Towards artificial general intelligence with hybrid Tianjic chip architecture.

There are two general approaches to developing artificial general intelligence (AGI)1: computer-science-oriented and neuroscience-oriented. Because of the fundamental differences in their formulations and coding schemes, these two approaches rely on distinct and incompatible platforms2-8, retarding the development of AGI. A general platform that could support the prevailing computer-science-based artificial neural networks as well as neuroscience-inspired models and algorithms is highly desirable. Here we present the Tianjic chip, which integrates the two approaches to provide a hybrid, synergistic platform. The Tianjic chip adopts a many-core architecture, reconfigurable building blocks and a streamlined dataflow with hybrid coding schemes, and can not only accommodate computer-science-based machine-learning algorithms, but also easily implement brain-inspired circuits and several coding schemes. Using just one chip, we demonstrate the simultaneous processing of versatile algorithms and models in an unmanned bicycle system, realizing real-time object detection, tracking, voice control, obstacle avoidance and balance control. Our study is expected to stimulate AGI development by paving the way to more generalized hardware platforms.

Two Birds with One Stone: Broadband Electromagnetic Wave Absorption and Anticorrosion Performance in 2-18 GHz for Prussian Blue Analog Derivatives Aimed for Practical Applications.

Nowadays, the exploration of electromagnetic (EM) wave absorbers with anticorrosion to improve the survivability and environmental adaptability of military targets in the harsh environments is becoming an attractive and unavoidable challenge. Herein, through modulation of the metal composition in the precursors, the core@shell structure Prussian blue analog-derived NiCo@C, CoFe@C, NiFe@C, and NiCoFe@C are obtained with excellent EM wave absorption performance. As for NiCoFe@C, ascribed to the coupling effect of the dual magnetic alloy, a minimum reflection loss (RL) of -47.6 dB and an effective absorption bandwidthof 5.83 GHz are realized, which cover the whole Ku-band. Meanwhile, four absorbers display the lower corrosion current density (10-4 -10-6 A cm-2 ) and larger polarization resistance (104 -106  Ω) under acid, neutral, and alkaline corrosion conditions over uninterrupted 30 days. Furthermore, due to the spatial barrier effect and the passivation effect of the graphitic carbon shell , the continuous salt spray test has little effect on RL performance and inconspicuously changes the surface morphologies of coating, demonstrating its excellent bifunctional performance. This work lays the foundation for the development of metal-organic frameworks-derived materials with both anticorrosion and EM wave absorption performance.

NDRG1 facilitates the replication and persistence of Kaposi's sarcoma-associated herpesvirus by interacting with the DNA polymerase clamp PCNA.

Kaposi's sarcoma-associated herpesvirus (KSHV) latently infects host cells and establishes lifelong persistence as an extra-chromosomal episome in the nucleus. To persist in proliferating cells, the viral genome typically replicates once per cell cycle and is distributed into daughter cells. This process involves host machinery utilized by KSHV, however the underlying mechanisms are not fully elucidated. In present study, we found that N-Myc downstream regulated gene 1 (NDRG1), a cellular gene known to be non-detectable in primary B cells and endothelial cells which are the major cell types for KSHV infection in vivo, was highly upregulated by KSHV in these cells. We further demonstrated that the high expression of NDRG1 was regulated by latency-associated nuclear antigen (LANA), the major viral latent protein which tethers the viral genome to host chromosome and plays an essential role in viral genome maintenance. Surprisingly, knockdown of NDRG1 in KSHV latently infected cells resulted in a significant decrease of viral genome copy number in these cells. Interestingly, NDRG1 can directly interact with proliferating cell nuclear antigen (PCNA), a cellular protein which functions as a DNA polymerase clamp during DNA replication. Intriguingly, we found that NDRG1 forms a complex with LANA and PCNA and serves as a scaffold protein bridging these two proteins. We further demonstrated that NDRG1 is critical for mediating LANA to recruit PCNA onto terminal repeat (TR) of KSHV genome, and facilitates viral DNA replication and episome persistence. Taken together, our findings suggest that NDRG1 plays an important role in KSHV viral genome replication, and provide new clues for understanding of KSHV persistence.

