Neural Components Research Articles

Endocrine disrupting chemicals (EDCs) and the neuroendocrine system: Beyond estrogen, androgen, and thyroid.

Hundreds of anthropogenic chemicals occupy our bodies, a situation that threatens the health of present and future generations. This chapter focuses on endocrine disrupting compounds (EDCs), both naturally occurring and man-made, that affect the neuroendocrine system to adversely impact health, with an emphasis on reproductive and metabolic pathways. The neuroendocrine system is highly sexually dimorphic and essential for maintaining homeostasis and appropriately responding to the environment. Comprising both neural and endocrine components, the neuroendocrine system is hormone sensitive throughout life and touches every organ system in the body. The integrative nature of the neuroendocrine system means that EDCs can have multi-system effects. Additionally, because gonadal hormones are essential for the sex-specific organization of numerous neuroendocrine pathways, endocrine disruption of this programming can lead to permanent deficits. Included in this review is a brief history of the neuroendocrine disruption field and a thorough discussion of the most common and less well understood neuroendocrine disruption modes of action. Also provided are extensive examples of how EDCs are likely contributing to neuroendocrine disorders such as obesity, and evidence that they have the potential for multi-generational effects.

Chapter 20 - The circadian system in Parkinson's disease, multiple system atrophy, and progressive supranuclear palsy

Circadian organization of physiology and behavior is an important biologic process that allows organisms to anticipate and prepare for predictable changes in the environment. Circadian disruptions are associated with a wide range of health issues. In patients with neurodegenerative diseases, alterations of circadian rhythms are among the most common and debilitating symptoms. Although a growing awareness of these symptoms has occurred during the last decade, their underlying neuropathophysiologic circuitry remains poorly understood and, consequently, no effective therapeutic strategies are available to alleviate these health issues. Recent studies have examined the neuropathologic status of the different neural components of the circuitry governing the generation of circadian rhythms in neurodegenerative diseases. In this review, we will dissect the potential contribution of dysfunctions in the different nodes of this circuitry to circadian alterations in patients with parkinsonism-linked neurodegenerative diseases (namely, Parkinson's disease, multiple system atrophy, and progressive supranuclear palsy). A deeper understanding of these mechanisms will provide not only a better understanding of disease neuropathophysiology but also holds promise for the development of more effective and mechanisms-based therapies.

Effects of Coffee and Its Components on the Gastrointestinal Tract and the Brain-Gut Axis.

Coffee is one of the most popular beverages consumed worldwide. Roasted coffee is a complex mixture of thousands of bioactive compounds, and some of them have numerous potential health-promoting properties that have been extensively studied in the cardiovascular and central nervous systems, with relatively much less attention given to other body systems, such as the gastrointestinal tract and its particular connection with the brain, known as the brain–gut axis. This narrative review provides an overview of the effect of coffee brew; its by-products; and its components on the gastrointestinal mucosa (mainly involved in permeability, secretion, and proliferation), the neural and non-neural components of the gut wall responsible for its motor function, and the brain–gut axis. Despite in vitro, in vivo, and epidemiological studies having shown that coffee may exert multiple effects on the digestive tract, including antioxidant, anti-inflammatory, and antiproliferative effects on the mucosa, and pro-motility effects on the external muscle layers, much is still surprisingly unknown. Further studies are needed to understand the mechanisms of action of certain health-promoting properties of coffee on the gastrointestinal tract and to transfer this knowledge to the industry to develop functional foods to improve the gastrointestinal and brain–gut axis health.

Artificial synapses that exploit ionic modulation for perception and integration

We propose the first two-terminal sodium-ion artificial synapse (SIAS) with nanoparticle/electrolyte architecture to emulate behaviors of a biological synapse. The device emulates essential synaptic functions such as postsynaptic current and paired-pulse facilitation by employing sodium-ion doping. The synaptic event uses only 0.55 pW at an ultra-low operating voltage of −6 mV, sensitive responsiveness. Pain perception nociceptors, with pain sensitization/desensitization properties, are realized using SIAS. Artificial neural components with multiple inputs also implement dendritic integration functions such as spatiotemporal learning rules and classical conditional learning. Furthermore, benefiting from the nanostructured active layer and solid-state polymer electrolyte, the flexible device on stainless steel substrate induced different deformation curvatures. The proposed SIAS device with capabilities of parallel information processing and pain perception could apply to future bioinspired flexible sensing-processing nervous systems.

Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning.

