A dynamic approach to support outbreak management using reinforcement learning and semi-connected SEIQR models.

BACKGROUND: Containment measures slowed the spread of COVID-19 but led to a global economic crisis. We establish a reinforcement learning (RL) algorithm that balances disease control and economic activities. METHODS: To train the RL agent, we design an RL environment with 4 semi-connected regions to represent the COVID-19 epidemic in Tokyo, Osaka, Okinawa, and Hokkaido, Japan. Every region is governed by a Susceptible-Exposed-Infected-Quarantined-Removed (SEIQR) model and has a transport hub to connect with other regions. The allocation of the synthetic population and inter-regional traveling is determined by population-weighted density. The agent learns the best policy from interacting with the RL environment, which involves obtaining daily observations, performing actions on individual movement and screening, and receiving feedback from the reward function. After training, we implement the agent into RL environments describing the actual epidemic waves of the four regions to observe the agent's performance. RESULTS: For all epidemic waves covered by our study, the trained agent reduces the peak number of infectious cases and shortens the epidemics (from 165 to 35 cases and 148 to 131 days for the 5th wave). The agent is generally strict on screening but easy on movement, except for Okinawa, where the agent is easy on both actions. Action timing analyses indicate that restriction on movement is elevated when the number of exposed or infectious cases remains high or infectious cases increase rapidly, and stringency on screening is eased when the number of exposed or infectious cases drops quickly or to a regional low. For Okinawa, action on screening is tightened when the number of exposed or infectious cases increases rapidly. CONCLUSIONS: Our experiments exhibit the potential of the RL in assisting policy-making and how the semi-connected SEIQR models establish an interactive environment for imitating cross-regional human flows.

Feature-aware unsupervised lesion segmentation for brain tumor images using fast data density functional transform.

We demonstrate that isomorphically mapping gray-level medical image matrices onto energy spaces underlying the framework of fast data density functional transform (fDDFT) can achieve the unsupervised recognition of lesion morphology. By introducing the architecture of geometric deep learning and metrics of graph neural networks, gridized density functionals of the fDDFT establish an unsupervised feature-aware mechanism with global convolutional kernels to extract the most likely lesion boundaries and produce lesion segmentation. An AutoEncoder-assisted module reduces the computational complexity from [Formula: see text] to [Formula: see text], thus efficiently speeding up global convolutional operations. We validate their performance utilizing various open-access datasets and discuss limitations. The inference time of each object in large three-dimensional datasets is 1.76 s on average. The proposed gridized density functionals have activation capability synergized with gradient ascent operations, hence can be modularized and embedded in pipelines of modern deep neural networks. Algorithms of geometric stability and similarity convergence also raise the accuracy of unsupervised recognition and segmentation of lesion images. Their performance achieves the standard requirement for conventional deep neural networks; the median dice score is higher than 0.75. The experiment shows that the synergy of fDDFT and a naïve neural network improves the training and inference time by 58% and 51%, respectively, and the dice score raises to 0.9415. This advantage facilitates fast computational modeling in interdisciplinary applications and clinical investigation.

Lightweight Deep Neural Network Embedded with Stochastic Variational Inference Loss Function for Fast Detection of Human Postures.

Fusing object detection techniques and stochastic variational inference, we proposed a new scheme for lightweight neural network models, which could simultaneously reduce model sizes and raise the inference speed. This technique was then applied in fast human posture identification. The integer-arithmetic-only algorithm and the feature pyramid network were adopted to reduce the computational complexity in training and to capture features of small objects, respectively. Features of sequential human motion frames (i.e., the centroid coordinates of bounding boxes) were extracted by the self-attention mechanism. With the techniques of Bayesian neural network and stochastic variational inference, human postures could be promptly classified by fast resolving of the Gaussian mixture model for human posture classification. The model took instant centroid features as inputs and indicated possible human postures in the probabilistic maps. Our model had better overall performance than the baseline model ResNet in mean average precision (32.5 vs. 34.6), inference speed (27 vs. 48 milliseconds), and model size (46.2 vs. 227.8 MB). The model could also alert a suspected human falling event about 0.66 s in advance.