Male hormones activate EphA2 to facilitate Kaposi's sarcoma-associated herpesvirus infection: Implications for gender disparity in Kaposi's sarcoma.

There is increasing consensus that males are more vulnerable than females to infection by several pathogens. However, the underlying mechanism needs further investigation. Here, it was showed that knockdown of androgen receptor (AR) expression or pre-treatment with 5α-dihydrotestosterone, the AR agonist, led to a considerably dysregulated Kaposi's sarcoma-associated herpesvirus (KSHV) infection. In endothelial cells, membrane-localized AR promoted the endocytosis and nuclear trafficking of KSHV. The AR interacted with ephrin receptor A2 (EphA2) and increased its phosphorylation at residue Ser897, which was specifically upregulated upon KSHV infection. This phosphorylation resulted from the AR-mediated recruitment of Src, which resulted in the activation of p90 ribosomal S6 kinase 1 (RSK1), which directly phosphorylates EphA2 at Ser897. Finally, the EphA2-mediated entry of KSHV was abolished in a Ser897Asn EphA2 mutant. Taken together, membrane-localized AR was identified as a KSHV entry factor that cooperatively activates Src/RSK1/EphA2 signaling, which subsequently promotes KSHV infection of both endothelial and epithelial cells.

Lighting Deviation Correction for Integrating-Sphere Multispectral Imaging Systems.

In an integrating sphere multispectral imaging system, measurement inconsistency can arise when acquiring the spectral reflectances of samples. This is because the lighting condition can be changed by the measured samples, due to the multiple light reflections inside the integrating sphere. Besides, owing to non-uniform light transmission of the lens and narrow-band filters, the measured reflectance is spatially dependent. To deal with these problems, we propose a correction method that consists of two stages. The first stage employs a white board to correct non-uniformity and a small white patch to correct lighting deviation, both under the assumption of ideal Lambertian reflection. The second stage uses a polynomial regression model to further remove the lighting inconsistency when measuring non-Lambertian samples. The method is evaluated on image data acquired in a real multispectral imaging system. Experimental results illustrate that our method eliminates the measurement inconsistency considerably. This consequently improves the spectral and colorimetric accuracy in color measurement, which is crucial to practical applications.

Guanylate-Binding Protein 1 Inhibits Nuclear Delivery of Kaposi's Sarcoma-Associated Herpesvirus Virions by Disrupting Formation of Actin Filament.

Kaposi's sarcoma-associated herpesvirus (KSHV) is a typical gammaherpesvirus that establishes persistent lifelong infection in host cells. In order to establish successful infection, KSHV has evolved numerous immune evasion strategies to bypass or hijack the host immune system. However, host cells still produce immune cytokines abundantly during primary KSHV infection. Whether the immune effectors produced are able to inhibit viral infection and how KSHV successfully conquers these immune effectors remain largely unknown. The guanylate-binding protein 1 (GBP1) gene is an interferon-stimulated gene and exerts antiviral functions on several RNA viruses; however, its function in DNA virus infection is less well understood. In this study, we found that KSHV infection increases both the transcriptional and protein levels of GBP1 at the early stage of primary infection by activating the NF-κB pathway. The overexpression of GBP1 significantly inhibited KSHV infection, while the knockdown of GBP1 promoted KSHV infection. The GTPase activity and dimerization of GBP1 were demonstrated to be responsible for its anti-KSHV activity. Furthermore, we found that GBP1 inhibited the nuclear delivery of KSHV virions by disrupting the formation of actin filaments. Finally, we demonstrated that replication and transcription activator (RTA) promotes the degradation of GBP1 through a proteasome pathway. Taken together, these results provide a new understanding of the antiviral mechanism of GBP1, which possesses potent anti-KSHV activity, and suggest the critical role of RTA in the evasion of the innate immune response during primary infection by KSHV.IMPORTANCE GBP1 can be induced by various cytokines and exerts antiviral activities against several RNA viruses. Our study demonstrated that GBP1 can exert anti-KSHV function by inhibiting the nuclear delivery of KSHV virions via the disruption of actin filaments. Moreover, we found that KSHV RTA can promote the degradation of GBP1 through a proteasome-mediated pathway. Taken together, our results elucidate a novel mechanism of GBP1 anti-KSHV activity and emphasize the critical role of RTA in KSHV evasion of the host immune system during primary infection.

