NeuralRecon: Real-Time Coherent 3D Scene Reconstruction from Monocular Video.

We present a novel framework named NeuralRecon for real-time 3D scene reconstruction from a monocular video. Unlike previous methods that estimate single-view depth maps separately on each key-frame and fuse them later, we propose to directly reconstruct local surfaces represented as sparse TSDF volumes for each video fragment sequentially by a neural network. A learning-based TSDF fusion module based on gated recurrent units is used to guide the network to fuse features from previous fragments. This design allows the network to capture local smoothness prior and global shape prior of 3D surfaces when sequentially reconstructing the surfaces, resulting in accurate, coherent, and real-time surface reconstruction. The fused features can also be used to predict semantic labels, allowing our method to reconstruct and segment the 3D scene simultaneously. Furthermore, we purpose an efficient self-supervised fine-tuning scheme that refines scene geometry based on input images through differentiable volume rendering. This fine-tuning scheme improves reconstruction quality on the fine-tuned scenes as well as the generalization to similar test scenes. The experiments on ScanNet, 7-Scenes and Replica datasets show that our system outperforms state-of-the-art methods in terms of both accuracy and speed.

Associations between residential greenness and obesity phenotypes among adults in Southwest China.

BACKGROUND: Although exposure to greenness has generally benefited human metabolic health, the association between greenness exposure and metabolic obesity remains poorly studied. We aimed to investigate the associations between residential greenness and obesity phenotypes and the mediation effects of air pollutants and physical activity (PA) level on the associations. METHODS: We used the baseline of the China Multi-Ethnic Cohort (CMEC) study, which enrolled 87,613 adults. Obesity phenotypes were defined based on obesity and metabolic status, including metabolically unhealthy obesity (MUO), non-obesity (MUNO), metabolically healthy obesity (MHO), and non-obesity (MHNO). Greenness exposure was measured as the 3-year mean values of the normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) within the 500-m buffer zones around the participants' residence. Multivariable logistic regression was used to estimate the associations between greenness and obesity phenotypes. Stratified analyses by age, sex, educational level, and urbanicity were performed to identify how the effect varies across different subgroups. Causal mediation analysis was used to examine the mediation effects of air pollutants and PA level. RESULTS: Compared with MHNO, each interquartile range (IQR) increase in greenness exposure was associated with reduced risks of MHO (ORNDVI [95% CI] = 0.87 [0.81, 0.93]; OREVI = 0.91 [0.86, 0.97]), MUO (ORNDVI = 0.83 [0.78, 0.88]; OREVI = 0.86 [0.81, 0.91]), and MUNO (ORNDVI = 0.88 [0.84, 0.91]; OREVI = 0.89 [0.86, 0.92]). For each IQR increase in both NDVI and EVI, the risks of MHO, MUO, and MUNO were reduced more in men, participants over 60 years, those with a higher level of education, and those living in urban areas, compared to their counterparts. Concentrations of particulate matter (PM) and PA level partially mediated the associations between greenness exposure and obesity phenotypes. CONCLUSIONS: Exposure to residential greenness was associated with decreased risks of MHO, MUO, and MUNO, which was mediated by concentrations of PM and PA level, and modified by sex, age, educational level, and urbanicity.

Poly(3-hexylthiophene)/perovskite Heterointerface by Spinodal Decomposition Enabling Efficient and Stable Perovskite Solar Cells.

The best research-cell efficiency of perovskite solar cells (PSCs) is comparable with that of mature silicon solar cells (SSCs); However, the industrial development of PSCs lags far behind SSCs. PSC is a multiphase and multicomponent system, whose consequent interfacial energy loss and carrier loss seriously affect the performance and stability of devices. Here, by using spinodal decomposition, a spontaneous solid phase segregation process, in situ introduces a poly(3-hexylthiophene)/perovskite (P3HT/PVK) heterointerface with interpenetrating structure in PSCs. The P3HT/PVK heterointerface tunes the energy alignment, thereby reducing the energy loss at the interface; The P3HT/PVK interpenetrating structure bridges a transport channel, thus decreasing the carrier loss at the interface. The simultaneous mitigation of energy and carrier losses by P3HT/PVK heterointerface enables n-i-p geometry device a power conversion efficiency of 24.53% (certified 23.94%) and excellent stability. These findings demonstrate an ingenious strategy to optimize the performance of PSCs by heterointerface via Spinodal decomposition.

