Structural insights into Ubr1-mediated N-degron polyubiquitination.

The N-degron pathway targets proteins that bear a destabilizing residue at the N terminus for proteasome-dependent degradation1. In yeast, Ubr1-a single-subunit E3 ligase-is responsible for the Arg/N-degron pathway2. How Ubr1 mediates the initiation of ubiquitination and the elongation of the ubiquitin chain in a linkage-specific manner through a single E2 ubiquitin-conjugating enzyme (Ubc2) remains unknown. Here we developed chemical strategies to mimic the reaction intermediates of the first and second ubiquitin transfer steps, and determined the cryo-electron microscopy structures of Ubr1 in complex with Ubc2, ubiquitin and two N-degron peptides, representing the initiation and elongation steps of ubiquitination. Key structural elements, including a Ubc2-binding region and an acceptor ubiquitin-binding loop on Ubr1, were identified and characterized. These structures provide mechanistic insights into the initiation and elongation of ubiquitination catalysed by Ubr1.

Enhancing cryo-EM structure prediction with DeepTracer and AlphaFold2 integration.

Understanding the protein structures is invaluable in various biomedical applications, such as vaccine development. Protein structure model building from experimental electron density maps is a time-consuming and labor-intensive task. To address the challenge, machine learning approaches have been proposed to automate this process. Currently, the majority of the experimental maps in the database lack atomic resolution features, making it challenging for machine learning-based methods to precisely determine protein structures from cryogenic electron microscopy density maps. On the other hand, protein structure prediction methods, such as AlphaFold2, leverage evolutionary information from protein sequences and have recently achieved groundbreaking accuracy. However, these methods often require manual refinement, which is labor intensive and time consuming. In this study, we present DeepTracer-Refine, an automated method that refines AlphaFold predicted structures by aligning them to DeepTracers modeled structure. Our method was evaluated on 39 multi-domain proteins and we improved the average residue coverage from 78.2 to 90.0% and average local Distance Difference Test score from 0.67 to 0.71. We also compared DeepTracer-Refine with Phenixs AlphaFold refinement and demonstrated that our method not only performs better when the initial AlphaFold model is less precise but also surpasses Phenix in run-time performance.

Fast and automated protein-DNA/RNA macromolecular complex modeling from cryo-EM maps.

Cryo-electron microscopy (cryo-EM) allows a macromolecular structure such as protein-DNA/RNA complexes to be reconstructed in a three-dimensional coulomb potential map. The structural information of these macromolecular complexes forms the foundation for understanding the molecular mechanism including many human diseases. However, the model building of large macromolecular complexes is often difficult and time-consuming. We recently developed DeepTracer-2.0, an artificial-intelligence-based pipeline that can build amino acid and nucleic acid backbones from a single cryo-EM map, and even predict the best-fitting residues according to the density of side chains. The experiments showed improved accuracy and efficiency when benchmarking the performance on independent experimental maps of protein-DNA/RNA complexes and demonstrated the promising future of macromolecular modeling from cryo-EM maps. Our method and pipeline could benefit researchers worldwide who work in molecular biomedicine and drug discovery, and substantially increase the throughput of the cryo-EM model building. The pipeline has been integrated into the web portal https://deeptracer.uw.edu/.

ComplexQA: a deep graph learning approach for protein complex structure assessment.