An important challenge in hyperspectral imaging tasks is to cope with the large number of spectral bins. Common spectral data reduction methods do not take prior knowledge about the task into account. Consequently, sparsely occurring features that may be essential for the imaging task may not be preserved in the data reduction step. Convolutional neural network (CNN) approaches are capable of learning the specific features relevant to the particular imaging task, but applying them directly to the spectral input data is constrained by the computational efficiency. We propose a novel supervised deep learning approach for combining data reduction and image analysis in an end-to-end architecture. In our approach, the neural network component that performs the reduction is trained such that image features most relevant for the task are preserved in the reduction step. Results for two convolutional neural network architectures and two types of generated datasets show that the proposed Data Reduction CNN (DRCNN) approach can produce more accurate results than existing popular data reduction methods, and can be used in a wide range of problem settings. The integration of knowledge about the task allows for more image compression and higher accuracies compared to standard data reduction methods.

Quenching of an antiferromagnet into high resistivity states using electrical or ultrashort optical pulses

Ultra-fast dynamics, insensitivity to external magnetic fields, or absence of magnetic stray fields are examples of properties that make antiferromagnets of potential use in the development of spintronic devices. Similar to their ferromagnetic counterparts, antiferromagnets can store information in the orientations of the collective magnetic order vector. However, also in analogy to ferromagnets, the readout magnetoresistivity signals in simple antiferromagnetic films have been weak and the extension of the electrical reorientation mechanism to optics has not been achieved. Here we report reversible and reproducible quenching of an antiferromagnetic CuMnAs film by either electrical or ultrashort optical pulses into nano-fragmented domain states. The resulting resistivity changes approach 20\% at room temperature, which is comparable to the giant magnetoresistance ratios in ferromagnetic multilayers. We also obtain a signal readout by optical reflectivity. The analog time-dependent switching and relaxation characteristics of our devices can mimic functionality of spiking neural network components.

System for Automated Infrared Image Processing Based on Neural Network Technologies

The article describes the main problems that non-destructive control experts encounter in processing and analyzing infrared (IR) images obtained using thermographic cameras: noisiness, information redundancy, excessive data volume, etc. A gradual process of automated digital processing and analysis of images on the basis of a software system is considered, which allows not only overcoming the drawbacks of the raw data described above, but also drawing a conclusion about the state of the object based on the obtained image independently. The ability to make expert decision is provided by a neural network component in a system on the basis of Kohonen network used for cluster image analysis. There are the examples of system operation applied to IR images of the lining of solid domestic waste (SDW) incinerator. The use of such systems allows reducing workload of an expert in processing large volumes of data and reducing the probability of errors due to such effect as vision "blurring".

A Fascial Reinterpretation of the Classical Female Pelvic Surgical Anatomy: Seeing Things from a Different Angle