[Clinical Value of Detecting ABL Kinase Domain Mutations in Patients with Chronic Myeloid Leukemia Based on High-Throughput Sequencing Technology].

OBJECTIVE: To compare the efficacy and clinical value of high-throughput sequencing (HTS) and Sanger sequencing in detecting ABL kinase domain mutations in patients with chronic myeloid leukemia (CML). METHODS: A total of 198 samples of 147 CML patients from July 2017 to March 2021 in Henan Cancer Hospital were collected and underwent high-throughput sequencing and Sanger sequencing to detect the mutations in ABL kinase domain, and the relevant clinical data were collected for comparative analysis. RESULTS: The proportion of total mutations and ≥2 mutations detected by high-throughput sequencing were significantly higher than those detected by Sanger sequencing (P =0.01; P =0.046). ≥2 mutations were detected in 22 cases, of which 5 cases (22.7%) had compound mutations. High-throughput sequencing can detect low level mutations that cannot be detected by Sanger sequencing. In 198 samples, 25 (12.6%) were low level mutations, 33 (16.7%) were high level mutations and 10 (5.1%) were mixed high and low level mutations. In the analysis of related clinical factors, the total mutation rate and the low level mutation rate in the optimal period, failure period and warning period were gradually increased (total mutation rate, P =0.016; low level mutation rate, P =0.005). The mutation rate of the samples with additional chromosomal abnormalities was also significantly increased (P =0.009). The mutation rate of patients who received first- and second-line treatment was significantly lower than that of patients who received third- or higher-line treatment (P =0.006). Analysis based on variant allele frequency (VAF) of the mutation site was helpful to visually evaluate the clonal evolution status of TKI-resistance CML cells. CONCLUSION: High-throughput sequencing is more sensitive and accurate than Sanger sequencing in mutation detection, which is helpful to accurately and visually evaluate TKI treatment response and optimize treatment strategy for CML.

The immunogenicity of a novel cytotoxic T lymphocyte epitope from tumor antigen PL2L60 could be enhanced by 4-chlorophenylalanine substitution at position 1.

PIWIL2, a member of PIWI/AGO family, is expressed in germline stem cells and precancerous stem cells, but not in adult somatic cells. PIWIL2 plays an important role in tumor development. It is considered as a cancer­testis antigen (CT80). It has been reported that the spliced fragment of PIWIL2, PL2L60, was widely expressed in cancer cell lines. In this study, HLA-A2-restricted epitopes from PL2L60 were predicted by online tools. To improve the activity of the native epitope, a candidate peptide P281 with potent binding affinity was chosen to investigate the modification strategy. A series of aromatic amino acids were introduced to substitute the first residue of P281. Then, we tested the binding affinity and stability of the peptide analogs and their ability to elicit specific immune responses both in vitro and in vivo. Our results indicated that the cytotoxic T lymphocytes (CTLs) induced by [4-Cl-Phe1]P281 could elicit more potent activities than that of P281 and other analogs. The CTLs induced by this analog could lyze target cells in HLA-A2-restricted and antigen-specific manners. [4-Cl-Phe1]P281 also showed the best resistance against degradation in human serum. In conclusion, the introduction of the unnatural amino acid, 4-Cl-Phe, into the first position could enhance the activity of the native epitope to induce cytotoxic T lymphocytes. It might be a good strategy to modify other promising native epitopes. The novel epitopes identified in this study could be used as novel candidates to the immunotherapy of HLA-A2 positive patients with tumors expressing PL2L60.

Cost-performance analysis of nutrient removal in a full-scale oxidation ditch process based on kinetic modeling.

