Numerical Simulation of Maximum Spreading of an Impacting Ferrofluid Droplet under a Vertical Magnetic Field.

Theoretical modeling is proposed to predict the maximum spreading of water-based ferrofluid droplets impacting upon dry surfaces influenced by a vertical magnetic field. Constructed on the principle of energy balance, this model demonstrates excellent agreement with numerical findings across various impact velocities, contact angles, and magnetic strengths. Notably, as magnetic field strength escalates, magnetic forces prevail over viscous and capillary forces, exerting a significant influence on spreading dynamics and diminishing the maximum spreading diameter of ferrofluid droplets if the impacting shape is spherical. However, for freely falling droplets, the shape becomes prolate before impacting and the promoted surface energy balances the magnetic inhibitory effect on droplet spreading, thus resulting in an almost unchanged maximum spreading diameter. By postulating complete conversion of initial kinetic energy into magnetic energy, a scaling law is derived for maximum spreading diameter under extremely high magnetic fields. Further interpolation with viscous dissipation and capillary effects enables universal rescaling under diverse impact conditions. Through comparison with numerical outcomes, the validity of our theoretical model is affirmed, establishing a balanced formula between distinct energy components for predicting maximum spreading diameter of ferrofluid droplets accurately.

Ultra-Photostable Bacterial-Seeking Near-Infrared CPDs for Simultaneous NIR-II Bioimaging and Antibacterial Therapy.

Bacterial infections can pose significant health risks as they have the potential to cause a range of illnesses. These infections can spread rapidly and lead to complications if not promptly diagnosed and treated. Therefore, it is of great significance to develop a probe to selectively target and image pathogenic bacteria while simultaneously killing them, as there are currently no effective clinical solutions available. This study presents a novel approach using near-infrared carbonized polymer dots (NIR-CPDs) for simultaneous in vivo imaging and treatment of bacterial infections. The core-shell structure of the NIR-CPDs facilitates their incorporation into bacterial cell membranes, leading to an increase in fluorescence brightness and photostability. Significantly, the NIR-CPDs exhibit selective bacterial-targeting properties, specifically identifying Staphylococcus aureus (S. aureus) while sparing Escherichia coli (E. coli). Moreover, under 808 nm laser irradiation, the NIR-CPDs exhibit potent photodynamic effects by generating reactive oxygen species that target and damage bacterial membranes. In vivo experiments on infected mouse models demonstrate not only precise imaging capabilities but also significant therapeutic efficacy, with marked improvements in wound healing. The study provides the dual-functional potential of NIR-CPDs as a highly effective tool for the advancement of medical diagnostics and therapeutics in the fight against bacterial infections.

Albumin-seeking dyes with adjustable assemblies in situ enable programmable imaging windows and targeting tumor imaging.

Cyanine dyes are widely used organic probes for in vivo imaging due to their tunable fluorescence. They can form complexes with endogenous albumin, resulting in enhanced brightness and photostability. However, this binding is uncontrollable and irreversible, leading to considerable nonspecific background signals and unregulated circulation time. Methods: Here, we connect varying numbers of 4-(4-iodophenyl) butanoic acid (IP) as albumin-binding moieties (ABM) to the cyanine dye, enabling dynamic and controllable binding with albumin. Meanwhile, we provide a blocking method to completely release the dye from covalent capture with albumin, resulting in specific targeting fluorescence. Furthermore, we evaluate the pharmacokinetics and tumor targeting of the developed dyes. Results: The engineered dyes can dynamically and selectively bind with multiple albumins to change the in situ size of assemblies and circulation time, providing programmable regulation over the imaging time window. The nucleophilic substitution of meso-Cl with water-soluble amino acids or targeting peptides for IP-engineered dye further addresses the nonspecific signals caused by albumin, allowing for adjustable angiography time and efficient tumor targeting. Conclusion: This study rationalizes the binding modes of dyes and proteins, applicable to a wide range of near-infrared (NIR) dyes for improving their in vivo molecular imaging.

Biomimetic NIR-II fluorescent proteins created from chemogenic protein-seeking dyes for multicolor deep-tissue bioimaging.