Potassium-Based Dual-Ion Batteries Operating at -60 °C Enabled By Co-Intercalation Anode Chemistry.

Battery performance at subzero is restricted by sluggish interfacial kinetics. To resolve this issue, potassium-based dual-ion batteries (K-DIBs) based on the polytriphenylamine (PTPAn) cathode with anion storage chemistry and the hydrogen titanate (HTO) anode with K+ /solvent co-intercalation mechanism are constructed. Both the PTPAn cathode and the HTO anode do not undergo the desolvation process, which can effectively accelerate the interfacial kinetics at subzero. As revealed by theoretical calculations and experimental analysis, the strong K+ /solvent binding energy in the dilute electrolyte, the charge shielding effect of the crystal water, and the uniform SEI layer with high content of the flexible organic species synergically promote HTO to undergo K+ /solvent co-intercalation behavior. The special co-intercalation mechanism and anion storage chemistry enable HTO||PTPAn K-DIBs with superior rate performance and cycle durability, maintaining a capacity retention of 94.1% after 6000 cycles at -40 °C and 91% after 1000 cycles at -60 °C. These results provide a step forward for achieving high-performance energy storage devices at low temperatures.

Biochar Decreases Cr Toxicity and Accumulation in Sunflower Grown in Cr(VI)-Polluted Soil.

Biochar is preferentially recommended for the remediation of heavy metal-polluted soils. Sunflower is an important high-biomass oil crop with a promising potential for phytoremediation of Cr(VI)-polluted soil. However, how biochar affects sunflower growth and Cr accumulation in Cr(VI)-polluted soil needs to be elucidated. Here, a pot culture experiment was conducted to study whether soil amendment with biochar (0, 0.1%, 1%, and 5%, w/w) can mitigate Cr toxicity and accumulation in sunflower seedlings grown in soils artificially polluted with different levels of Cr(VI) (0, 50, and 250 mg Cr(VI)/kg soil). The addition of Cr(VI) exhibited significant phytotoxicity, as evidenced by inhibited plant growth and even the death of seedlings at 250 mg/kg Cr(VI). Overall, biochar amendment showed positive effects on plant growth and Cr immobilization, dependent on both the biochar dose and Cr addition level. When 50 mg/kg Cr(VI) was added, 1% biochar showed positive effects similar to 5% biochar on improving plant growth and mineral nutrition (particularly K), reducing Cr content in shoots and roots, and decreasing Cr availability and Cr(VI) content in the soil. In comparison with non-amendment, 1% and 5% biochar caused 85% and 100% increase in shoot dry weights, and 75% and 86% reduction in shoot Cr concentrations, respectively. When 250 mg/kg Cr(VI) was added, a 5% dose produced much better benefits than 1%, while a 0.1% dose did not help plants to survive. Overall, an appropriate dose of biochar enhanced Cr(VI) immobilization and subsequently decreased its toxicity and accumulation in sunflower seedlings. Our findings confirm that biochar can be used as an efficient amendment for the remediation of Cr(VI)-polluted soils and cleaner production of sunflower oil and biomass.

DNA Origami: From Molecular Folding Art to Drug Delivery Technology.

DNA molecules that store genetic information in living creatures can be repurposed as building blocks to construct artificial architectures, ranging from the nanoscale to the microscale. The precise fabrication of self-assembled DNA nanomaterials and their various applications have greatly impacted nanoscience and nanotechnology. More specifically, the DNA origami technique has realized the assembly of various nanostructures featuring rationally predesigned geometries, precise addressability, and versatile programmability, as well as remarkable biocompatibility. These features have elevated DNA origami from academic interest to an emerging class of drug delivery platform for a wide range of diseases. In this minireview, the latest advances in the burgeoning field of DNA-origami-based innovative platforms for regulating biological functions and delivering versatile drugs are presented. Challenges regarding the novel drug vehicle's safety, stability, targeting strategy, and future clinical translation are also discussed.