A full-scale oxidation ditch process for treating sewage was simulated with the ASM2d model and optimized for minimal cost with acceptable performance in terms of ammonium and phosphorus removal. A unified index was introduced by integrating operational costs (aeration energy and sludge production) with effluent violations for performance evaluation. Scenario analysis showed that, in comparison with the baseline (all of the 9 aerators activated), the strategy of activating 5 aerators could save aeration energy significantly with an ammonium violation below 10%. Sludge discharge scenario analysis showed that a sludge discharge flow of 250-300 m3/day (solid retention time (SRT), 13-15 days) was appropriate for the enhancement of phosphorus removal without excessive sludge production. The proposed optimal control strategy was: activating 5 rotating disks operated with a mode of "111100100" ("1" represents activation and "0" represents inactivation) for aeration and sludge discharge flow of 200 m3/day (SRT, 19 days). Compared with the baseline, this strategy could achieve ammonium violation below 10% and TP violation below 30% with substantial reduction of aeration energy cost (46%) and minimal increment of sludge production (< 2%). This study provides a useful approach for the optimization of process operation and control.

Design and stabilization analysis of luxury cruise with dynamic positioning systems based on sampled-data control.

This paper studies the sampled-data control issue for a luxury cruise (LC) with dynamic positioning system (DPS). The design method and mathematical model of LC is given. By constructing an improved time-dependent Lyapunov-Krasovskii function (LKF) by adding new useful terms, the sampling pattern is fully captured and less conservatism of the results are obtained. Based on the constructed the LKF, the new stability criterion is obtained and the sampled-data controller for LC with DPS is designed. Finally, an example is exhibited to prove that the proposed approach is valid and applicable.

Cognition and Interaction: From the Perspective of Daily Therapeutic Landscape of the Coastal Zone.

This study explored the relationship between mental and physical therapeutic effects through three dimensions: man-environment relationships, a sense of place, and symbolic landscapes. The study used a combination of qualitative and quantitative research methods. Local residents living in the coastal area of Xinglin Bay were the research objects. Quantitative data analysis revealed that the frequency of residents' visits was an important variable affecting their physical and treatment perceptions. For those who visit frequently, these visits can evoke memories, which can better express their sensory experience. The text analysis showed that residents picked up two major landscape elements to form the sense of place and symbolic landscape: one is the water body in the coastal zone, and the other is the cultural symbol of the peninsula. Based on untoward event experience, the residents assembled the elements into a new spatial relationship with therapeutic affordance.

[Strategy Optimization and Clinical Analyze of Multiple Nucleotide Polymorphism Analysis in the Chimerism Detection after Allogeneic Hematopoietic Stem Cell Transplantation].

AbstractObjective: To investigate the sample selection, result correction and clinical application value of multi nucleotide polymorphism chimerism detection method based on Next-generation sequencing. METHODS: The chimerism samples from November 2018 to June 2020 were collected, and Pearson correlation coefficient (r) was used to analyze the consistency of bone marrow and peripheral blood results detected by MNPseq; according to the different information integrity before transplantation, the calibration model was constructed to analyze the correction value of the micro chimerism results in each model; the clinical results were retrospectively analyzed to verify the reliability and practicability of chimerism results correction and the clinical value of MNPseq method. RESULTS: The results of bone marrow and peripheral blood chimerism detected by MNPseq method were consistent with each other and showed significant correlation (r=0.985, P<0.01). The three groups of calibration models were constructed according to different pre-transplant information. For the no donor and pre-transplant patients information group, the correction value was 1%; while for the group with pre-transplant patients and without donor information, 0.61% of the chimerism rate and 13 heterotopic points were used as the correction value; 0.26% of the chimerism rate and 21.57% of the heterotopic points were used as the correction value for the group with pre-transplantation patients and donor information. After correction, the number of the patients with incomplete chimerism decreased from 276 (74.19%) to 141 (37.91%) (P<0.01). Among 18 (18/141, 12.77%) patients with incomplete chimerism, the results of MNPseq in the patients were 25-39 days earlier than those in STR and flow MRD, and the result showed statistical significance. CONCLUSION: MNPseq method can be used to monitor chimerism with peripheral blood instead of bone marrow samples, and the results can be corrected to detect the changes of graft status in vivo in a more timely manner.