Near-infrared-I/II fluorescent proteins (NIR-I/II FPs) are crucial for in vivo imaging, yet the current NIR-I/II FPs face challenges including scarcity, the requirement for chromophore maturation, and limited emission wavelengths (typically < 800 nm). Here, we utilize synthetic protein-seeking NIR-II dyes as chromophores, which covalently bind to tag proteins (e.g., human serum albumin, HSA) through a site-specific nucleophilic substitution reaction, thereby creating proof-of-concept biomimetic NIR-II FPs. This chemogenic protein-seeking strategy can be accomplished under gentle physiological conditions without catalysis. Proteomics analysis identifies specific binding site (Cys 477 on DIII). NIR-II FPs significantly enhance chromophore brightness and photostability, while improving biocompatibility, allowing for high-performance NIR-II lymphography and angiography. This strategy is universal and applicable in creating a wide range of spectrally separated NIR-I/II FPs for real-time visualization of multiple biological events. Overall, this straightforward biomimetic approach holds the potential to transform fluorescent protein-based bioimaging and enables in-situ albumin targeting to create NIR-I/II FPs for deep-tissue imaging in live organisms.

Site-specific albumin tagging with NIR-II fluorogenic dye for high-performance and super-stable bioimaging.

Synthetic near-infrared-II (NIR-II) dyes are promising for deep tissue imaging, yet they are generally difficult to target a given biomolecule with high specificity. Furthermore, the interaction mechanism between albumin and cyanine molecules, which is usually regarded as uncertain "complexes" such as crosslinked nanoparticles, remains poorly understood. Methods: Here, we propose a new class of NIR-II fluorogenic dyes capable of site-specific albumin tagging for in situ albumin seeking/targeting or constructing high-performance cyanine@albumin probes. We further investigate the interaction mechanism between NIR-II fluorogenic dyes and albumin. Results: We identify CO-1080 as an optimal dye structure that produces a stable/bright NIR-II cyanine@albumin probe. CO-1080 exhibits maximum supramolecular binding affinity to albumin while catalyzing their covalent attachment. The probe shows exact binding sites located on Cys476 and Cys101, as identified by proteomic analysis and docking modeling. Conclusion: Our cyanine@albumin probe substantially improves the pharmacokinetics of its free dye counterpart, enabling high-performance NIR-II angiography and lymphography. Importantly, the site-specific labeling tags between NIR-II fluorogenic dyes and albumin occur under mild conditions, offering a specific and straightforward synthesis strategy for NIR-II fluorophores in the fields of targeting bioimaging and imaging-guided surgery.

Association between inflammatory biomarkers and mortality in individuals with type 2 diabetes: NHANES 2005-2018.

PURPOSE: This study aimed to examine independent association between inflammatory biomarkers and all-cause mortality as well as cardio-cerebrovascular disease (CCD) mortality among U.S. adults with diabetes. METHODS: A cohort of 6412 U.S. adults aged 20 or older was followed from the start until December 31, 2019. Statistical models such as Cox proportional hazards model (Cox) and Kaplan-Meier (K-M) survival curves were employed to investigate the associations between the inflammatory biomarkers and all-cause mortality and CCD mortality. RESULTS: After adjusting for confounding factors, the highest quartile of inflammatory biomarkers (NLR HR = 1.99; 95 % CI:1.54-2.57, MLR HR = 1.93; 95 % CI:1.46-2.54, SII HR = 1.49; 95 % CI:1.18-1.87, SIRI HR = 2.32; 95 % CI:1.81-2.96, nLPR HR = 2.05; 95 % CI:1.61-2.60, dNLR HR = 1.94; 95 % CI:1.51-2.49, AISI HR = 1.73; 95 % CI:1.4 1-2.12)) were positively associated with all-cause mortality compared to those in the lowest quartile. K-M survival curves indicated that participants with an inflammatory biomarker above a certain threshold had a higher risk of both all-cause mortality and CCD mortality (Log rank P < 0.05). CONCLUSION: Some biomarkers such as NLR, MLR, SII, AISI, SIRI, and dNLR, are significantly associated with all-cause mortality and CCD mortality among U.S. adults with diabetes. The risk of both outcomes increased when the biomarkers surpassed a specific threshold.

Insights into the mechanisms underlying ethanol-induced changes in the dough mechanical properties and quality characteristics of fresh noodles.