MOTIVATION: In recent years, the end-to-end deep learning method for single-chain protein structure prediction has achieved high accuracy. For example, the state-of-the-art method AlphaFold, developed by Google, has largely increased the accuracy of protein structure predictions to near experimental accuracy in some of the cases. At the same time, there are few methods that can evaluate the quality of protein complexes at the residue level. In particular, evaluating the quality of residues at the interface of protein complexes can lead to a wide range of applications, such as protein function analysis and drug design. In this paper, we introduce a new deep graph neural network-based method ComplexQA, to evaluate the local quality of interfaces for protein complexes by utilizing the residue-level structural information in 3D space and the sequence-level constraints. RESULTS: We benchmark our method to other state-of-the-art quality assessment approaches on the HAF2 and DBM55-AF2 datasets (high-quality structural models predicted by AlphaFold-Multimer), and the BM5 docking dataset. The experimental results show that our proposed method achieves better or similar performance compared with other state-of-the-art methods, especially on difficult targets which only contain a few acceptable models. Our method is able to suggest a score for each interfac e residue, which demonstrates a powerful assessment tool for the ever-increasing number of protein complexes. AVAILABILITY: https://github.com/Cao-Labs/ComplexQA.git. Contact: caora@plu.edu.

Cryo-EM model validation recommendations based on outcomes of the 2019 EMDataResource challenge.

This paper describes outcomes of the 2019 Cryo-EM Model Challenge. The goals were to (1) assess the quality of models that can be produced from cryogenic electron microscopy (cryo-EM) maps using current modeling software, (2) evaluate reproducibility of modeling results from different software developers and users and (3) compare performance of current metrics used for model evaluation, particularly Fit-to-Map metrics, with focus on near-atomic resolution. Our findings demonstrate the relatively high accuracy and reproducibility of cryo-EM models derived by 13 participating teams from four benchmark maps, including three forming a resolution series (1.8 to 3.1 Å). The results permit specific recommendations to be made about validating near-atomic cryo-EM structures both in the context of individual experiments and structure data archives such as the Protein Data Bank. We recommend the adoption of multiple scoring parameters to provide full and objective annotation and assessment of the model, reflective of the observed cryo-EM map density.

ZoomQA: residue-level protein model accuracy estimation with machine learning on sequential and 3D structural features.

MOTIVATION: The Estimation of Model Accuracy problem is a cornerstone problem in the field of Bioinformatics. As of CASP14, there are 79 global QA methods, and a minority of 39 residue-level QA methods with very few of them working on protein complexes. Here, we introduce ZoomQA, a novel, single-model method for assessing the accuracy of a tertiary protein structure/complex prediction at residue level, which have many applications such as drug discovery. ZoomQA differs from others by considering the change in chemical and physical features of a fragment structure (a portion of a protein within a radius $r$ of the target amino acid) as the radius of contact increases. Fourteen physical and chemical properties of amino acids are used to build a comprehensive representation of every residue within a protein and grade their placement within the protein as a whole. Moreover, we have shown the potential of ZoomQA to identify problematic regions of the SARS-CoV-2 protein complex. RESULTS: We benchmark ZoomQA on CASP14, and it outperforms other state-of-the-art local QA methods and rivals state of the art QA methods in global prediction metrics. Our experiment shows the efficacy of these new features and shows that our method is able to match the performance of other state-of-the-art methods without the use of homology searching against databases or PSSM matrices. AVAILABILITY: http://zoomQA.renzhitech.com.

DeepTracer for fast de novo cryo-EM protein structure modeling and special studies on CoV-related complexes.

Information about macromolecular structure of protein complexes and related cellular and molecular mechanisms can assist the search for vaccines and drug development processes. To obtain such structural information, we present DeepTracer, a fully automated deep learning-based method for fast de novo multichain protein complex structure determination from high-resolution cryoelectron microscopy (cryo-EM) maps. We applied DeepTracer on a previously published set of 476 raw experimental cryo-EM maps and compared the results with a current state of the art method. The residue coverage increased by over 30% using DeepTracer, and the rmsd value improved from 1.29 Å to 1.18 Å. Additionally, we applied DeepTracer on a set of 62 coronavirus-related cryo-EM maps, among them 10 with no deposited structure available in EMDataResource. We observed an average residue match of 84% with the deposited structures and an average rmsd of 0.93 Å. Additional tests with related methods further exemplify DeepTracer's competitive accuracy and efficiency of structure modeling. DeepTracer allows for exceptionally fast computations, making it possible to trace around 60,000 residues in 350 chains within only 2 h. The web service is globally accessible at https://deeptracer.uw.edu.