Stable carbon isotopic characterization of rice vinegar protein as an intrinsic reference for discriminating the authenticity of brewed rice vinegar.

Rice vinegar plays an important role in daily life. However, some unscrupulous manufacturers may deliberately add synthetic acetic acid in vinegar products to reduce fermentation time and save production costs. To protect the rights and health of consumers, vinegar authenticity must be controlled. The rice vinegar protein was used as an intrinsic reference and its stable carbon isotope ratio (δ13Cprotein) was analyzed by elemental analyzer-isotope ratio mass spectrometry. The stable carbon isotope ratio difference between the acetic acid and the rice vinegar protein (Δδ13Cacetic acid-protein) was calculated to evaluate vinegar authenticity. Sixteen rice vinegar samples were analyzed and a stable carbon isotopic pattern of rice vinegar was established by the 95% confidence interval for Δδ13Cacetic acid-protein (0.27‰-2.10‰). An acetic acid adulteration curve of Δδ13Cacetic acid-protein was also assumed according to the data from rice vinegar samples, and its validity was confirmed by rice vinegar deliberately blended with acetic acid at different ratios (25, 50, and 75%). The Δδ13Cacetic acid-protein values of the adulterated vinegars decreased with increasing amounts blended acetic acid, but the δ13Cprotein values did not, showing that rice vinegar protein could be used as an intrinsic reference for identifying the adulterated rice vinegar. The rice vinegar adulterated with acetic acid at higher than approximately 10% could be detected.

Inorganic arsenic speciation analysis in food using HPLC/ICP-MS: Method development and validation.

Arsenic (As) compounds can be classified as organic or inorganic, with inorganic arsenic (iAs) having significantly higher toxicity than organic As. As may accumulate in food materials that have been exposed to As-contaminated environments. Thus, the "Sanitation Standard for Contaminants and Toxins in Foods" published by the Ministry of Health and Welfare set the standard limits for iAs content in rice, seaweed, seafood, and marine oils to safeguard public health. Therefore, a robust analytical method must be developed to selectively and quantitatively determine iAs content in rice, seaweed, seafood, and marine oils. Herein, we reported and verified the method of combined high-performance liquid chromatography/inductively coupled plasma-mass spectrometry (HPLC/ICP-MS) to determine iAs content in a wide variety of food. The fish oil samples were spiked with different concentrations of the As(III) standard solution, and their iAs analyzes were obtained via extraction procedures using the 1% (w/w) nitric acid (HNO3) solution containing 0.2 M hydrogen peroxide (H2O2) under sonication. The extracts were subsequently analyzed for their As(V) contents using HPLC/ICP-MS with aqueous ammonium carbonate as the mobile phase. The As(III) species had completely oxidized into the As(V) species, which prevented interferences between organic and iAs during chromatography. The method showed good extraction efficiencies (generally >90%) for the iAs samples, and their limits of quantification in fish oil were 0.02 mg/kg. The method was verified via the iAs speciation analytes of rice, seaweed, seafood, and marine oil matrices. The average recoveries for the fortified samples of each matrix ranged from 87.5 to 112.4%, with their coefficients of variation being less than 10%. Surveillance studies were conducted on the iAs contents of food samples purchased from local Taiwanese markets. The results showed that the only Hijiki (Sargassum fusiforme) higher than the maximum limit of the sanitation standard for iAs in seaweed, whereas the remaining samples met their corresponding requirements. This method is quick and straightforward, and it can be applied for the routine analysis of iAs content in a wide variety of food products to ensure public health safety.

Determination of cannabinoids in hemp oil based cosmetic products by LC-tandem MS.

