled to train the model. This one-time training process took 3 h. The output, which has the tracked coordinates of the forepaw being tested, was fed into a script in R to plot Δ y between consecutive frames. The positive peaks were counted as one step, and large negative peaks were counted as a reset or side switch. The counts for each video were then compared with an independent manual rater.</p><p><strong>Results: </strong>There was good absolute agreement between the two scoring methods, using the two-way random effect model, kappa = 0.9, <i>p</i> < 0.0001. It takes 10-15 min to go through each video manually, but the DLC-assisted scoring resulted in 3-4 min per video. These results show that DLC-assisted scoring produced results that could be on par with manual scoring. In addition, this shows a feasible avenue to integrate AI/ML in parkinsonian behavioral studies to reduce the workload for analysis and eventually, fully automating such tasks.</p>","PeriodicalId":520989,"journal":{"name":"AI in neuroscience","volume":"1 1","pages":"54-59"},"PeriodicalIF":0.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12356275/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144878029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}