The classical surgical anatomy of the female pelvis was born with radical hysterectomy [1] and focused on the pivotal role of the lateral parametrium, a conceptually complex structure, an artifact of surgical anatomy [2] without which the whole classical model would collapse. Here, using natural planes, we tried to simplify the puzzle of the virtual spaces surrounding this structure [3,4]. With the aim of better conceptualizing the classical model of the female pelvic surgical anatomy, we broadened its perspective, which had been narrowly focused on the historic gynecologic setting, by developing a comprehensive model of pelvic retroperitoneal compartmentalization. This dissection was based on the invariable anatomic (fasciae) rather than the surgical-anatomic (parametrium) structures and aimed at providing a holistic, more user-friendly approach intended for surgical and educational purposes [5]. Because each compartment has its own surgical function (hence the name), the excavation of a single compartment may be used as a rational guide to tailor surgery to the site of the pathologic condition to be treated or the type of procedure required, whereas the compartments' sequential development may be useful in planning surgical strategies. Redefining the classical model according to the anatomic fascial planes of dissection potentially allows for an intrinsic surgical reproducibility, minimizing dissective bias. A reinterpretation of the known anatomy is required to enhance education. The breaking down of such a complex system (the pelvis) into smaller parts (compartments) will hopefully provide a useful guide for conceptualization and navigation; surgical navigation requires a holistic mental map and a few invariable anatomic reference points or landmarks. A step-by-step laparoscopic demonstration of the fascial model, developed on a fresh frozen female pelvis, and its correlation with the classical female retroperitoneal surgical anatomy. Cadaver Laboratory, Department of Legal Medicine, University of Turin. The first part of the video shows the progressive development of the 3 hemicompartments in the right hemipelvis and of the fourth median compartment after the identification of 3 invariable anatomic reference points: the obliterated umbilical artery, the ureter, and the sacrouterine ligament as superficial landmarks of 3 deeper fascial-ligamentous structures: the umbilicovesical fascia, the urogenital-hypogastric fascia, and the sacropubic ligament, respectively (Figure 1). The areas delimited by the aforementioned deep fascial ligamentous structures have been designated as compartments: • the right parietal hemicompartment, so called because it is bordered by the sidewall of the pelvis, lateral to the umbilicovesical fascia • the right vascular hemicompartment, so called because of the presence of the internal iliac vessel's visceral branches between the umbilicovesical fascia and the urogenital-hypogastric fascia • the visceral compartment, so called because it contains the pelvic organs between the sacropubic ligaments • the right neural hemicompartment, so called because of the presence of the organ-specific vegetative bundles, medial to the urogenital-hypogastric fascia. The second part of the video describes the retrorectal, presacral, and retropubic connection areas between the neural, vascular, and parietal hemicompartments of each hemipelvis, justifying their overall crescent shape. Finally, the spaces of classical surgical anatomy included in each hemicompartment are listed not only according to their anatomic criterion, but also according to their functional criterion. In fact, the parietal compartment should be developed for the evaluation of the pelvic lymph node status or during exenterative and urogynecologic procedures. The vascular compartment must be prepared when sectioning of the vascular visceral pedicles at their origin is required. Development of the neural compartment is required whenever visceral neural components are to be spared. The visceral compartment has to be developed for complete organ mobilization and exposure. Taken as a whole, our 4-compartment model of pelvic anatomic surgery is intended for use in planning and optimizing surgical strategies. Moreover, it is potentially able to simplify surgical teaching and training, allowing the fitting together of puzzle-like pieces of disjointed organ-specific retroperitoneal spaces according to their function (Figure 2). The correlation of this approach to clinical outcomes is still being determined.

Photoreceptor cells and RPE contribute to the development of diabetic retinopathy.

Diabetic retinopathy (DR) is a leading cause of blindness. It has long been regarded as vascular disease, but work in the past years has shown abnormalities also in the neural retina. Unfortunately, research on the vascular and neural abnormalities have remained largely separate, instead of being integrated into a comprehensive view of DR that includes both the neural and vascular components. Recent evidence suggests that the most predominant neural cell in the retina (photoreceptors) and the adjacent retinal pigment epithelium (RPE) play an important role in the development of vascular lesions characteristic of DR. This review summarizes evidence that the outer retina is altered in diabetes, and that photoreceptors and RPE contribute to retinal vascular alterations in the early stages of the retinopathy. The possible molecular mechanisms by which cells of the outer retina might contribute to retinal vascular damage in diabetes also are discussed. Diabetes-induced alterations in the outer retina represent a novel therapeutic target to inhibit DR.

Neural component analysis: A spatial filter for electroencephalogram analysis

BackgroundSpatial filtering and source separation are valuable tools in the analysis of EEG data. However, despite the well-known spatial localisation of individual cognitive processes within the brain, the available methods for source separation, such as the widely used blind source separation technique, do not take into account the spatial distributions and locations of sources. This can result in sub-optimal source identification. New methodWe present a new method for deriving a spatial filter for EEG data that attempts to identify sources that are maximally spatially distinct from one another in terms of the spatial distributions of their projections. ResultsWe first evaluate our method with simulated EEG and show that it is able to separate EEG signals into components with distinct spatial distributions that closely resemble the original simulated sources. We also evaluate our method with real EEG and show it is able to identify a spatial filter that can be used to significantly improve classification accuracy of the P300 event-related potential (ERP). Comparison with existing methodsWe compare our method to a state of the art blind source separation methods, fast independent component analysis (ICA) and common spatial patterns (CSP). We evaluate the methods suitability for a common source separation application, analysis of ERPs. ConclusionsOur results show that our method is well suited to identifying spatial filters for EEG analysis. This has potential applications in a wide range of EEG signal processing applications.

The importance of the freeway space in orofacial myofunctional therapy

The purpose of this article is to explain the importance of evaluating and documenting the freeway space (interdental rest position), in initial examinations and at the completion of therapy in patients exhibiting an open mandibular resting position. The freeway space measurements obtained can be shared with parents, patients, and referral sources. To this end, the muscular, neural, and sensory components related to the freeway space are discussed, as well as airway, dental eruption patterns, tactile feedback, and orthodontic relapse considerations. An extraoral method for evaluating the dimensions of the anterior freeway space is proposed.