This study investigated the effects of ethanol (0 %∼6%) on the dough mechanical properties and quality characteristics of fresh noodles and elucidated the relationship between the above changes and physicochemical, structural, and molecular properties of gluten. Ethanol reduced the water absorption (from 59.00 % to 52.33 %), stability time (from 8.17 min to 3.33 min) and viscoelasticity of dough, and increased the development time, weakening degree and compliance. Ethanol also decreased the fracture stress of dough sheet, and increased fracture elongation and adhesiveness (from 46.15 g·s to 75.88 g·s). Ethanol decreased the noodles' hardness (from 5347.41 g to 4442.34 g), break force, tensile distance, and water absorption, while cooking loss was increased. SEM and CLSM showed that ethanol destroyed the compactness of internal structure and inhibited the formation of gluten network in noodles. According to the results of SE-HPLC and RP-HPLC, ethanol dissolved part of the gliadin and inhibited the polymerization of protein.

High-precision detection and navigation surgery of colorectal cancer micrometastases.

Surgical resection is an effective treatment for colorectal cancer (CRC) patients, whereas occult metastases hinder the curative effect. Currently, there is no effective method to achieve intraoperatively diagnosis of tumor-positive lymph nodes (LNs). Herein, we adopt a near-infrared-II (NIR-II) organic donor-pi-acceptor-pi-donor probe FE-2PEG, which exhibits bright fluorescence over 1100 nm, excellent photostability, blood circulation time, and biocompatibility, to achieve high-performance bioimaging with improved temporal and spatial resolution. Importantly, the FE-2PEG shows efficient passive enrichment in orthotopic CRC, metastatic mesenteric LNs, and peritoneal metastases by enhanced permeability and retention effect. Under NIR-II fluorescence-guided surgery (FGS), the peritoneal micrometastases were resected with a sensitivity of 94.51%, specificity of 86.59%, positive predictive value (PPV) of 96.57%, and negative predictive value of 79.78%. The PPV still achieves 96.07% even for micrometastases less than 3 mm. Pathological staining and NIR-II microscopy imaging proved that FE-2PEG could successfully delineate the boundary between the tumor and normal tissues. Dual-color NIR-II imaging strategy with FE-2PEG (1100 ~ 1300 nm) and PbS@CdS quantum dots (> 1500 nm) successfully protects both blood supply and normal tissues during surgery. The NIR-II-based FGS provides a promising prospect for precise intraoperative diagnosis and minimally invasive surgery of CRC.

Dye-Triplet-Sensitized Downshifting Nanoprobes with Ratiometric Dual-NIR-IIb Emission for Accurate In Vivo Detection.

Despite the emerging near-infrared-IIb (NIR-IIb, 1500-1700 nm) bioimaging significantly improving the in vivo penetration depth and resolution, quantitative detection with accuracy remains challenging due to its inhomogeneous fluorescence signal attenuation in biological tissue. Here, ratiometric dual-NIR-IIb in vivo detection with excitation wavelengths of 808 and 980 nm is presented using analyte-responsive dye-triplet-sensitized downshifting nanoprobes (DSNPs). NIR cyanine dye IR-808, a recognizer of biomarker hypochlorite (ClO-), is introduced to trigger a triplet energy transfer process from the dye to Er3+ ions of DSNPs under 808 nm excitation, facilitating the formation of an analyte-responsive 1525 nm NIR-IIb assay channel. Meanwhile, DSNPs also enable emitting intrinsic nonanalyte-dependent downshifting fluorescence at the same NIR-IIb window under 980 nm excitation, serving as a self-calibrated signal to alleviate the interference from the probe amount and depth. Due to the two detected emissions sharing identical light propagation and scattering, the ratiometric NIR-IIb signal is demonstrated to ignore the depth of penetration in biotissue. The arthritis lesions are distinguished from normal tissue using ratiometric probes, and the amount of ClO- can be accurately output by the established detection curves.

Protein@Cyanine-Based NIR-II Lymphography Enables the Supersensitive Visualization of Lymphedema and Tumor Lymphatic Metastasis.