Cosmetic products containing hemp seed oil as permitted raw materials required the specific compound delta-9-tetrahydrocannabinol (THC) below 10 µg/g. THC was the main psychoactive constituent of cannabis. Since hemp seed oil became an increasingly popular ingredient in cosmetics over the last few years, an efficient and reliable analytical method for THC and other cannabinoids in cannabis-infused cosmetic products was in need. A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for the determination of delta-9-tetrahydrocannabinol (THC), cannabidiol (CBD), and cannabinol (CBN) in hemp seed oil based cosmetic products was developed. Method validation was performed by fertilizing blank samples with analytes and internal standards (THC-d3, CBD-d3, and CBN-d3). Chromatographic method utilized a Xbridge BEH Shield RP18 column with gradient elution containing 10 mM ammonium formate in water and methanol provided successful separation of THC, CBD, and CBN in cosmetic matrix. The combination of MS detection in positive electrospray ionization (ESI) and multiple reaction monitoring (MRM) mode offered rapid run time 13 minutes with limit of quantification (LOQ) of 0.05 µg/g. The intra- and inter-day recoveries were 79.23-114.04% and 83.55-111.61% with spiking levels ranged between 0.05 µg/g and 0.5 µg/g, respectively. Surveillance results of 90 cosmetic products showed 22, 34, and 5 products containing THC (0.06-1777 µg/g), CBD (0.47-37217 µg/g), and CBN (2.2-25.2 µg/g), respectively. This validated method offered accurate, reliable, and fast way for the determination of drug contaminations including THC, CBD, and CBN in cosmetics. The surveillance results for commercial cosmetic products purchased in Taiwan between 2018-2020 provided valuable background references for THC, CBD, and CBN in hemp seed oil based cosmetic products, and could be used for administration purpose.

Simultaneous Analysis of the Stable Carbon Isotope Ratios of Acetoin and Acetic Acid by GC-C-IRMS for Adulteration Detection in Brewed Rice Vinegar Products.

In this study, we developed a method to simultaneously measure the stable carbon isotope ratio for acetic acid (δ 13Cacetic acid) and acetoin (δ13Cacetoin) in rice vinegar by gas chromatography-combustion-isotope ratio mass spectrometry. The method showed good precision and accuracy. With this method, data from 16 brewed rice vinegars and 10 acetic acid samples were used to evaluate the feasibility of adulteration detection. On the basis that all δ13Cacetoin values of brewed rice vinegars are nearly constant, a characteristic pattern of the stable carbon isotope in rice vinegar was built with the 95% confidence intervals for δ13Cacetic acid (-26.97 to -25.38‰), δ13Cacetoin (-28.14 to -27.09‰), and Δδ13C (0.61 to 2.27‰). An adulteration detection curve of Δδ13C was proposed based on the results of vinegar and acetic acid samples and confirmed by vinegar spiked with different amounts of acetic acid. This method could be useful in estimating the blending ratio of adulterated rice vinegar products. Products containing more than 10% of synthetic acetic acid could be possibly identified.

Novel multi-analyte method for detection of thirty photoinitiators in breakfast cereal and packaged juice.

Multilayer print designs are commonly used in commercial food packaging to attract consumers. UV-curable ink is generally used in this type of printing due to its ease of application, space saving, and rapid drying; however, there have been a number of health alerts related to the contamination of food by photoinitiators in UV-curable ink. In this study, we established a multi-analyte method by which to detect 30 photoinitiators simultaneously. We then applied this method to the analysis of five breakfast cereals and ten types of packaged juice to detect the presence of photoinitiator contamination. Sample treatment was performed using the QuEChERS (quick, easy, cheap, effective, rugged, and safe) method for the extraction of photoinitiators. Chromatographic separation of two isomers, methylbenzophenone (MBP) and isopropylthioxanthone (ITX), was achieved using a pentafluorophenyl propyl (PFP) column (1.7 µm, 100 × 2.1 mm i.d.) and MeOH: 5 mM formic acid-ammonium formate (pH 4.0) in gradient elution. The average recovery of photoinitiators from cereal was between 62.0 and 120.3%, with a coefficient of variation between 0.4 and 14.4%. The average recovery of photoinitiators from packaged juices was between 84.4 and 122.9% with a coefficient of variation between 0.5 and 9.5%. The contamination results were as follows: 13.1 ng/g triphenyl phosphate (TPP) was detected in one breakfast cereal, and 2-hydroxy-4-methoxy benzophenone (BP-3), 1-hydroxycyclohexyl phenyl-ketone (Irgacure 184), methyl-2-benzoylbenzoate (MOBB), and 2,4-diethyl-9H-thioxanthen-9-one (DETX) were detected in one of the packaged juices at levels ranging from 2.2 to 152.9 ng/g.

