Journal of Neuroscience Methods最新文献

{"title":"Spinal cord trunk preparation for analyzing cross-segmental primary afferent signal transmission and modulation","authors":"Caifeng Shao ,&nbsp;Mingwei Zhao ,&nbsp;Kun Yang","doi":"10.1016/j.jneumeth.2025.110440","DOIUrl":"10.1016/j.jneumeth.2025.110440","url":null,"abstract":"<div><h3>Background</h3><div>The spinal cord dorsal horn is pivotal for primary afferent signal transmission and modulation. Primary afferent fibers from each dorsal root arrive at the dorsal horn and travel 1–2 segments caudally and rostrally. Usually, in vitro spinal cord slices or in vivo preparations are employed for primary afferent stimulation and patch-clamp recordings to assess input signals. However, the spinal cord slices lose \"intact\" cross-segmental pathways, and in vivo studies are technically challenging.</div></div><div><h3>New method</h3><div>Here, we describe the preparation of a spinal cord trunk for analyzing afferent signal cross-segmental transmission in adult rats. By combining patch-clamp recording, Lissauer's tract stimulation, and ambient temperature manipulation, our methods enable accessing primary afferent pathways within several segments.</div></div><div><h3>Results</h3><div>Our present spinal trunk preparation can be maintained healthy for about 5 h. Lissauer’s tract stimulation induced evoked excitatory postsynaptic currents (eEPSCs) recorded in 6–10 mm rostrally in the ipsilateral dorsal horn. The eEPSCs, spontaneous EPSCs (sEPSCs), and neural excitability can be modulated by ambient temperature rise. Neuropharmacological studies can also be conducted on this spinal trunk preparation.</div></div><div><h3>Compared with existing methods</h3><div>Compared with conventional in vitro spinal cord slices, our present method maintains a relatively intact cross-segment pathway in the dorsal horn; compared with in vivo study, it avoids mechanical vibration and other technical challenges in living animals.</div></div><div><h3>Conclusion</h3><div>The rodent spinal cord trunk can be maintained for an extended period in a fully submerged chamber; combined with patch clamp recordings, our protocol facilitates the study of primary afferent transmission and modulation in the dorsal horn within adjacent segments.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110440"},"PeriodicalIF":2.7,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143747426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Inverse relationship between nodal strength and nodal power: Insights from separate resting fMRI and EEG datasets","authors":"Tien-Wen Lee ,&nbsp;Gerald Tramontano","doi":"10.1016/j.jneumeth.2025.110438","DOIUrl":"10.1016/j.jneumeth.2025.110438","url":null,"abstract":"<div><h3>Background</h3><div>Regional neural response and network properties have traditionally been studied separately. However, growing evidence suggests a close interplay between regional activity and inter-regional connectivity. This study aimed to examine the relationship between global functional connectivity and regional spontaneous activity, termed the global-to-local relationship.</div></div><div><h3>New method</h3><div>Resting-state fMRI data were parcellated using MOSI (modular analysis and similarity measurements), enabling multi-resolution functional partitioning. For each parcellated cluster, the mean amplitude of low-frequency fluctuations (node power) and its average functional connectivity with the remaining cortex (node strength) were computed. Correlation analyses assessed their relationship. A supplementary analysis was conducted on EEG data (1–30 Hz).</div></div><div><h3>Results</h3><div>A significant negative correlation between node strength and regional power was observed in MRI datasets. One-sample t-tests confirmed robustness across different MOSI resolutions, with individual <em>P</em> values at the level 10⁻⁴ to 10⁻⁵. The negative relationship was also found in EEG data but was restricted to delta (1–4 Hz) and theta (4–8 Hz) bands.</div></div><div><h3>Comparison with existing methods</h3><div>This study introduces two key novel aspects. First, it applies MOSI to the entire cortex, enhancing the comprehensiveness of network analysis. Second, it examines the global influence on regional neural activity, rather than limiting the focus to a specific network.</div></div><div><h3>Conclusions</h3><div>A robust negative relationship between node strength and node power was consistently observed across both MRI and EEG datasets, particularly in lower frequency bands (up to 8 Hz). These findings suggest a framework for investigating how global connectivity shapes regional neural activity, with inhibitory coupling as a potential underlying mechanism.