Single-cell characterization of self-renewing primary trophoblast organoids as modeling of EVT differentiation and interactions with decidual natural killer cells.

BACKGROUND: Extravillous trophoblast cell (EVT) differentiation and its communication with maternal decidua especially the leading immune cell type natural killer (NK) cell are critical events for placentation. However, appropriate in vitro modelling system and regulatory programs of these two events are still lacking. Recent trophoblast organoid (TO) has advanced the molecular and mechanistic research in placentation. Here, we firstly generated the self-renewing TO from human placental villous and differentiated it into EVTs (EVT-TO) for investigating the differentiation events. We then co-cultured EVT-TO with freshly isolated decidual NKs for further study of cell communication. TO modelling of EVT differentiation as well as EVT interaction with dNK might cast new aspect for placentation research. RESULTS: Single-cell RNA sequencing (scRNA-seq) was applied for comprehensive characterization and molecular exploration of TOs modelling of EVT differentiation and interaction with dNKs. Multiple distinct trophoblast states and dNK subpopulations were identified, representing CTB, STB, EVT, dNK1/2/3 and dNKp. Lineage trajectory and Seurat mapping analysis identified the close resemblance of TO and EVT-TO with the human placenta characteristic. Transcription factors regulatory network analysis revealed the cell-type specific essential TFs for controlling EVT differentiation. CellphoneDB analysis predicted the ligand-receptor complexes in dNK-EVT-TO co-cultures, which relate to cytokines, immunomodulation and angiogenesis. EVT was known to affect the immune properties of dNK. Our study found out that on the other way around, dNKs could exert effects on EVT causing expression changes which are functionally important. CONCLUSION: Our study documented a single-cell atlas for TO and its applications on EVT differentiation and communications with dNKs, and thus provide methodology and novel research cues for future study of human placentation.

Surface Lattice Modulation through Chemical Delithiation toward a Stable Nickel-Rich Layered Oxide Cathode.

Nickel-rich layered oxides (NLOs) are considered as one of the most promising cathode materials for next-generation high-energy lithium-ion batteries (LIBs), yet their practical applications are currently challenged by the unsatisfactory cyclability and reliability owing to their inherent interfacial and structural instability. Herein, we demonstrate an approach to reverse the unstable nature of NLOs through surface solid reaction, by which the reconstructed surface lattice turns stable and robust against both side reactions and chemophysical breakdown, resulting in improved cycling performance. Specifically, conformal La(OH)3 nanoshells are built with their thicknesses controlled at nanometer accuracy, which act as a Li+ capturer and induce controlled reaction with the NLO surface lattices, thereby transforming the particle crust into an epitaxial layer with localized Ni/Li disordering, where lithium deficiency and nickel stabilization are both achieved by transforming oxidative Ni3+ into stable Ni2+. An optimized balance between surface stabilization and charge transfer is demonstrated by a representative NLO material, namely, LiNi0.83Co0.07Mn0.1O2, whose surface engineering leads to a highly improved capacity retention and excellent rate capability with a strong capability to inhibit the crack of NLO particles. Our study highlights the importance of surface chemistry in determining chemical and structural behaviors and paves a research avenue in controlling the surface lattice for the stabilization of NLOs toward reliable high-energy LIBs.

JMJD6 protects against isoproterenol-induced cardiac hypertrophy via inhibition of NF-κB activation by demethylating R149 of the p65 subunit.