Visualization of the lymphatic system is clinically indispensable for the diagnosis and/or treatment of lymphatic diseases. Although indocyanine green (ICG) lymphography becomes an alternate imaging modality compared to traditional lymphoscintigraphy, it is still far from ideal due to the insufficient detection depth and low spatiotemporal resolution. Herein, protein@cyanine probes are rationally developed to solve the limitations of the current near-infrared-I (NIR-I) lymphography. The protein@cyanine probes are synthesized following a chlorine-containing dye-labeling strategy based on structure-selectivity (facile covalent binding between the dye and protein with a 1:1 molar ratio). As expected, the probes display exceptional NIR-II imaging ability with much-improved imaging contrast/resolution and controllable pharmacokinetics, superior to the clinical ICG. The protein@cyanine probes locate lymph nodes and delineate lymphatic vessels with super-high sensitivity and signal-to-background ratio, enabling real-time diagnosing lymphatic diseases such as lymphedema and tumor lymphatic metastasis. In particular, the NIR-II lymphography provides an opportunity to discover the disparate morbidity rate of primary lymphedema in different types of mice. Given the fact of lacking clinically transferable NIR-II probes, this work not only provides a promising strategy for enriching of the current library of NIR-II probes, but also promotes the clinical translation of NIR-II lymphography technology.

Surfactant-chaperoned donor-acceptor-donor NIR-II dye strategy efficiently circumvents intermolecular aggregation to afford enhanced bioimaging contrast.

Fluorescence emission in the near-infrared-II (NIR-II) optical window affords reduced autofluorescence and light scattering, enabling deep-tissue visualization for both disease detection and surgical navigation. Small-molecule NIR-II dyes are preferable for clinical bioimaging applications, as the flexibility in their molecular synthesis allows for precise control of their optical and pharmacokinetic properties. Among the various types of dye, donor-acceptor-donor-based (D-A-D) dyes demonstrate exceptional photostability, whereas the frequently used PEGylation approach does not keep their intrinsic brightness enough in water environments due to their inherent effect of self-assembly. Here, we demonstrate that the commercially-available surfactants can serve as a dispersant to prevent molecular aggregation of PEGylated D-A-D dyes. Due to the favorable energetics for co-assembly between D-A-D dyes and surfactants, the formed surfactant-chaperoned dye strategy dramatically increases dye brightness. Accordingly, this effect provides remarkably improved performance for in vivo bioimaging applications. In parallel, we also investigate the D-A-D dye uptake and signal enhancement properties in the liver of murine models and demonstrate that the lumen-lining Kupffer cells can potentially disassemble PEGylated D-A-D aggregates such that their inherent brightness is restored. This phenomenon is similar to the surfactant-chaperoned dye strategy and our investigations provide a positive addition to better use of the current NIR-II fluorophores, especially for visualizing high-brightness required events.

Near-infrared-II ratiometric fluorescence probes for non-invasive detection and precise navigation surgery of metastatic sentinel lymph nodes.

Sentinel lymph node (SLN) biopsy is the key diagnostic procedure to determine tumor metastasis and treatment plan. Current SLN biopsy has considerable drawbacks in that SLNs (both malignant and normal) must be removed by navigation surgery, followed by a time-consuming pathological examination. The selective, non-invasive, and real-time diagnosis of metastatic status in SLNs is becoming essential. Methods: Here, we design two lanthanide-doped nanoparticles as a pair of NIR-II ratiometric fluorescence probes, one of which is conjugated with tumor-targeting moiety, while the other is conjugated with PEG as an internal reference. The NIR-II ratiometric fluorescence signal (I1060 nm/I1525 nm) from two well-separated channels were used to identify the tumor-draining SLNs. The precise navigation surgery of metastatic SLNs was performed and we further evaluated their surgery outcomes. Results: The NIR-II ratiometric fluorescence facilitates an ideal fluorescence-guided surgery with only resection of tumor-positive SLNs, thereby avoiding unnecessary removal of the normal SLNs. In addition, our system has a time-saving operation procedure and can be performed under the operation light without altering the appearance of surgical settings. Conclusion: The present study enables non-invasive and real-time detection metastatic status in SLNs with high sensitivity and selectivity. Our investigations will provide a new direction for SLN biopsy and substantially improve cancer surgery outcomes.

Near-Infrared Carbonized Polymer Dots for NIR-II Bioimaging.

Carbon dots (CDs) or carbonized polymer dots (CPDs) are an emerging class of optical materials that have exceptional applications in optoelectronic devices, catalysis, detection, and bioimaging. Although cell studies of CPDs have produced impressive results, in vivo imaging requires available CPDs to fluoresce in the near-infrared-II (NIR-II) window (1000-1700 nm). Here, a two-step bottom-up strategy is developed to synthesize NIR-CPDs that provide bright emissions in both NIR-I and NIR-II transparent imaging windows. The designed strategy includes a hydrothermal reaction to form a stable carbon core with aldehyde groups, followed by the Knoevenagel reaction to tether the molecular emission centers. This procedure is labor-saving, cost-efficient, and produces a high yield. The NIR-CPDs enable high-performance NIR-II angiography and real-time imaging of the disease degree of colitis noninvasively. This technology may therefore provide a next-generation synthesis strategy for CPDs with rational molecular engineering that can accurately tune the absorption/emission properties of NIR-emissive CPDs.