Cutaneous Eccrine Adenolipoma with a Spindle Cell Component

Abstract Introduction/Objective Cutaneous adenolipoma is a rare, benign variant of solitary lipoma characterized by a mature adipocytic proliferation with entrapped eccrine or apocrine sweat glands. It is found predominately in middle-aged females, with a predilection for lower extremities including the thigh and gluteal regions. Cutaneous adenolipoma is presumed to be a hamartomatous process. Herein, we report a unique case of a cutaneous eccrine adenolipoma with a spindle cell component, namely, an early evolving spindle cell adenolipoma. Methods A 43-year-old female with a recurrent history of hidradenitis presented with a 2.7 cm x 1.5 cm x 0.7 cm soft, mobile nodule on the right posterior thigh. It had been present for at least four months and clinically resembled a lipoma. An excisional punch biopsy was performed for pathological evaluation. A concurrent nevus lipomatosus superficialis was diagnosed from the right perineum. Results Gross examination of the thigh nodule revealed fragments of a tan-yellow, lobulated, fatty lesion. Microscopic examination demonstrated a benign lipomatous proliferation with entrapped eccrine glands associated with a surrounding spindle cell stroma. Initial differential diagnosis included a cutaneous eccrine adenolipoma variant and a cutaneous mixed solitary hamartoma composed of admixed neural, eccrine and lipomatous components. Immunohistochemical staining showed uptake for CD34 and CD10 in the spindle cell stromal component, but negative for desmin and S100. Based on our histopathological findings, our diagnosis of an eccrine cutaneous adenolipoma with a spindle cell component was rendered. This unique lesion displayed features of an evolving spindle cell lipoma with entrapped sweat glands, which may be viewed as an early stage of a dermal spindle cell adenolipoma. Conclusion Cutaneous eccrine adenolipoma is a rare, benign lipomatous neoplasm with entrapped sweat glands that can also show a spindle cell component. We share this rare lesion exemplifying the histomorphological spectrum of a lipomatous hamartoma and to highlight the recognition of a cutaneous adenolipoma.

Auditory information enhances post-sensory visual evidence during rapid multisensory decision-making

Despite recent progress in understanding multisensory decision-making, a conclusive mechanistic account of how the brain translates the relevant evidence into a decision is lacking. Specifically, it remains unclear whether perceptual improvements during rapid multisensory decisions are best explained by sensory (i.e., ‘Early’) processing benefits or post-sensory (i.e., ‘Late’) changes in decision dynamics. Here, we employ a well-established visual object categorisation task in which early sensory and post-sensory decision evidence can be dissociated using multivariate pattern analysis of the electroencephalogram (EEG). We capitalize on these distinct neural components to identify when and how complementary auditory information influences the encoding of decision-relevant visual evidence in a multisensory context. We show that it is primarily the post-sensory, rather than the early sensory, EEG component amplitudes that are being amplified during rapid audiovisual decision-making. Using a neurally informed drift diffusion model we demonstrate that a multisensory behavioral improvement in accuracy arises from an enhanced quality of the relevant decision evidence, as captured by the post-sensory EEG component, consistent with the emergence of multisensory evidence in higher-order brain areas.

Giant cell tumors of the sacrum: is non-operative treatment effective?

Giant cell tumors of sacrum in which surgery could endanger important neural components were treated with short term denosumab, angioembolisation and radiotherapy in different combinations to provide a non-operative function preserving treatment option. Between April 2013 and April 2017, 13 sacral GCTs [proximal extent of disease-S1 (10), S2 (2) and S3 (1)] were treated. Age ranged from 20 to 50years. One patient had loss of bladder control at presentation. Treatment protocol included short term denosumab, angioembolisation and radiotherapy in different combinations. Patients were evaluated every 10-12weeks. If disease ceased to progress no further treatment was advised. In case of progress, patient was advised additional denosumab and/or angioembolisation and/or radiotherapy till disease stopped progressing. 10 patients have non-progressive disease and are asymptomatic, 2 have non-progressive disease with occasional pain, 1 patient died. Follow-up duration (since final non-progression of disease) ranged from 15 to 54months (mean 31months). Total number of angio embolisation sessions ranged from 0 to 12 (mean = 4), total number of denosumab doses ranged from 5 to 16 (mean = 9). Five patients did not receive any radiotherapy, 5 received 50.4Gy and one patient each received 50.4 + 30 + 12Gy, 50.4 + 30Gy and 50.4 + 12Gy. The patient with loss of bladder control at presentation recovered. There were no other long-term complications. This study offers a non-surgical management option that provides good mid-term local control while preserving neurological function in these complex lesions.