Single Incision non-thoracoscopic Nuss procedure for children with pectus excavatum: protocol for a multicenter, non-masked, randomized controlled trial.

Background: Nuss procedure is the most common method of surgical treatment to pectus excavatum (PE). A significant percentage of surgeons choose to use thoracoscopic assistance during the Nuss procedure (TNP) to avoid cardiac injury. However, our previous findings confirm the safety of single incision Non-thoracoscopic Nuss Procedure (SINTNP). Hence, Further studies, particularly prospective randomized controlled trials, are necessary to assess the value of SINTNP for PE. Methods: This study is a prospective, superiority, multicenter, non-masked, randomized controlled trial that investigates the outcome and hospitalization medical expense of SINTNP compared to TNP for PE. A total of 320 eligible patients according to sample size calculation by retrospective data will be randomly assigned to the SINTNP group or the TNP group at a 1:1 ratio using stratified blocked randomization and the zone length was set as four. Patients aged between 3 and 18 years old for the first surgery and without combination of complex anomalies such as Marfan syndrome and congenital heart disease will be considered for the study. The co-primary endpoint is thoracic related complications and medical expense during hospitalization. Thoracic related complications were defined as pneumothorax, pleural effusion, pneumonia and incision infection. The secondary endpoints include surgery duration and length of hospital stay.The registration number for this study protocol is ChiCTR230073081 (Chinese Clinical Trial Registry, A Primary Registry of International Clinical Trial Registry Platform, World Health Organization).

Efficient photocatalytic hydrogen production by space separation of photo-generated charges from S-scheme ZnIn2S4/ZnO heterojunction.

ZnIn2S4/ZnO heterostructures have been achieved by a simple in-situ growth solvothermal method. Under full spectrum irradiation, the optimal photocatalyst 2ZnIn2S4/ZnO exhibits H2 evolution rate of 13,638 (water/ethanol = 1:1) and 3036 (water) µmol·g-1h-1, which is respectively 4 and 5 times higher than that of pure ZnIn2S4. In situ illumination X-ray photoelectron spectroscopy (ISI-XPS) analysis and density functional theory (DFT) calculations show that the electrons of ZnIn2S4 are removed to ZnO through hybridization and form an internal electric field between ZnIn2S4 and ZnO. The optical properties of the catalyst and the effect of internal electric field (IEF) can increase photo-generated electrons (e-)-holes (h+) transport rate and enhance light collection, resulting in profitable photocatalytic properties. The photoelectrochemical and EPR results show that a stepped (S-scheme) heterojunction is formed in the ZnIn2S4/ZnO redox center, which greatly promotes separation of e--h+ pairs and efficient H2 evolution. This research offers an effective method for constructing an efficient S-Scheme photocatalytic system for H2 evolution.

Dual DNAzyme-catalytic assembly of G-quadruplexes for inducing the aggregation of gold nanoparticles and developing a novel antibiotic assay method.