Histone modification plays an important role in pathological cardiac hypertrophy and heart failure. In this study we investigated the role of a histone arginine demethylase, Jumonji C domain-containing protein 6 (JMJD6) in pathological cardiac hypertrophy. Cardiac hypertrophy was induced in rats by subcutaneous injection of isoproterenol (ISO, 1.2 mg·kg-1·d-1) for a week. At the end of the experiment, the rats underwent echocardiography, followed by euthanasia and heart collection. We found that JMJD6 levels were compensatorily increased in ISO-induced hypertrophic cardiac tissues, but reduced in patients with heart failure with reduced ejection fraction (HFrEF). Furthermore, we demonstrated that JMJD6 overexpression significantly attenuated ISO-induced hypertrophy in neonatal rat cardiomyocytes (NRCMs) evidenced by the decreased cardiomyocyte surface area and hypertrophic genes expression. Cardiac-specific JMJD6 overexpression in rats protected the hearts against ISO-induced cardiac hypertrophy and fibrosis, and rescued cardiac function. Conversely, depletion of JMJD6 by single-guide RNA (sgRNA) exacerbated ISO-induced hypertrophic responses in NRCMs. We revealed that JMJD6 interacted with NF-κB p65 in cytoplasm and reduced nuclear levels of p65 under hypertrophic stimulation in vivo and in vitro. Mechanistically, JMJD6 bound to p65 and demethylated p65 at the R149 residue to inhibit the nuclear translocation of p65, thus inactivating NF-κB signaling and protecting against pathological cardiac hypertrophy. In addition, we found that JMJD6 demethylated histone H3R8, which might be a new histone substrate of JMJD6. These results suggest that JMJD6 may be a potential target for therapeutic interventions in cardiac hypertrophy and heart failure.

DeepTracer-ID: De novo protein identification from cryo-EM maps.

The recent revolution in cryo-electron microscopy (cryo-EM) has made it possible to determine macromolecular structures directly from cell extracts. However, identifying the correct protein from the cryo-EM map is still challenging and often needs additional sequence information from other techniques, such as tandem mass spectrometry and/or bioinformatics. Here, we present DeepTracer-ID, a server-based approach to identify the candidate protein in a user-provided organism de novo from a cryo-EM map, without the need for additional information. Our method first uses DeepTracer to generate a protein backbone model that best represents the cryo-EM map, and this model is then searched against the library of AlphaFold2 predictions for all proteins in the given organism. This method is highly accurate and robust for high-resolution cryo-EM maps: in all 13 experimental maps tested blindly, DeepTracer-ID identified the correct proteins as the top candidates. Eight of the maps were of known structures, while the other five unpublished maps were validated by prior protein annotation and careful inspection of the model refined into the map. The program also showed promising results for both homomeric and heteromeric protein complexes. This platform is possible because of the recent breakthroughs in large-scale three-dimensional protein structure prediction.

Neural representations of cryo-EM maps and a graph-based interpretation.

BACKGROUND: Advances in imagery at atomic and near-atomic resolution, such as cryogenic electron microscopy (cryo-EM), have led to an influx of high resolution images of proteins and other macromolecular structures to data banks worldwide. Producing a protein structure from the discrete voxel grid data of cryo-EM maps involves interpolation into the continuous spatial domain. We present a novel data format called the neural cryo-EM map, which is formed from a set of neural networks that accurately parameterize cryo-EM maps and provide native, spatially continuous data for density and gradient. As a case study of this data format, we create graph-based interpretations of high resolution experimental cryo-EM maps. RESULTS: Normalized cryo-EM map values interpolated using the non-linear neural cryo-EM format are more accurate, consistently scoring less than 0.01 mean absolute error, than a conventional tri-linear interpolation, which scores up to 0.12 mean absolute error. Our graph-based interpretations of 115 experimental cryo-EM maps from 1.15 to 4.0 Å resolution provide high coverage of the underlying amino acid residue locations, while accuracy of nodes is correlated with resolution. The nodes of graphs created from atomic resolution maps (higher than 1.6 Å) provide greater than 99% residue coverage as well as 85% full atomic coverage with a mean of 0.19 Å root mean squared deviation. Other graphs have a mean 84% residue coverage with less specificity of the nodes due to experimental noise and differences of density context at lower resolutions. CONCLUSIONS: The fully continuous and differentiable nature of the neural cryo-EM map enables the adaptation of the voxel data to alternative data formats, such as a graph that characterizes the atomic locations of the underlying protein or macromolecular structure. Graphs created from atomic resolution maps are superior in finding atom locations and may serve as input to predictive residue classification and structure segmentation methods. This work may be generalized to transform any 3D grid-based data format into non-linear, continuous, and differentiable format for downstream geometric deep learning applications.