Myelin Water Imaging Demonstrates Lower Brain Myelination in Children and Adolescents With Poor Reading Ability.

Magnetic resonance imaging (MRI) provides a means to non-invasively investigate the neurological links with dyslexia, a learning disability that affects one’s ability to read. Most previous brain MRI studies of dyslexia and reading skill have used structural or diffusion imaging to reveal regional brain abnormalities. However, volumetric and diffusion MRI lack specificity in their interpretation at the microstructural level. Myelin is a critical neural component for brain function and plasticity, and as such, deficits in myelin may impact reading ability. MRI can estimate myelin using myelin water fraction (MWF) imaging, which is based on evaluation of the proportion of short T2 myelin-associated water from multi-exponential T2 relaxation analysis, but has not yet been applied to the study of reading or dyslexia. In this study, MWF MRI, intelligence, and reading assessments were acquired in 20 participants aged 10–18 years with a wide range of reading ability to investigate the relationship between reading ability and myelination. Group comparisons showed markedly lower MWF by 16–69% in poor readers relative to good readers in the left and right thalamus, as well as the left posterior limb of the internal capsule, left/right anterior limb of the internal capsule, left/right centrum semiovale, and splenium of the corpus callosum. MWF over the entire group also correlated positively with three different reading scores in the bilateral thalamus as well as white matter, including the splenium of the corpus callosum, left posterior limb of the internal capsule, left anterior limb of the internal capsule, and left centrum semiovale. MWF imaging from T2 relaxation suggests that myelination, particularly in the bilateral thalamus, splenium, and left hemisphere white matter, plays a role in reading abilities. Myelin water imaging thus provides a potentially valuable in vivo imaging tool for the study of dyslexia and its remediation.

A perspective on C. elegans neurodevelopment: from early visionaries to a booming neuroscience research

The formation of the nervous system and its striking complexity is a remarkable feat of development. C. elegans served as a unique model to dissect the molecular events in neurodevelopment, from its early visionaries to the current booming neuroscience community. Soon after being introduced as a model, C. elegans was mapped at the level of genes, cells, and synapses, providing the first metazoan with a complete cell lineage, sequenced genome, and connectome. Here, I summarize mechanisms underlying C. elegans neurodevelopment, from the generation and diversification of neural components to their navigation and connectivity. I point out recent noteworthy findings in the fields of glia biology, sex dimorphism and plasticity in neurodevelopment, highlighting how current research connects back to the pioneering studies by Brenner, Sulston and colleagues. Multifaceted investigations in model organisms, connecting genes to cell function and behavior, expand our mechanistic understanding of neurodevelopment while allowing us to formulate emerging questions for future discoveries.

Inauguration in Development for Data Deduplication Under Neural Network Circumstances

The Neural network system is an educational paradigm that unites several neural networks to solve a problem. This paper explores the relationship between the ensemble and its networks of neural components, both from the viewpoint of regression and classification, which reveals that certain networks are stronger than other neural networks. This result is surprising because the rest of the neural networks enter the ensemble at present. To prove that a GASEN algorithm efficiently selects the appropriate neural networks to construct an ensemble from different neural networks available. At first several neuronal networks were taught by GASEN. Then the network allocates random weights and uses genetic algorithms to establish these weights to classify the fitness of the neural system in one ensemble to a certain degree. Ultimately, it used the weights designed for the ensemble for certain neural networks. A comprehensive analytical analysis reveals that, in comparison to typical assemblies, such as luggage, GASEN can generate network assemblies with much smaller sizes but with a higher generalization efficiency. This study, in addition, gives the mistake a gradual regression, demonstrating that the performance of GASEN could be that it can greatly reduce its bias and uncertainty such that GASEN is well aware of its operating mechanism.

Efficient phase coding in hippocampal place cells.