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110438"},"PeriodicalIF":2.7,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143780250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Three-dimensional imaging and computational quantitation as a novel approach to assess nerve fibers, enteric glial cells, mast cells, and the proximity of mast cells to the nerve fibers in human sigmoid mucosal biopsies from healthy subjects","authors":"Pu-Qing Yuan ,&nbsp;Tao Li ,&nbsp;Swapna Mahurkar-Joshi ,&nbsp;Jessica Sohn ,&nbsp;Lin Chang ,&nbsp;Yvette Taché","doi":"10.1016/j.jneumeth.2025.110436","DOIUrl":"10.1016/j.jneumeth.2025.110436","url":null,"abstract":"<div><h3>Background</h3><div>The visualization and quantitation of nerve fibers (NFs), enteric glial cells (EGCs), mast cells (MCs), and their spatial configurations in the human colonic mucosa represent considerable challenges due to the meshed network of these components and the arborizing of NFs in a three-dimensional (3D) structure.</div></div><div><h3>New method</h3><div>We developed a novel approach combining tissue clearing, 3D imaging and computerized quantitation of NFs, EGCs and MCs in sigmoid mucosal biopsies of healthy subjects using a modified CLARITY tissue clearing protocol and adapting Imaris Surfaces Rendering Technology.</div></div><div><h3>Results</h3><div>The cleared colonic biopsies are compatible with immunostaining using 10 marker antibodies and capable of generating 3D images rendering clear spatial views and computational quantitation of NFs, MCs, EGCs, in particular the proximity of MCs to NFs with Imaris 9.7–9.9.</div></div><div><h3>Comparison with existing methods</h3><div>Our modified tissue clearing protocol shortened the membrane lipid removal time to 1 day from the original 1–2 weeks and total tissue clearing time to 3–4 days from the original 2–4 weeks. The 3D images displayed a clear spatial landscape of NFs, MCs and EGCs in the biopsies which cannot be portrayed with 2D images acquired from sections. Computerized quantitation is faster than measuring manually, allowing us to quantify a larger number of samples with less bias.</div></div><div><h3>Conclusion</h3><div>The novel approach enables faster tissue clearing/immunolabeling, high-quality 3D imaging and precise computational quantitation of NFs, cells and proximity of MCs to NFs in human sigmoid biopsies which may allow new insight to detect alterations in colonic-related diseases.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110436"},"PeriodicalIF":2.7,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143767687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Minimizing hearing loss: A novel surgical approach for round window exposure in miniature pigs","authors":"Zhifeng Chen ,&nbsp;Xihang Chen ,&nbsp;Haiqiao Du ,&nbsp;Hongdong Liu ,&nbsp;Yun Hu ,&nbsp;Yun Gao ,&nbsp;Xiao Yang ,&nbsp;Juanjuan Li ,&nbsp;Xianhai Zeng ,&nbsp;Xiaojun Ji ,&nbsp;Weiwei Guo ,&nbsp;Wei Chen ,&nbsp;Shiming Yang","doi":"10.1016/j.jneumeth.2025.110439","DOIUrl":"10.1016/j.jneumeth.2025.110439","url":null,"abstract":"<div><h3>Background</h3><div>A delicate operation for exposing the round window without affecting the hearing function is critical for auditory research. Despite the different surgical approaches proposed for miniature pigs, hearing protection during operation is still challenging. The efficient solution to this problem is significant for accurate auditory research.</div></div><div><h3>New method</h3><div>Eight healthy Diannan miniature pigs, aged three months and of both sexes, with normal hearing, were utilized. Three pigs were employed for anatomical studies. Based on the findings from these anatomical investigations, we propose the facial nerve medial approach designed to expose the round window while preserving hearing function in the porcine model. The remaining five pigs were used to verify the surgical approach and hearing function.</div></div><div><h3>Results</h3><div>The anatomy study confirmed that the lateral side of the round window was the genu of the facial nerve. Removing the medial wall of the genu part of the facial nerve canal allows access to the tympanic cavity and exposes the round window. The facial nerve medial approach effectively exposed the round window in vivo, with no significant change in auditory brainstem response threshold immediately post-operation.</div></div><div><h3>Comparison with existing methods</h3><div>The facial nerve medial approach to exposing the round window preserved the sound transmission structures of the ear, maintaining hearing function post-operation.</div></div><div><h3>Conclusions</h3><div>The facial nerve medial approach with a retroauricular incision allows access to the tympanic cavity and exposes the round window without impacting hearing, making it ideal for accurate auditory research in miniature pigs.