Multi-residue analysis using liquid chromatography tandem mass spectrometry for detection of 20 coccidiostats in poultry, livestock, and aquatic tissues.

In this study, we developed a novel analysis method based on liquid chromatography/tandem mass spectrometry (LC-MS/MS) to allow the simultaneous identification of 20 coccidiostats in eight matrix categories, including the muscles of chicken, swine, cow, and fish as well as chicken eggs, bovine milk, and porcine viscera. In the pretreatment procedure, acetonitrile/methanol (95:5, v/v) containing 1% formic acid, 5 g of sodium acetate, and 6.0 g of anhydrous magnesium sulfate was used for extraction, followed by a clean-up procedure using n-hexane saturated with ACN to facilitate the elimination of analytes from high lipid samples. Chromatographic separations were achieved using a Poroshell 120SB C18 column and operated with a gradient mobile phase system consisting of methanol (with 0.1% formic acid) and 5 mM ammonium formate, and the MS detection was monitored simultaneously. The method was validated in accordance with the Guidelines for the Validation of Food Chemical Methods by the Taiwan Food and Drug Administration. The limit of quantitation among 8 matrices were 0.5-2 ng g-1. The proposed method proved highly effective in detecting the presence of targeted veterinary drugs, providing a high degree of precision and accuracy over a broad range of matrices.

Identification of 3-MCPD esters to verify the adulteration of extra virgin olive oil.

The adulteration of olive oil is an important issue around the world. This paper reports an indirect method by which to identify 3-monochloropropane-1,2-diol (3-MCPD) esters in olive oils. Following sample preparation, the samples were spiked with 1,2-bis-palmitoyl-3-chloropropanediol standard for analysis using gas chromatograph-tandem mass spectrometry. The total recovery ranged from 102.8% to 105.5%, the coefficient of variation ranged from 1.1% to 10.1%, and the limit of quantification was 0.125 mg/kg. The content of 3-MCPD esters in samples of refined olive oil (0.97-20.53 mg/kg) exceeded those of extra virgin olive oil (non-detected to 0.24 mg/kg). These results indicate that the oil refining process increased the content of 3-MCPD esters, which means that they could be used as a target compound for the differentiation of extra virgin olive oil from refined olive oil in order to prevent adulteration.

Temporal trend of polychlorinated dibenzo-p-dioxin/polychlorinated dibenzofuran and dioxin like-polychlorinated biphenyl concentrations in food from Taiwan markets during 2004-2012.

The levels of polychlorinated dibenzo-p-dioxin (PCDD) and polychlorinated dibenzofuran (PCDF) or polychlorinated biphenyl (PCB) in foodstuffs have decreased over the past decade in many countries. However, the trend for the levels of these compounds in foodstuffs in Taiwan remains unknown. In this study, we compared the distribution of PCDD/F and PCB in nine foodstuff categories acquired from Taiwan markets from 2004 to 2012. The levels expressed as World Health Organization toxic equivalents (WHO-TEQs) in the different foodstuffs tested were as follows: fish, average 0.463 pg WHO98-TEQ/g sample > seafood, 0.163 pg WHO98-TEQ/g > eggs, 0.150 pg WHO98-TEQ/g > oils, 0.126 pg WHO98-TEQ/g > meats, 0.095 pg WHO98-TEQ/g > dairy products, 0.054 pg WHO98-TEQ/g > cereals, 0.017 pg WHO98-TEQ/g > vegetables, 0.013 pg WHO98-TEQ/g > fruits, 0.009 pg WHO98-TEQ/g. Levels were particularly high in crab (average: 0.6 pg WHO98-TEQ/g sample (1.243 pg WHO98-TEQ/g sample) and large marine fish (0.6). In Taiwan, a decreasing trend of PCDD/Fs or dioxin-like PCBs (dl-PCBs) was observed in meat, dairy, eggs, and vegetables, whereas an elevated trend was observed in cereals or the levels were nearly equal in fruits and oils at alternative time shift. Dl-PCBs contributed to 60-65% toxicity equivalence levels in fish and seafood, but only to 13-40% in meat and cereal samples. The decreasing trend was consistent with the results in other countries; however, the trends in cereals, fruits, and oils were in contrast to previous results reported in other countries. Cereals and fruits are important crops in southern Taiwan, and the local pollution generated by industries or incinerators may seriously affect the distribution of PCDD/Fs and dl-PCBs. To ensure food safety, a risk assessment for residents living in different areas should be adopted for all food categories simultaneously in the future.