By utilizing a target biorecognition reaction to induce the self-assembly of G-quadruplexes and the aggregation of gold nanoparticles (Au NPs), this work develops a novel colorimetric biosensing method for kanamycin (Kana) antibiotic detection. The compact G-quadruplex structure was assembled from its two half-split sequences which were designed in two hairpin substrates of the Mg2+-dependent DNAzyme (MNAzyme). Besides hybridizing with the aptamer strand, the MNAzyme sequence was also split into two half fragments to be designed in the two substrates. Upon the aptamer-recognition reaction toward Kana, the MNAzyme strand could be quantitatively released to cause the exposure of the split G-quadruplex-sequences on two hairpin substrate-modified Au NPs and simultaneous release of two half fragments of the MNAzyme-sequence. Thus, the K+-assisted self-folding of G-quadruplexes causes the cross-linking of the two Au NPs to realize the Au NP aggregation-based colorimetric signal output (measured at the largest absorption peak near 520 nm). Meanwhile, the self-assembled formation of the second MNAzyme drastically amplified the signal response. Under the optimal conditions, a wide linear range from 0.1 pg mL-1 to 10 ng mL-1 and an ultrahigh sensitivity with the detection limit of 76 fg mL-1 were obtained. The dose-recovery experiments in real samples showed satisfactory results with recoveries from 98.4 to 105.4% and relative errors compared with the ELISA method less than 4.1%. Due to the high selectivity, excellent repeatability and stability, and simple manipulation, this method indicates a promising potential for practical applications. A novel homogeneous biosensing method was developed for the convenient detection of the kanamycin antibiotic. The target biorecognition-induced and dual DNAzyme-catalytic assembly of G-quadruplexes enabled the amplified aggregation of gold nanoparticles for the simple, cheap, stable, and ultrasensitive colorimetric signal transduction of the method.

Target biorecognition-triggered assembly of a G-quadruplex DNAzyme-decorated nanotree for the convenient and ultrasensitive detection of antibiotic residues.

The abuse of kanamycin (Kana) in many fields has led to increasing antibiotic pollution problems and serious threats to public health. Therefore, determining how to develop methods to realize the convenient detection of antibiotics in complicated environmental matrices is highly desirable. In this study, we utilized a target biorecognition-triggered hybridization chain reaction (HCR) assembly of a G-quadruplex DNAzyme (G-DNAzyme)-decorated nanotree to develop a novel homogeneous colorimetric biosensing method for the convenient and ultrasensitive detection of Kana antibiotic residues in real samples. Through the designed aptamer-recognition reaction, an Mg2+-dependent DNAzyme (MNAzyme) strand can be liberated. Thus, its catalyzed cleavage of the hairpin substrates anchored at a DNA nanowire will cause the assembled formation of an HCR-initiator; this process can be greatly amplified by the exonuclease III-assisted target recycling and the MNAzyme-catalyzed release of another MNAzyme strand. Based on the DNA-nanowire-accelerated HCR assembly of many G-DNAzyme-decorated DNA duplexes on the two sides of the nanowire, a DNA nanotree decorated by numerous G-DNAzymes will form to realize the ultrasensitive colorimetric signal output. Under the optimal conditions, this method exhibited a wide five-order-of-magnitude linear range and a very low detection limit of 28 fg mL-1. In addition, excellent selectivity, repeatability, and reliability were also demonstrated for this homogeneous bioassay method. These unique features along with its automatic manipulation and low assay cost show promise for practical applications.

Dual CHA-mediated high-efficient formation of a tripedal DNA walker for constructing a novel proteinase-free dual-mode biosensing strategy.

DNA walkers have been recognized as a type of powerful signal amplification tool for biosensors, but how to adopt a proper strategy to increase their amplification efficiency is still highly desirable. Herein we design a dual-catalytic hairpin assembly (CHA)-mediated strategy for the high-efficient formation of a tripedal Mg2+-dependent DNAzyme (MNAzyme)-DNA walker, and thus develop a novel proteinase-free dual-mode biosensing method for the kanamycin (Kana) antibiotic assay. The first CHA is initiated by a target-biorecognition reaction, which can produce the DNA walker and also induce the target recycling. The second CHA is initiated by a special base sequence designed as a one-half substrate of the MNAzyme. Upon the first CHA-triggered DNA walking at a magnetic bead (MB) track, this "pseudo-target" sequence can be released to induce another CHA-cycle for the formation of the same DNA walker. Meanwhile, the other one-half substrate strand exposed on the MB surface will trigger the quantitative hybridization chain reaction (HCR)-assembly of a G-quadruplex DNAzyme (G-DNAzyme)-enriched double-stranded DNA polymer. So the enzymatic reaction of G-DNAzymes enabled the convenient colorimetric and photoelectrochemical dual-mode signal transduction of the method. Due to the dual-CHA facilitation to the tripedal and three-dimensional DNA walking and synergetic signal amplification of HCR, this method exhibits very low detection limits of 9.4 and 0.55 fg mL-1, respectively. In combination with its wide linear range, automated manipulation, and excellent selectivity, repeatability and reliability, the proposed method is expected to be used for the convenient semiquantitative screening and accurate determination of possible antibiotic residues in complicated matrices.