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110439"},"PeriodicalIF":2.7,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143767686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of neural responses to whisker and ultrasound stimulation using a novel dual-stimulation protocol","authors":"Ye Yuan,&nbsp;Tian Liu,&nbsp;Jue Wang","doi":"10.1016/j.jneumeth.2025.110435","DOIUrl":"10.1016/j.jneumeth.2025.110435","url":null,"abstract":"<div><h3>Background</h3><div>The sensory system allows organisms to perceive and respond to environmental stimuli. This study investigates neural response differences between whisker and ultrasound stimulation in rats to evaluate cortical specificity to sensory inputs.</div></div><div><h3>New method</h3><div>A novel dual-stimulation protocol combining a step motor and ultrasound system was developed to alternately stimulate the C2 whisker and corresponding barrel column region. Experiments were conducted under varying stimulation sequences (whisker-ultrasound and ultrasound-whisker) and time intervals (10 ms, 25 ms, and 100 ms). Neural response signals were recorded, and statistical analyses (ANOVA and T-test) were performed to compare response amplitudes and peak latencies.</div></div><div><h3>Results</h3><div>Whisker stimulation consistently elicited significantly stronger neural responses than ultrasound stimulation (<em>*p &lt; 0.05</em>), regardless of sequence or interval. The efficiency of neural responses to ultrasound was closely tied to frequency, with higher frequencies producing greater amplitudes and faster latencies. Notably, at a 25 ms interval in the ultrasound-whisker sequence, whisker responses were significantly enhanced compared to whisker stimulation alone, suggesting a pre-activation effect of ultrasound.</div></div><div><h3>Comparison with existing methods</h3><div>Unlike single-modal whisker or ultrasound stimulation, the dual-stimulation protocol can enhance sensory responses, highlighting its neuromodulatory potential.</div></div><div><h3>Conclusion</h3><div>This study reveals distinct cortical activation patterns induced by whisker and ultrasound stimulation. While whisker stimulation is more sensitive, ultrasound stimulation—when optimized for frequency and timing—can effectively modulate neural responses under dual-stimulation protocol. These findings provide insights into ultrasound-based neuromodulation and sensory processing.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110435"},"PeriodicalIF":2.7,"publicationDate":"2025-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive protocol for simplified mouse DRG fixation, processing and F4/80 immunohistochemistry: Overcoming common challenges","authors":"Nicolette Tay ,&nbsp;Ammar Alshammari ,&nbsp;Simranpreet Kaur ,&nbsp;Allison Pettit ,&nbsp;Erica Mu ,&nbsp;Anna Reid ,&nbsp;Ingrid Winkler ,&nbsp;Irina Vetter ,&nbsp;Hana Starobova","doi":"10.1016/j.jneumeth.2025.110434","DOIUrl":"10.1016/j.jneumeth.2025.110434","url":null,"abstract":"<div><h3>Background</h3><div>Dorsal root ganglia (DRGs) contain the cell bodies of sensory neurons and non-neuronal cells that play a role in the pathophysiology of painful inflammatory conditions, such as neuropathic pain. Immunohistochemistry (IHC) is a valuable tool for visualising and quantifying immune cell markers in DRGs, providing important insights into these mechanisms. However, isolating DRGs while preserving cell morphology for IHC staining is technically challenging due to their small size and location within the spinal column.</div></div><div><h3>Objective</h3><div>Using F4/80, a pan monocyte-macrophage marker, we present an optimised protocol for the fixation, harvesting, processing, and IHC staining of formalin-fixed-paraffin-embedded (FFPE) mouse DRGs. This method is designed to maintain tissue integrity and ensure compatibility with downstream histopathological analysis.</div></div><div><h3>New Method</h3><div>The entire spinal column of mouse was fixed in 10 % neutral-buffered formalin at room temperature for 24 h before DRG isolation. DRGs were processed for 9 h, and antigen retrieval was performed using proteinase K.</div></div><div><h3>Results</h3><div>The optimised immersion-fixation approach preserved cellular morphology and antigenicity, ensuring high-quality histological outcomes.</div></div><div><h3>Comparison with Existing Methods</h3><div>While transcardial perfusion remains the gold standard for tissue fixation, it is time-intensive, requires training and raises ethical concerns. Our optimised method of whole spinal column fixation with subsequent tissue isolation is non-invasive and reduces the time between death and fixation in comparison to post-isolation fixation. Additionally, it delivers histological quality likely comparable to that of perfusion-based techniques.</div></div><div><h3>Conclusion</h3><div>This protocol is supported by a grading system to help evaluate variables and select conditions best suited to their experimental goals.