Identification and quantification of Cu-chlorophyll adulteration of edible oils.

Cu-pyropheophytin a, the major Cu-pigment of Cu-chlorophyll, was determined in edible oil by high-resolution mass spectrometry with a high-performance liquid chromatography-quadrupole (HPLC-Q)-Orbitrap system and by HPLC coupled with a photodiode-array detector. Respective limit of detection and limit of quantification levels of 0.02 µg/g and 0.05 µg/g were obtained. Twenty-nine commercial oil products marked as olive oil, grapeseed oil and blended oil, all sourced directly from a food company that committed adulteration with Cu-chlorophyll, were investigated. In this company, four green dyes illegally used in oils were seized during factory investigation by the health authorities. The food additive Cu-pyropheophytin a was found in all confiscated samples in concentrations between 0.02 and 0.39 µg/g. Survey results of another 235 commercial oil samples manufactured from other companies, including olive pomace oil, extra virgin olive oil, olive oil, grapeseed oil and blended oil, indicated high positive incidences of 63%, 39%, 44%, 97% and 8%, respectively, with a concentration range between 0.02 and 0.54 µg/g. High Cu-chlorophyll concentrations are indications for fraudulent adulteration of oils.

Effective extraction method through alkaline hydrolysis for the detection of starch maleate in foods.

A high-performance liquid chromatography (HPLC) method was developed for the determination of maleic acid which was released from starch maleate (SM) through the alkaline hydrolysis reaction. The proper alkaline hydrolysis conditions and LC separation are reported in this study. The starch samples were treated with 50% methanol for 30 minutes, and then hydrolyzed by 0.5N KOH for 2 hours to release maleic acid. A C18 column and gradient mobile phase consisting of 0.1% phosphoric acid and methanol at a flow rate of 1.0 mL/minute were used for separation. The method showed a good linearity in the range of 0.01-1.0 ìg/mL, with a limit of quantification (LOQ) at 10 mg/kg in starch. The recoveries in corn starch, noodle, and fish balls were between 93.9% and 108.4%. The relative standard deviation (RSD) of precision was <4.9% (n = 3). This valid method was rapid, sensitive, precise, and suitable for routine monitoring of the illegal adulteration of SM in foods.

Trace elements and heavy metals in poultry and livestock meat in Taiwan.

Trace elements and heavy metals were determined in poultry and livestock meat by inductively coupled plasma mass spectrometry after graphite block digestion. Samples were collected from markets in Taiwan and included beef (20), mutton (20), pork (30), chicken (30), duck (10) and goose (10). Mean concentrations of Mn, Co, As, Se, Mo, Cd, Sb and Pb in meat samples were 0.106-0.365 µg/g, 0.002-0.033 µg/g, 0.005-0.035 µg/g, 0.108-0.349 µg/g, 0.029-0.140 µg/g, <0.002-0.003 µg/g, <0.002-0.004 µg/g and 0.009-0.046 µg/g, respectively. Arsenic concentrations in both pork and chicken were higher than in other meat. Lead concentrations were higher in duck. Risk assessment from these data did not indicate harm for public health.