Shape Prior Guided Instance Disparity Estimation for 3D Object Detection.

In this paper, we propose a novel system named Disp R-CNN for 3D object detection from stereo images. Many recent works solve this problem by first recovering point clouds with disparity estimation and then apply a 3D detector. The disparity map is computed for the entire image, which is costly and fails to leverage category-specific prior. In contrast, we design an instance disparity estimation network (iDispNet) that predicts disparity only for pixels on objects of interest and learns a category-specific shape prior for more accurate disparity estimation. To address the challenge from scarcity of disparity annotation in training, we propose to use a statistical shape model to generate dense disparity pseudo-ground-truth without the need of LiDAR point clouds, which makes our system more widely applicable. Experiments on the KITTI dataset show that, when LiDAR ground-truth is not used at training time, Disp R-CNN outperforms previous state-of-the-art methods based on stereo input by 20 percent in terms of average precision for all categories. The code and pseudo-ground-truth data are available at the project page: https://github.com/zju3dv/disprcnn.

Dietary Approaches to Stop Hypertension (DASH) diet, Mediterranean diet and blood lipid profiles in less-developed ethnic minority regions.

Different from developed countries, there is a paucity of research examining how the Dietary Approaches to Stop Hypertension (DASH) and Mediterranean diets relate to lipids in less-developed ethnic minority regions (LEMR). A total of 83 081 participants from seven ethnic groups were retrieved from the baseline data of the China Multi-Ethnic Cohort study, which was conducted in less-developed Southwest China between May 2018 and September 2019. Multivariable linear regression models were then used to examine the associations of the DASH and alternative Mediterranean diet (AMED) scores, assessed by modified DASH score and AMED, as well as their components with total cholesterol (TC), LDL-cholesterol, HDL-cholesterol, TAG and TC/HDL-cholesterol. The DASH scores were negatively associated with TC, HDL-cholesterol and TAG. Comparing the highest quintiles with the lowest DASH scores, TC decreased 0·0708 (95 % CI -0·0923, -0·0493) mmol/l, HDL-cholesterol decreased 0·0380 (95 % CI -0·0462, -0·0299) mmol/l and TAG decreased 0·0668 (95 % CI -0·0994, -0·0341) mmol/l. The AMED scores were negatively associated with TC, LDL-cholesterol and HDL-cholesterol. Comparing the highest quintiles with the lowest AMED scores, TC decreased 0·0816 (95 % CI -0·1035, -0·0597) mmol/l, LDL-cholesterol decreased 0·0297 (95 % CI -0·0477, -0·0118) mmol/l and HDL-cholesterol decreased 0·0275 (95 % CI -0·0358, -0·0192) mmol/l. Although both the DASH diet and the Mediterranean diet were negatively associated with blood lipids, those associations showed different patterns in LEMR, particularly for TAG and HDL-cholesterol.

Longitudinal clonal tracking in humanized mice reveals sustained polyclonal repopulation of gene-modified human-HSPC despite vector integration bias.