Modulation of DNA conformation in electrolytic nanodroplets.

The behavior of deoxyribonucleic acid (DNA) molecules in confinement is of profound importance in various bioengineering and medical applications. In the present study, all-atom molecular dynamics simulation is utilized to investigate the transition of the double-strand DNA (dsDNA) conformation in the electrolytic nanodroplet. Three typical conformations, i.e., C-shaped, folded S-shaped, and double C-shaped, are observed for different droplet sizes and ionic concentrations. To reveal the physics underlying this phenomenon, the characteristics of the dsDNA molecules, such as the overcharging intensity, the end-to-end distance, the radius of gyration, etc. are analyzed in detail based on the numerical results. It is found that the transition can be ascribed to the buckling of the polymer molecules under the compression due to the confinement of the nanodroplet, and it can be modulated by the ionic concentration in the electrolyte. Generally, nanoscale confinement dominates dsDNA behavior over the electrostatic effects in smaller nanodroplets, while the latter becomes more important for larger nanodroplets. This competition results in the persistence length increasing with the nanodroplet radii. Based on these discussions, a non-dimensional elasto-capillary number µ is proposed to classify the dsDNA conformations into three regions.

Joint Detection of Community and Structural Hole Spanner of Networks in Hyperbolic Space.

Community detection and structural hole spanner (the node bridging different communities) identification, revealing the mesoscopic and microscopic structural properties of complex networks, have drawn much attention in recent years. As the determinant of mesoscopic structure, communities and structural hole spanners discover the clustering and hierarchy of networks, which has a key impact on transmission phenomena such as epidemic transmission, information diffusion, etc. However, most existing studies address the two tasks independently, which ignores the structural correlation between mesoscale and microscale and suffers from high computational costs. In this article, we propose an algorithm for simultaneously detecting communities and structural hole spanners via hyperbolic embedding (SDHE). Specifically, we first embed networks into a hyperbolic plane, in which, the angular distribution of the nodes reveals community structures of the embedded network. Then, we analyze the critical gap to detect communities and the angular region where structural hole spanners may exist. Finally, we identify structural hole spanners via two-step connectivity. Experimental results on synthetic networks and real networks demonstrate the effectiveness of our proposed algorithm compared with several state-of-the-art methods.

Missed rib fractures on initial chest CT in trauma patients: time patterns, clinical and forensic significance.

OBJECTIVE: To analyze missed rib fractures and proper time for evaluation on CT at different ages and to determine factors that favor missed fractures. METHODS: One hundred patients with rib fractures who underwent CT were classified into three groups according to their age: young, middle-aged, and elderly. CT was performed within 1 to 6 weeks after trauma. The imaging features and temporal changes of rib fractures were analyzed. RESULTS: At the first CT during the initial week, 638 ribs were detected with one or several fractures, overall 838 fractures were confirmed, and 6 were suspected. In the next 2-6 weeks, 47 occult rib fractures were additionally detected. The number of additionally diagnosed fractures was the highest in respectively the 3rd week among younger, 4th week in the middle-aged, and 6th week in the elderly groups. The detection of occult rib fractures was significantly delayed in the middle-aged and elderly groups compared with the young group (p < 0.05). The time to form bony callus was also significantly (p < 0.05) delayed with age, with significantly (p < 0.05) more time needed to form bony callus in the middle-aged (23.8 ± 4.5 days) and elderly (28.48 ± 5.1 days) groups than in the young group (18.0 ± 2.2 days). CONCLUSIONS: Most rib fractures can be detected within the first week after trauma. Detection of occult rib fractures will be delayed with increase of age, and repeated CT scanning should be appropriately postponed in patients at different ages. Trabecula, inner and outer plates, costal angle, and cartilage are the primary locations for occult and subtle fractures which should be carefully evaluated. KEY POINTS: • More rib fractures can be detected on repeated CT scans, especially for subtle and occult rib fractures. • Detection of all rib fractures helps relieve the patient's concerns and determine the degree of personal injury for appropriate evaluation.