Determination of urine melamine by validated isotopic ultra-performance liquid chromatography/tandem mass spectrometry.

Little is known about melamine (MEL) analysis in children's urine. In this study, an isotopic ultra-performance liquid chromatography/tandem mass spectrometry (UPLC/MS/MS) method was developed and systematically validated for the analysis of MEL in urine. The method is easily performed and comprises acidification, solid-phase extraction (SPE) and UPLC/MS/MS analysis. (13)C(3)N(3)((15)NH(2))(3) was used as the internal standard (IS) for calibration. Transition ions m/z 127 > 85 of MEL and m/z 133 > 89 of the IS were used for quantification and m/z 127 > 68 of MEL was used for quantitative confirmation. Recovery and precision were assessed to guarantee the applicability of the method. The limit of quantification (LOQ) was 0.01 microg/mL while the calculated method detection limit was 0.006 microg/mL. The mean recoveries ranged from 96-99%. The method was then applied to analyze urine samples from children who had potentially consumed MEL-tainted dairy products during screening in Taiwan. Ten nephrolithiasis cases and 20 age- and gender-matched controls were selected for this study. Three out of the 10 nephrolithiasis cases had elevated levels of MEL. Comparatively, twenty age- and gender-matched non-nephrolithiasis controls consuming Taiwan brand milk powder all showed MEL levels lower than the detection limit except for two children with background levels of 0.02 microg/mL. The background level in these children urine samples was established by UPLC/MS/MS analysis. Positive results of urine MEL tests might be associated with nephrolithiasis in these candidates. Measurement of urine MEL concentration can be helpful in confirming MEL-related nephrolithiasis, but its clinical application needs further clarification.

Condom use by women recently diagnosed with a sexually transmitted infection: effects of study methodology on the apparent influence of hormonal/surgical contraception.

OBJECTIVE/GOAL: The objective of this study was to investigate potential predictors of consistent condom use (CCU), including the influence of hormonal contraception/surgical sterilization (HC/SS). STUDY: Regression methods were used to predict CCU and other measures of CU among 214 sexually active, 18- to 45-year-old women previously diagnosed with a sexually transmitted infection. RESULTS: CCU was significantly associated with younger age, African American ethnicity, having casual partners, recent HIV testing, condom use self-efficacy, and concern about partner relationship. HC/SS was not significantly associated with the likelihood of CCU, before (HC/SS, 21.3%, non-HC/SS, 25.3%; odds ratio [OR], 0.798; P=0.4914) or after (OR, 1.209; P=0.5995) controlling for confounders (age, ethnicity, casual partners). Controlling for age and ethnicity eliminated initial significant or near-significant inverse associations between HC/SS and 3 alternative measures of interval condom use ("any use," "number of unprotected acts," "proportion condom-protected contacts") and substantially diminished the association between HC/SS and "condom use at last sex." CONCLUSIONS: Choice of condom use measure and control of confounding variables can substantially affect results when studying potential predictors of condom use such as HC/SS.

Correlates of sexual risk for HIV infection in female members of heterosexual California Latino couples: an application of a Bernoulli process model.

Individual HIV risk estimates were generated from reported sexual behavior for 1,146 California Latino Couples Study participants (573 couples). These Bernoulli process model-based estimates proved strongly associated with individual sexually transmitted disease history. Mean estimated background risk from sexual contacts other than with their primary partner was substantially lower for the females than the males (1.4 vs. 7.4 per 10,000). After including their chance of infection from each other, mean net estimated risk was higher for the females than the males (9.2 vs. 8.6 per 10,000). Individual background risk was predicted by individual demographic and psychosocial characteristics (females: coefficient of concordance C = 0.84 predicting any (nonzero) risk; adjusted R(2) = 36% predicting level of risk, given any risk; males: C = 0.78; R(2) = 24%). Characteristics of women with higher risk primary partners were also identifiable (C = 0.65; R(2) = 13%). There was no significant negative association between the male partner's background sexual risk and the aggregate infectivity of the woman from him (taking into account the total number of their condom-protected and unprotected acts of different types).