Ultrabright Renal-Clearable Cyanine-Protein Nanoprobes for High-Quality NIR-II Angiography and Lymphography.

The renal-clearable aspect of imaging agent with minimum toxicity issues and side effects is essential for clinical translation, yet clinical near-infrared-I/II (NIR-I/II) fluorophores with timely renal-clearance pathways are very limited. Herein, we rationally develop the cyanine-protein composite strategy through covalent bonding of ß-lactoglobulin (ß-LG) and chloride-cyanine dye to produce a brilliant and stable NIR-I/II fluorophore (e.g., ß-LG@IR-780). The ß-LG acts as a protecting shell with small molecular weight (18.4 kDa) and ultrasmall size (<5 nm), thus endowing the ß-LG@IR-780 with excellent biocompatibility and renal excretion. Our ß-LG@IR-780 probe enables noninvasive and precise NIR-II visualization of the physiological and pathological conditions of the vascular and lymphatic drainage system, facilitating intraoperative imaging-guided surgery and postoperative noninvasive monitoring. The minimum accumulation of our probes in the main organs improves the overall biosafety. This study provides a facile methodology for new-generation NIR-II fluorophores and largely improves the brightness and pharmacokinetics of small molecular dyes.

DEMOC: a deep embedded multi-omics learning approach for clustering single-cell CITE-seq data.

Advances in single-cell RNA sequencing (scRNA-seq) technologies has provided an unprecedent opportunity for cell-type identification. As clustering is an effective strategy towards cell-type identification, various computational approaches have been proposed for clustering scRNA-seq data. Recently, with the emergence of cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq), the cell surface expression of specific proteins and the RNA expression on the same cell can be captured, which provides more comprehensive information for cell analysis. However, existing single cell clustering algorithms are mainly designed for single-omic data, and have difficulties in handling multi-omics data with diverse characteristics efficiently. In this study, we propose a novel deep embedded multi-omics clustering with collaborative training (DEMOC) model to perform joint clustering on CITE-seq data. Our model can take into account the characteristics of transcriptomic and proteomic data, and make use of the consistent and complementary information provided by different data sources effectively. Experiment results on two real CITE-seq datasets demonstrate that our DEMOC model not only outperforms state-of-the-art single-omic clustering methods, but also achieves better and more stable performance than existing multi-omics clustering methods. We also apply our model on three scRNA-seq datasets to assess the performance of our model in rare cell-type identification, novel cell-subtype detection and cellular heterogeneity analysis. Experiment results illustrate the effectiveness of our model in discovering the underlying patterns of data.

Super-stable cyanine@albumin fluorophore for enhanced NIR-II bioimaging.

Near-infrared-II (NIR-II) dyes could be encapsulated by either exogenous or endogenous albumin to form stable complexes for deep tissue bioimaging. However, we still lack a complete understanding of the interaction mechanism of the dye@albumin complex. Studying this principle is essential to guide efficient dye synthesis and develop NIR-II probes with improved brightness, photostability, etc. Methods: Here, we screen and test the optical and chemical properties of dye@albumin fluorophores, and systematically investigate the binding sites and the relationship between dye structures and binding degree. Super-stable cyanine dye@albumin fluorophores are rationally obtained, and we also evaluate their pharmacokinetics and long-lasting NIR-II imaging abilities. Results: We identify several key parameters of cyanine dyes governing the supramolecular/covalent binding to albumin, including a six-membered ring with chlorine (Cl), the small size of side groups, and relatively high hydrophobicity. The tailored fluorophore (IR-780@albumin) exhibits much-improved photostability, serving as a long-lasting imaging probe for NIR-II bioimaging. Conclusion: Our study reveals that the chloride-containing cyanine dyes with the above-screened chemical structure (e.g. IR-780) could be lodged into albumin more efficiently, producing a much more stable fluorescent probe. Our finding partly solves the photobleaching issue of clinically-available cyanine dyes, enriching the probe library for NIR-II bioimaging and imaging-guided surgery.