Consistency of spatial dynamics of HIV-1 and HCV among HIV-1/HCV coinfected drug users in China.

BACKGROUND: As the transmission routes of human immunodeficiency virus type 1 (HIV-1) and hepatitis C virus (HCV) are similar, previous studies based on separate research on HIV-1 and HCV assumed a similar transmission pattern. However, few studies have focused on the possible correlation of the spatial dynamics of HIV-1 and HCV among HIV-1/HCV coinfected patients. METHODS: A total of 310 HIV-1/HCV coinfected drug users were recruited in Yingjiang and Kaiyuan prefectures, Yunnan Province, China. HIV-1 env, p17, pol and HCV C/E2, NS5B fragments were amplified and sequenced from serum samples. The genetic characteristics and spatial dynamics of HIV-1 and HCV were explored by phylogenetic, bootscanning, and phylogeographic analyses. RESULTS: Among HIV-1/HCV coinfected drug users, eight HCV subtypes (1a, 1b, 3a, 3b, 6a, 6n, 6v, and 6u) and two HIV-1 subtypes (subtype B and subtype C), three HIV-1 circulating recombinant forms (CRF01_AE, CRF07_BC and CRF08_BC), and four unique recombinant forms (URF_BC, URF_01B, URF_01C and URF_01BC) were identified. HCV subtype 3b was the most predominant subtype in both Yingjiang and Kaiyuan prefectures. The dominant circulating HIV-1 subtypes for drug users among the two areas were CRF08_BC and URF_BC. Maximum clade credibility trees revealed that both HIV-1 and HCV were transmitted from Yingjiang to Kaiyuan. CONCLUSIONS: The spatial dynamics of HIV-1 and HCV among HIV-1/HCV coinfected drug users seem to have high consistency, providing theoretical evidence for the prevention of HIV-1 and HCV simultaneously.

Non-volatile acylphloroglucinol components from Eucalyptus robusta inhibit Zika virus by impairing RdRp activity of NS5.

Eucalyptus is a large genus of the Myrtaceae family with high value in various fields of industry. Recently, attention has been focused on the functional properties of Eucalyptus extracts. These extracts have been traditionally used to combat various infectious diseases, and volatile oils are usually considered to play a major role. But the positive effects of non-volatile acylphloroglucinols, a class of specialized metabolites with relatively high content in Eucalyptus, should not be neglected. Herein, non-volatile acylphloroglucinols from leaves of Eucalyptus robusta were evaluated for their abilities to inhibit Zika virus (ZIKV) which is associated with severe neurological damage and complications. The results showed eucalyprobusone G, a new symmetrical acylphloroglucinol dimer, possessed the significant ability to inhibit ZIKV without inducing cytotoxicity. The EC50 values of eucalyprobusone G against the African lineage (MR766) and Asian lineage (SZ-WIV01) of ZIKV were 0.43 ± 0.08 and 10.10 ± 3.84 µM which were 110 times and 5.8 times better than those of the reference compound ribavirin, respectively. Further action mode research showed that eucalyprobusone G impairs the viral binding and RdRp activity of NS5. The results broaden the functional properties of Eucalyptus robusta and indicate acylphloroglucinol dimers could be developed as anti-ZIKV agents.