A novel baseline-correction method for standard addition based derivative spectra and its application to quantitative analysis of benzo(a)pyrene in vegetable oil samples.

In the present work, a baseline-correction method based on peak-to-derivative baseline measurement was proposed for the elimination of complex matrix interference that was mainly caused by unknown components and/or background in the analysis of derivative spectra. This novel method was applicable particularly when the matrix interfering components showed a broad spectral band, which was common in practical analysis. The derivative baseline was established by connecting two crossing points of the spectral curves obtained with a standard addition method (SAM). The applicability and reliability of the proposed method was demonstrated through both theoretical simulation and practical application. Firstly, Gaussian bands were used to simulate 'interfering' and 'analyte' bands to investigate the effect of different parameters of interfering band on the derivative baseline. This simulation analysis verified that the accuracy of the proposed method was remarkably better than other conventional methods such as peak-to-zero, tangent, and peak-to-peak measurements. Then the above proposed baseline-correction method was applied to the determination of benzo(a)pyrene (BaP) in vegetable oil samples by second-derivative synchronous fluorescence spectroscopy. The satisfactory results were obtained by using this new method to analyze a certified reference material (coconut oil, BCR(®)-458) with a relative error of -3.2% from the certified BaP concentration. Potentially, the proposed method can be applied to various types of derivative spectra in different fields such as UV-visible absorption spectroscopy, fluorescence spectroscopy and infrared spectroscopy.

Over-expression of microRNA169 confers enhanced drought tolerance to tomato.

Plant miRNA regulates multiple developmental and physiological processes, including drought responses. We found that the accumulation of Sly-miR169 in tomato (Solanum lycopersicum) was induced by drought stress. Consequently, Sly-miR169 targets, namely, three nuclear factor Y subunit genes (SlNF-YA1/2/3) and one multidrug resistance-associated protein gene (SlMRP1), were significantly down-regulated by drought stress. Constitutive over-expression of a miR169 family member, Sly-miR169c, in tomato plant can efficiently down-regulate the transcripts of the target genes. Compared with non-transgenic plants, transgenic plants over-expressing Sly-miR169c displayed reduced stomatal opening, decreased transpiration rate, lowered leaf water loss, and enhanced drought tolerance. Our study is the first to provide evidence that the Sly-miR169c negatively regulates stomatal movement in tomato drought responses.

Effect of the Recombinant Adenovirus-Mediated HIF-1 Alpha on the Expression of VEGF in the Hypoxic Brain Microvascular Endothelial Cells of Rats.

OBJECTIVE: To investigate the effect of recombinant adenovirus-mediated HIF-1 alpha (HIF-1α) on the expression of vascular endothelial growth factor (VEGFA) and HIF-1α in hypoxic brain microvascular endothelial cells (BMEC) in rats. METHODS: Primary cultured rat BMEC in vitro were treated without or with either recombinant adenovirus-mediated hypoxia-inducible factor-1 alpha (AdHIF-1α) or recombinant adenovirus empty vector (Ad) in the presence of CoCl2 (simulating hypoxia conditions), or were grown under normoxia conditions. The expression of VEGFA and HIF-1α was analyzed at 12h, 24h, 48h and 72h incubation time, respectively. We also accessed a GEO dataset of stroke to analyze in vivo the alteration of HIF-1α and VEGFA expression, and the correlations between HIF-1α, VEGFA and CD31 mRNA levels in vascular vessels after stroke. RESULTS: VEGFA and HIF-1α expression were significantly higher in at each time point in the AdHIF-1α than other groups (p<0.05), whereas the Ad group and hypoxia group, showed no statistically significant difference (p>0.05). Moreover, VEGFA and HIF-1α levels were significantly higher in BMEC under hypoxia conditions than normoxia conditions (p <0.05). Both HIF-1α and VEGFA expression significantly increased after stroke in vivo with 1.30 and 1.57 fold-change in log2, respectively. There were significantly positive associations between HIF-1α, VEGFA and CD31 mRNA levels in vivo after stroke. CONCLUSION: Hypoxia-induced HIF-1α and VEGFA expression in vascular vessels, and recombinant AdHIF-1α could up-regulate VEGFA, and enhance HIF-1ααlevels in BMEC in vitro, which may play an important role in the recovery of stroke.