BACKGROUND: Current understanding of hematopoiesis is largely derived from mouse models that are physiologically distant from humans. Humanized mice provide the most physiologically relevant small animal model to study human diseases, most notably preclinical gene therapy studies. However, the clonal repopulation dynamics of human hematopoietic stem and progenitor cells (HSPC) in these animal models is only partially understood. Using a new clonal tracking methodology designed for small sample volumes, we aim to reveal the underlying clonal dynamics of human cell repopulation in a mouse environment. METHODS: Humanized bone marrow-liver-thymus (hu-BLT) mice were generated by transplanting lentiviral vector-transduced human fetal liver HSPC (FL-HSPC) in NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice implanted with a piece of human fetal thymus. We developed a methodology to track vector integration sites (VIS) in a mere 25 µl of mouse blood for longitudinal and quantitative clonal analysis of human HSPC repopulation in mouse environment. We explored transcriptional and epigenetic features of human HSPC for possible VIS bias. RESULTS: A total of 897 HSPC clones were longitudinally tracked in hu-BLT mice-providing a first-ever demonstration of clonal dynamics and coordinated expansion of therapeutic and control vector-modified human cell populations simultaneously repopulating in the same humanized mice. The polyclonal repopulation stabilized at 19 weeks post-transplant and the contribution of the largest clone doubled within 4 weeks. Moreover, 550 (~ 60%) clones persisted over 6 weeks and were highly shared between different organs. The normal clonal profiles confirmed the safety of our gene therapy vectors. Multi-omics analysis of human FL-HSPC revealed that 54% of vector integrations in repopulating clones occurred within ± 1 kb of H3K36me3-enriched regions. CONCLUSIONS: Human repopulation in mice is polyclonal and stabilizes more rapidly than that previously observed in humans. VIS preference for H3K36me3 has no apparent negative effects on HSPC repopulation. Our study provides a methodology to longitudinally track clonal repopulation in small animal models extensively used for stem cell and gene therapy research and with lentiviral vectors designed for clinical applications. Results of this study provide a framework for understanding the clonal behavior of human HPSC repopulating in a mouse environment, critical for translating results from humanized mice models to the human settings.

Exo III-Catalyzed Release of a Zn2+-Ligation DNAzyme to Drive the Strand Displacement Reaction and Gold Aggregation for the Homogeneous Bioassay of Kanamycin Antibiotics.

Herein, we combine the exonuclease III (Exo III)-catalyzed release of a Zn2+-dependent ligation DNAzyme with the DNAzyme-driven strand displacement reaction (SDR) to develop a novel homogeneous colorimetric bioassay method for kanamycin (Kana) antibiotic detection. Upon the biorecognition reaction between Kana and a designed hairpin DNA, the DNAzyme-containing strand can be catalytically released by Exo III. Then, this DNAzyme will catalyze the ligation of two oligonucleotides to cause a SDR and the aggregation of gold nanoparticles (Au NPs) labeled by two linker DNA strands. Due to the aggregation of Au NPs for colorimetric signal transduction and the Exo III and SDR-assisted dual signal amplification, this method shows a wide linear range of 5 orders of magnitude and a very low detection limit down to 8.1 fg mL-1. Together with its excellent selectivity, repeatability, reliability, and convenient manipulation, the proposed method shows a great potential for the food quality monitoring application.

Design and Solvothermal Synthesis of Polyoxometalate-Based Cu(II)-Pyrazolate Photocatalytic Compounds for Solar-Light-Driven Hydrogen Evolution.

Polyoxometalates (POMs) are known for their photocatalytic hydrogen production activity, but their solubility and limited stability often restrict their practical applications. Herein, we designed and solvothermally synthesized two new Cu-H2bpz (3,3',5,5'-tetramethyl-4,4'-bipyrazole, abbreviated as H2bpz) compounds, namely, Cu0.5(H2bpz)(NO3) (1) and Cu(Hbpz)(Cl)·DMF (2), and three new polyoxometalate-based Cu(II)-pyrazolate compounds, namely, Cu(PW12O40)0.5(H2bpz)2(H2O)·(OH)0.5(H2O)5.5 (3), Cu(HPMo12O40)(H2bpz)2(H2O)2·(H2O)4 (4), and Cu2(SiW12O40)(H2bpz)3(H2O)3·(H2O)6 (5). Compound 3 (Cu(PW12O40)0.5(H2bpz)2(H2O)·(OH)0.5(H2O)5.5) exhibits the best photocatalytic activity of 44.4 µ L h-1 g-1, which may be related to the stability of compounds. Herein, the solvothermal method has been proven to be an effective method in synthesizing stable organic-inorganic hybrid compounds with soluble POMs, metal ions, and organic ligands. Thus, heterogeneous catalysts with outstanding solar-light-driven photocatalytic properties were obtained.