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110434"},"PeriodicalIF":2.7,"publicationDate":"2025-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating data mining with transcranial focused ultrasound to refine neuralgia treatment strategies","authors":"Muhammad Attique Khan ,&nbsp;Shrooq Alsenan ,&nbsp;Shabbab Ali Algamdi ,&nbsp;Haya Aldossary ,&nbsp;K. Narasimha Raju ,&nbsp;Jamel Baili ,&nbsp;Muhammad Asim Saleem","doi":"10.1016/j.jneumeth.2025.110433","DOIUrl":"10.1016/j.jneumeth.2025.110433","url":null,"abstract":"<div><h3>Background</h3><div>Neuralgia and other neuropathic pain are difficult to treat owing to their complicated etiology and a wide variety of responses to treatment. The novel neuromodulation technology transcranial focused ultrasound (tFUS) has intriguing implications in targeted non-invasive brain stimulation. Patient-specific variables and neurological processes must be better understood to enhance tFUS for personalized therapy.</div></div><div><h3>Methods</h3><div>In this research, a Machine Learning based Transcranial Focused Ultrasound Personalized Model (ML-tFUSPM) has been proposed to treat neuralgia by combining tFUS with data mining for personalized therapy. Data mining algorithms can examine patient demographics, pain factors, imaging data, and therapy outcomes to uncover response patterns and treatment predictors. According to these results, tFUS may be tailored to each patient by targeting brain regions involved in pain perception and control.</div></div><div><h3>Results</h3><div>Initial studies show that data-driven models and tFUS enhance therapeutic efficacy, side effects, and accuracy. This collaborative endeavor uses data analytics and neuromodulation to customize neuralgia treatment. The new model's emphasis on targeted treatments and predictive analytics gives clinicians evidence-based tools to manage pain more effectively and personally, which might transform the industry.</div></div><div><h3>Comparative analysis</h3><div>The experimental results show that the proposed method has a high accuracy ratio of 97 % compared to other methods.</div></div><div><h3>Conclusion</h3><div>According to this study, computational principles and cutting-edge technology may lead to game-changing neurology and pain management advances.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110433"},"PeriodicalIF":2.7,"publicationDate":"2025-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143710223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Eye-movement artifact correction in infant EEG: A systematic comparison between ICA and Artifact Blocking","authors":"Diksha Srishyla ,&nbsp;Sara Jane Webb ,&nbsp;Mayada Elsabbagh ,&nbsp;Christian O’Reilly ,&nbsp;BASIS Team","doi":"10.1016/j.jneumeth.2025.110405","DOIUrl":"10.1016/j.jneumeth.2025.110405","url":null,"abstract":"<div><h3>Background:</h3><div>Independent Component Analysis (ICA) is a well-established approach to clean EEG and remove the impact of signals of non-neural origin, such as those from muscular activity and eye movements. However, evidence suggests that ICA removes artifacts less effectively in infants than in adults. This study systematically compares ICA and Artifact Blocking (AB), an alternative approach proposed to improve eye-movement artifact correction in infant EEG.</div></div><div><h3>Methods:</h3><div>We analyzed EEG collected from 50 infants between 6 and 18 months of age as part of the International Infant EEG Data Integration Platform (EEG-IP), a longitudinal multi-study dataset. EEG was recorded while infants sat on their caregivers’ laps and watched videos. We used ICA and AB to correct for eye-movement artifacts in the EEG and calculated the proportion of effectively corrected segments, signal-to-noise ratio (SNR), power-spectral density (PSD), and multiscale entropy (MSE) in manually selected EEG segments with and without eye-movement artifacts.</div></div><div><h3>Results:</h3><div>On the one hand, the proportion of effectively corrected segments indicated that ICA corrected eye-movement artifacts (sensitivity) better than AB. SNR and PSD indicated that both AB and ICA correct eye-movement artifacts with equal sensitivity. MSE gave mixed results. On the other hand, AB caused less distortion to the clean segments (specificity) for SNR, PSD, and MSE.</div></div><div><h3>Conclusion:</h3><div>Our results suggest that ICA is more sensitive (i.e., it better removes artifacts) but less specific (it distorts clean signals) than AB for correcting eye-movement artifacts in infant EEG.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110405"},"PeriodicalIF":2.7,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143700677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Brain tumor segmentation with deep learning: Current approaches and future perspectives","authors":"Akash Verma,&nbsp;Arun Kumar Yadav","doi":"10.1016/j.jneumeth.2025.110424","DOIUrl":"10.1016/j.jneumeth.2025.110424","url":null,"abstract":"<div><h3>Background:</h3><div>Accurate brain tumor segmentation from MRI images is critical in the medical industry, directly impacts the efficacy of diagnostic and treatment plans. Accurate segmentation of tumor region can be challenging, especially when noise and abnormalities are present.