A hybrid and scalable brain-inspired robotic platform.

Recent years have witnessed tremendous progress of intelligent robots brought about by mimicking human intelligence. However, current robots are still far from being able to handle multiple tasks in a dynamic environment as efficiently as humans. To cope with complexity and variability, further progress toward scalability and adaptability are essential for intelligent robots. Here, we report a brain-inspired robotic platform implemented by an unmanned bicycle that exhibits scalability of network scale, quantity and diversity to handle the changing needs of different scenarios. The platform adopts rich coding schemes and a trainable and scalable neural state machine, enabling flexible cooperation of hybrid networks. In addition, an embedded system is developed using a cross-paradigm neuromorphic chip to facilitate the implementation of diverse neural networks in spike or non-spike form. The platform achieved various real-time tasks concurrently in different real-world scenarios, providing a new pathway to enhance robots' intelligence.

Impact of Vocal Fold Dehydration on Vocal Function and Its Treatment.

The change of vocal function after vocal fold dehydration due to dryness was discussed along with the treatment effect of different atomizing agents. Forty-eight staffs from The Central Hospital of Wuhan were recruited. All volunteers breathed dry air for vocal fold dehydration. After dry air inhalation, the subjects were randomly divided into four groups, with 12 cases each. Three groups were treatment groups, receiving 0.9% normal saline (IS), 5% hypertonic saline (HS) and double-distilled water (SW) atomizing inhalation therapy, respectively, and the last group was the control group without treatment. Voice data were collected for all subjects before and immediately after dry air inhalation using the Multi-Dimensional Voice Program (MDVP) system. Atomizing inhalation therapy was given 10 min after dry air inhalation, and voice data were collected using MDVP system at the following time points after atomizing inhalation treatment: 5 min, 20 min, 35 min, 50 min, 65 min, 80 min, 95 min, 110 min. In the control group, voice data were collected at the same time points and compared with those of treatment groups. The vocal function parameters collected before and after dry air inhalation as well as after treatment were subjected to test using SPSS 16.0 software. In the four groups, jitter (fundamental frequency perturbation), shimmer (amplitude perturbation), and amplitude perturbation quotient (APQ) were significantly increased after dry air inhalation (P<0.05). In IS, HS and SW groups, after atomizing inhalation treatment, there was an obvious reduction in jitter, shimmer and APQ, showing significant differences as compared with those after dry air inhalation (P<0.05). Moreover, these parameters were significantly lower than those in the control group (P<0.05). The jitter, shimmer and APQ in the IS group were significantly lower than those in the HS and SW groups (P<0.05). We are led to a conclusion: Vocal fold dehydration induced by dryness can reduce the stability of voice; such decreased voice stability can be improved by atomizing inhalation therapy; without proper treatment, voice stability caused by vocal fold dehydration cannot heal spontaneously; of three atomizing agents namely, IS, HS and SW, IS had the best treatment effect for decreased voice stability caused by vocal fold dehydration.

Pathological Pattern of Intrahepatic HBV in HCC is Phenocopied by PDX-Derived Mice: a Novel Model for Antiviral Treatment.