Supermolecule Cucurbituril Subnanoporous Carbon Supercapacitor (SCSCS).

It is quite challenging to prepare subnanometer porous materials from traditional porous precursors, and use of supramolecules as carbon sources was seldom reported due to the complex preparation and purification processes. We explore a facile one-pot method to fabricate supramolecular coordination compounds as carbon sources. The resultant CB[6]-derived carbons (CBC) have a high N content of 7.0-22.0%, surface area of 552-861 m2 g-1, and subnano/mesopores. The CBC electrodes have a narrow size distribution at 5.9 Å, and the supercapacitor exhibits an energy density of 117.1 Wh kg-1 and a potential window of over 3.8 V in a two-electrode system in the ionic liquid (MMIMBF4) electrolyte with appropriate cationic (5.8 Å) and anionic (2.3 Å) diameter. This work presents the facile fabrication of novel supermolecule cucurbituril subnanoporous carbon materials and the smart design of "pores and balls" for high-performance energy storage systems.

A novel CERNNE approach for predicting Parkinson's Disease-associated genes and brain regions based on multimodal imaging genetics data.

The detection and pathogenic factors analysis of Parkinson's disease (PD) has a practical significance for its diagnosis and treatment. However, the traditional research paradigms are commonly based on single neural imaging data, which is easy to ignore the complementarity between multimodal imaging genetics data. The existing researches also pay little attention to the comprehensive framework of patient detection and pathogenic factors analysis for PD. Based on functional magnetic resonance imaging (fMRI) data and single nucleotide polymorphism (SNP) data, a novel brain disease multimodal data analysis model is proposed in this paper. Firstly, according to the complementarity between the two types of data, the classical correlation analysis method is used to construct the fusion feature of subjects. Secondly, based on the artificial neural network, the fusion feature analysis tool named clustering evolutionary random neural network ensemble (CERNNE) is designed. This method integrates multiple neural networks constructed randomly, and uses clustering evolution strategy to optimize the ensemble learner by adaptive selective integration, selecting the discriminative features for PD analysis and ensuring the generalization performance of the ensemble model. By combining with data fusion scheme, the CERNNE is applied to forming a multi-task analysis framework, recognizing PD patients and predicting PD-associated brain regions and genes. In the multimodal data experiment, the proposed framework shows better classification performance and pathogenic factors predicting ability, which provides a new perspective for the diagnosis of PD.

The exploration of Parkinson's disease: a multi-modal data analysis of resting functional magnetic resonance imaging and gene data.

Parkinson's disease (PD) is the most universal chronic degenerative neurological dyskinesia and an important threat to elderly health. At present, the researches of PD are mainly based on single-modal data analysis, while the fusion research of multi-modal data may provide more meaningful information in the aspect of comprehending the pathogenesis of PD. In this paper, 104 samples having resting functional magnetic resonance imaging (rfMRI) and gene data are from Parkinson's Progression Markers Initiative (PPMI) and Alzheimer's Disease Neuroimaging Initiative (ADNI) database to predict pathological brain areas and risk genes related to PD. In the experiment, Pearson correlation analysis is adopted to conduct fusion analysis from the data of genes and brain areas as multi-modal sample characteristics, and the clustering evolution random forest (CERF) method is applied to detect the discriminative genes and brain areas. The experimental results indicate that compared with several existing advanced methods, the CERF method can further improve the diagnosis of PD and healthy control, and can achieve a significant effect. More importantly, we find that there are some interesting associations between brain areas and genes in PD patients. Based on these associations, we notice that PD-related brain areas include angular gyrus, thalamus, posterior cingulate gyrus and paracentral lobule, and risk genes mainly include C6orf10, HLA-DPB1 and HLA-DOA. These discoveries have a significant contribution to the early prevention and clinical treatments of PD.