</div></div><div><h3>Method:</h3><div>This research provides a systematic review of automatic brain tumor segmentation techniques, with a specific focus on the design of network architectures. The review categorizes existing methods into unsupervised and supervised learning techniques, as well as machine learning and deep learning approaches within supervised techniques. Deep learning techniques are thoroughly reviewed, with a particular focus on CNN-based, U-Net-based, transfer learning-based, transformer-based, and hybrid transformer-based methods.</div></div><div><h3>Scope and Coverage:</h3><div>This survey encompasses a broad spectrum of automatic segmentation methodologies, from traditional machine learning approaches to advanced deep learning frameworks. It provides an in-depth comparison of performance metrics, model efficiency, and robustness across multiple datasets, particularly the BraTS dataset. The study further examines multi-modal MRI imaging and its influence on segmentation accuracy, addressing domain adaptation, class imbalance, and generalization challenges.</div></div><div><h3>Comparison with existing methods:</h3><div>The analysis highlights the current challenges in Computer-aided Diagnostic (CAD) systems, examining how different models and imaging sequences impact performance. Recent advancements in deep learning, especially the widespread use of U-Net architectures, have significantly enhanced medical image segmentation. This review critically evaluates these developments, focusing the iterative improvements in U-Net models that have driven progress in brain tumor segmentation. Furthermore, it explores various techniques for improving U-Net performance for medical applications, focussing on its potential for improving diagnostic and treatment planning procedures.</div></div><div><h3>Conclusion:</h3><div>The efficiency of these automated segmentation approaches is rigorously evaluated using the BraTS dataset, a benchmark dataset, part of the annual Multimodal Brain Tumor Segmentation Challenge (MICCAI). This evaluation provides insights into the current state-of-the-art and identifies key areas for future research and development.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110424"},"PeriodicalIF":2.7,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143683191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-level visual processing in the lateral geniculate nucleus revealed using goal-driven deep learning","authors":"Mai Gamal ,&nbsp;Seif Eldawlatly","doi":"10.1016/j.jneumeth.2025.110429","DOIUrl":"10.1016/j.jneumeth.2025.110429","url":null,"abstract":"<div><h3>Background</h3><div>The Lateral Geniculate Nucleus (LGN) is an essential contributor to high-level visual processing despite being an early subcortical area in the visual system. Current LGN computational models focus on its basic properties, with less emphasis on its role in high-level vision.</div></div><div><h3>New method</h3><div>We propose a high-level approach for encoding mouse LGN neural responses to natural scenes. This approach employs two deep neural networks (DNNs); namely VGG16 and ResNet50, as goal-driven models. We use these models as tools to better understand visual features encoded in the LGN.</div></div><div><h3>Results</h3><div>Early layers of the DNNs represent the best LGN models. We also demonstrate that numerosity, as a high-level visual feature, is encoded, along with other visual features, in LGN neural activity. Results demonstrate that intermediate layers are better in representing numerosity compared to early layers. Early layers are better at predicting simple visual features, while intermediate layers are better at predicting more complex features. Finally, we show that an ensemble model of an early and an intermediate layer achieves high neural prediction accuracy and numerosity representation.</div></div><div><h3>Comparison with existing method(s)</h3><div>Our approach emphasizes the role of analyzing the inner workings of DNNs to demonstrate the representation of a high-level feature such as numerosity in the LGN, as opposed to the common belief about the simplicity of the LGN.</div></div><div><h3>Conclusions</h3><div>We demonstrate that goal-driven DNNs can be used as high-level vision models of the LGN for neural prediction and as an exploration tool to better understand the role of the LGN.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110429"},"PeriodicalIF":2.7,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143692425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}