MicroRNA-214 contributes to Ang II-induced cardiac hypertrophy by targeting SIRT3 to provoke mitochondrial malfunction.

Reduction of expression and activity of sirtuin 3 (SIRT3) contributes to the pathogenesis of cardiomyopathy via inducing mitochondrial injury and energy metabolism disorder. However, development of effective ways and agents to modulate SIRT3 remains a big challenge. In this study we explored the upstream suppressor of SIRT3 in angiotensin II (Ang II)-induced cardiac hypertrophy in mice. We first found that SIRT3 deficiency exacerbated Ang II-induced cardiac hypertrophy, and resulted in the development of spontaneous heart failure. Since miRNAs play crucial roles in the pathogenesis of cardiac hypertrophy, we performed miRNA sequencing on myocardium tissues from Ang II-infused Sirt3-/- and wild type mice, and identified microRNA-214 (miR-214) was significantly up-regulated in Ang II-infused mice. Similar results were also obtained in Ang II-treated neonatal mouse cardiomyocytes (NMCMs). Using dual-luciferase reporter assay we demonstrated that SIRT3 was a direct target of miR-214. Overexpression of miR-214 in vitro and in vivo decreased the expression of SIRT3, which resulted in extensive mitochondrial damages, thereby facilitating the onset of hypertrophy. In contrast, knockdown of miR-214 counteracted Ang II-induced detrimental effects via restoring SIRT3, and ameliorated mitochondrial morphology and respiratory activity. Collectively, these results demonstrate that miR-214 participates in Ang II-induced cardiac hypertrophy by directly suppressing SIRT3, and subsequently leading to mitochondrial malfunction, suggesting the potential of miR-214 as a promising intervention target for antihypertrophic therapy.

Erratum: 17ß-Estradiol Inhibites Tumor Necrosis Factor-α Induced Apoptosis of Human Nucleus Pulposus Cells via the PI3K/Akt Pathway.

The authors asked for the change of the figure 1. They wanted to send the figure that was 200 times bigger under the microscope as described in Figure 1 caption, however, they mistakenly uploaded the wrong picture. Reference: 1. Tao Wang, Si-Dong Yang, Sen Liu, Hui Wang, Huan Liu, Wen Yuan Ding: 17ß-Estradiol Inhibites Tumor Necrosis Factor-alpha Induced Apoptosis of Human Nucleus Pulposus Cells via the PI3K/Akt Pathway. Med Sci Monit, 2016; 22: 4312-4322. DOI: 10.12659/MSM.900310.

A Raspberry Pi-Based Traumatic Brain Injury Detection System for Single-Channel Electroencephalogram.

Traumatic Brain Injury (TBI) is a common cause of death and disability. However, existing tools for TBI diagnosis are either subjective or require extensive clinical setup and expertise. The increasing affordability and reduction in the size of relatively high-performance computing systems combined with promising results from TBI related machine learning research make it possible to create compact and portable systems for early detection of TBI. This work describes a Raspberry Pi based portable, real-time data acquisition, and automated processing system that uses machine learning to efficiently identify TBI and automatically score sleep stages from a single-channel Electroencephalogram (EEG) signal. We discuss the design, implementation, and verification of the system that can digitize the EEG signal using an Analog to Digital Converter (ADC) and perform real-time signal classification to detect the presence of mild TBI (mTBI). We utilize Convolutional Neural Networks (CNN) and XGBoost based predictive models to evaluate the performance and demonstrate the versatility of the system to operate with multiple types of predictive models. We achieve a peak classification accuracy of more than 90% with a classification time of less than 1 s across 16-64 s epochs for TBI vs. control conditions. This work can enable the development of systems suitable for field use without requiring specialized medical equipment for early TBI detection applications and TBI research. Further, this work opens avenues to implement connected, real-time TBI related health and wellness monitoring systems.