Neural codes have been postulated to build efficient representations of the external world. The hippocampus, an encoding system, employs neuronal firing rates and spike phases to encode external space. Although the biophysical origin of such codes is at a single neuronal level, the role of neural components in efficient coding is not understood. The complexity of this problem lies in the dimensionality of the parametric space encompassing neural components, and is amplified by the enormous biological heterogeneity observed in each parameter. A central question that spans encoding systems therefore is how neurons arrive at efficient codes in the face of widespread biological heterogeneities. To answer this, we developed a conductance-based spiking model for phase precession, a phase code of external space exhibited by hippocampal place cells. Our model accounted for several experimental observations on place cell firing and electrophysiology: the emergence of phase precession from exact spike timings of conductance-based models with neuron-specific ion channels and receptors; biological heterogeneities in neural components and excitability; the emergence of subthreshold voltage ramp, increased firing rate, enhanced theta power within the place field; a signature reduction in extracellular theta frequency compared to its intracellular counterpart; and experience-dependent asymmetry in firing-rate profile. We formulated phase-coding efficiency, using Shannon's information theory, as an information maximization problem with spike phase as the response and external space within a single place field as the stimulus. We employed an unbiased stochastic search spanning an 11-dimensional neural space, involving thousands of iterations that accounted for the biophysical richness and neuron-to-neuron heterogeneities. We found a small subset of models that exhibited efficient spatial information transfer through the phase code, and investigated the distinguishing features of this subpopulation at the parametric and functional scales. At the parametric scale, which spans the molecular components that defined the neuron, several nonunique parametric combinations with weak pairwise correlations yielded models with similar high phase-coding efficiency. Importantly, placing additional constraints on these models in terms of matching other aspects of hippocampal neural responses did not hamper parametric degeneracy. We provide quantitative evidence demonstrating this parametric degeneracy to be a consequence of a many-to-one relationship between the different parameters and phase-coding efficiency. At the functional scale, involving the cellular-scale neural properties, our analyses revealed an important higher-order constraint that was exclusive to models exhibiting efficient phase coding. Specifically, we found a counterbalancing negative correlation between neuronal gain and the strength of external synaptic inputs as a critical functional constraint for the emergence of efficient phase coding. These observations implicate intrinsic neural properties as important contributors in effectuating such counterbalance, which can be achieved by recruiting nonunique parametric combinations. Finally, we show that a change in afferent statistics, manifesting as input asymmetry onto these neuronal models, induced an adaptive shift in the phase code that preserved its efficiency. Together, our analyses unveil parametric degeneracy as a mechanism to harness widespread neuron-to-neuron heterogeneity towards accomplishing stable and efficient encoding, provided specific higher-order functional constraints on the relationship of neural gain to external inputs are satisfied.

Altered physiology of gastrointestinal vagal afferents following neurotrauma.

The adaptability of the central nervous system has been revealed in several model systems. Of particular interest to central nervous system-injured individuals is the ability for neural components to be modified for regain of function. In both types of neurotrauma, traumatic brain injury and spinal cord injury, the primary parasympathetic control to the gastrointestinal tract, the vagus nerve, remains anatomically intact. However, individuals with traumatic brain injury or spinal cord injury are highly susceptible to gastrointestinal dysfunctions. Such gastrointestinal dysfunctions attribute to higher morbidity and mortality following traumatic brain injury and spinal cord injury. While the vagal efferent output remains capable of eliciting motor responses following injury, evidence suggests impairment of the vagal afferents. Since sensory input drives motor output, this review will discuss the normal and altered anatomy and physiology of the gastrointestinal vagal afferents to better understand the contributions of vagal afferent plasticity following neurotrauma.

Deep Learning Model for Spatial Interpolation of Real-Time Seismic Intensity

Abstract Spatial distribution of seismic intensity plays an important role in emergency response during and immediately after an earthquake. In this study, we propose a deep learning model to predict the seismic intensity based on only the observation records at the seismic stations in a surrounding area. The deep learning model is trained using the observation records at both the input and target stations, and no geological information is used. Once the model is developed, for example, using the data from a temporal seismic array, the model can spatially interpolate the seismic intensity from the sparse layout of the seismic stations. The model consists of long short-term memory cells, which are well-established neural network components for time series analysis. We used observed seismograms in 1996 through 2019 at the Kyoshin Network (K-NET) and Kiban–Kyoshin Network (KiK-net) stations located in the northeastern part of Japan. In our deep learning model, approximately 85% of validation data is successfully classified into seismic intensity scales, which is better than adopting either the maximum or weighted average of the input data. We also apply the deep learning model to earthquake early warning (EEW). The model can predict the seismic intensity accurately and provides a long warning time. We concluded that our approach is a possible future solution for increasing the accuracy of EEW.