Can automated driving prevent crashes with distracted Pedestrians? An exploration of motion planning at unsignalized Mid-block crosswalks.

Pedestrian distraction may provoke severe difficulties in automated vehicle (AV) control, which may significantly affect the safety performance of AVs, especially at unsignalized mid-block crosswalks (UMCs). However, there is no available motion-planning model for AVs that considers the effect of pedestrian distraction on UMCs. This study aims to explore innovative approaches for safe and reasonable automated driving in response to distracted pedestrians with various speed profiles at UMCs. Based on two common model design concepts, two new models are established for AVs: a rule-based model that solves motion plans through a fixed calculation procedure incorporating several optimization models, and a learning-based model that replaces the deterministic optimization process with policy-gradient reinforcement learning. The developed models were assessed through simulation experiments in which pedestrian speed profiles were defined using empirical data from field surveys. The results reveal that the learning-based model has outstanding safety performance, whereas the rule-based model leads to remarkable safety problems. For distracted pedestrians with significant crossing-speed changes, rule-based AVs lead to a 5.1% probability of serious conflict and a 1.4% crash probability. The learning-based model is oversensitive to risk and always induces high braking rates, which results in unnecessary efficiency loss. To overcome this, a hybrid model based on the learning-based model was developed, which introduces a rule-based acceleration value to regularize the action space of the proposed learning-based model. The results indicate that the hybrid approach outperforms the other two models in preventing crash hazards from distracted pedestrians by employing appropriate braking behaviors. The high safety performance of the hybrid models can be attributed to the spontaneous slowing down of the vehicle that initiates before detecting pedestrians on UMCs. Although such a cautious driving pattern leads to extra delay, the time cost of the hybrid model is acceptable considering the significant improvements in ensuring pedestrian safety.

Electrostatic gating of single-molecule junctions based on the STM-BJ technique.

The gating of charge transport through single-molecule junctions is considered a critical step towards molecular circuits but remains challenging. In this work, we report an electrostatic gating method to tune the conductance of single-molecule junctions using the scanning tunneling microscope break junction (STM-BJ) technique incorporated with a back-gated chip as a substrate. We demonstrated that the conductance varied at different applied gating voltages (Vgs). The HOMO-dominated molecules show a decrease in conductance with an increase in Vg, and the LUMO-dominated molecules show the opposite trend. The measured conductance trends with Vg are consistent with the transition voltage spectroscopy measurements. Moreover, the transmission functions simulated from density functional theory (DFT) calculations and the finite element analysis all suggest that Vg changed the energy alignment of the molecular junction. This work provides a simple method for modulating the molecular orbitals' alignment relative to the Fermi energy (Ef) of metal electrodes to explore the charge transport properties at the single-molecule scale.

Indole Carbonized Polymer Dots Boost Full-Color Emission by Regulating Surface State.

Carbonized polymer dots (CPDs) are impressive imaging probes with great potential for enriching the library of metal-free fluorescent materials, yet current strategies have struggled to achieve products that emit full-color light in a single reaction system. Establishing an efficient and robust synthesis approach that unlocks the color barrier to the luminescence centers of specific CPDs remains a challenge. Herein, the surface-state engineering of pyridine and amide in the indole system to create a palette of resolvable full-color light-emissive CPDs is reported. Detailed structural analysis revealed that cationic polymerization and oxidation reactions potentially contribute to the formation of the main frameworks and emission centers of the final CPDs, with emissive oxygen- and nitrogen-based centers fixed by cross-linked polymer structures. This study provides valuable insight into the energy absorbance and photoluminescence mechanism of CPDs and introduces additional reactants (benzo heterocycle) into CPD research.

Self-Learning Perfect Optical Chirality via a Deep Neural Network.

Optical chirality occurs when materials interact differently with light in a specific circular polarization state. Chiroptical phenomena inspire wide interdisciplinary investigations, which require advanced designs to reach strong chirality for practical applications. The development of artificial intelligence provides a new vision for the manipulation of light-matter interaction beyond the theoretical interpretation. Here, we report a self-consistent framework named the Bayesian optimization and convolutional neural network that combines Bayesian optimization and deep convolutional neural network algorithms to calculate and optimize optical properties of metallic nanostructures. Both electric-field distributions at the near field and reflection spectra at the far field are calculated and self-learned to suggest better structure designs and provide possible explanations for the origin of the optimized properties, which enables wide applications for future nanostructure analysis and design.