BACKGROUND: Hepatitis B virus (HBV) is one of the most prominent risk factors for hepatocellular carcinoma (HCC) development and virus-mediated cases represents more than 80% of HCC in East Asia, where it is endemic. Currently, the HBV status of pathological HCC is not fully clarified, especially by comparison to nontumorous tissues. Lack of clinicopathological animal models of HCC impedes clinical application of antiviral treatment in the field. MATERIALS AND METHODS: A cohort sample of 14 HCC and corresponding stroma tissues were analyzed for pathological patterns of HBV antigens using immunohistochemistry; 10 fresh primary tumor tissues were inoculated into NOD/SCID mice and risk factors for patient-derived xenograft (PDX) model were identified by the univariate F test. Consistency of HBV features and cellular biomarkers between patient tissues and tumor grafts were examined. RESULTS: In HCC, HBV surface antigen (HBsAg) was mainly absent. Only 9.9% of samples showed HBsAg positivity in the tumor tissue that was limited to benign hepatocytes. In contrast, HBV core antigen (HBcAg) exhibited positive staining in all HCC tissues, located mainly in the cytoplasm of tumor cells. Of 14 HCC cases, three were diagnosed as occult infection of HBV based on HBcAg expression. The successful rate for the PDX model was 20% (2/10). Tumor lesions on hepatic lobes of V and VI, severe liver dysfunction and higher CA125 showed p-values of 0.01, 0.035, and 0.01, respectively. HBsAg absence in original tumors of #6 and 8 patients were faithfully reproduced by engraftments. Mixed distribution of HBcAg in cellular compartments of original tumor cells was also observed in mice. ki67 was dramatically increased in tumor grafts. CONCLUSION: We delineated pathological HBV profiles of HCC specimens and perilesional areas, which provided evidence for virus-based therapy in the future. PDX mice may phenocopy virological and cellular features of patient tissues, which is novel in the virus-related hepatocarcinogenesis field.

A simple and rapid fluorimetric method for simultaneous determination of protoporphyrin IX and zinc protoporphyrin IX in whole blood.

Derivative variable-angle synchronous fluorescence (DVASF) spectrometry improves the spectral resolution and selectivity of the fluorescence method. The feasibility of DVASF spectrometry for the simultaneous determination of protoporphyrin IX (PP) and zinc protoporphyrin IX (ZnPP) was investigated. PP and ZnPP were distinguished from each other simultaneously and rapidly by the DVASF method. The spectra were resolved well, and the two components were determined in a single scan, avoiding the spectral compensation factor for PP and chromatographic separation. The linear range of the calibration curve for PP was from 0.190 to 152 nmol/L and for ZnPP was from 0.383 to 230 nmol/L. The detection limits of PP and ZnPP were 0.098 nmol/L and 0.088 nmol/L, respectively. The within-run imprecision (RSD, n = 5) for PP was 4.1%, and for ZnPP was 4.2%. Mean recoveries (SD) of PP and ZnPP added to a blood sample were 86.4 (7.3)% and 72.9 (6.6)%, respectively. This method should be a potential tool in the rapid routine screening of large quantities of samples.

Fast Object Tracking on a Many-Core Neural Network Chip.

Fast object tracking on embedded devices is of great importance for applications such as autonomous driving, unmanned aerial vehicle, and intelligent monitoring. Whereas, most of previous general solutions failed to reach this goal due to the facts that (i) high computational complexity and heterogeneous operation steps in the tracking models and (ii) parallelism-limited and bloated hardware platforms (e.g., CPU/GPU). Although previously proposed devices leverage neural dynamics and near-data processing for efficient tracking, their flexibility is limited due to the tight integration with vision sensor and the effectiveness on various video datasets is yet to be fully demonstrated. On the other side, recently the many-core architecture with massive parallelism and optimized memory locality is being widely applied to improve the performance for flexibly executing neural networks. This motivates us to adapt and map an object tracking model based on attractor neural networks with continuous and smooth attractor dynamics onto neural network chips for fast tracking. In order to make the model hardware friendly, we add local-connection restriction. We analyze the tracking accuracy and observe that the model achieves comparable results on typical video datasets. Then, we design a many-core neural network architecture with several computation and transformation operations to support the model. Moreover, by discretizing the continuous dynamics to the corresponding discrete counterpart, designing a slicing scheme for efficient topology mapping, and introducing a constant-restricted scaling chain rule for data quantization, we build a complete mapping framework to implement the tracking model on the many-core architecture. We fabricate a many-core neural network chip to evaluate the real execution performance. Results show that a single chip is able to accommodate the whole tracking model, and a fast tracking speed of nearly 800 FPS (frames per second) can be achieved. This work enables high-speed object tracking on embedded devices which normally have limited resources and energy.