The investigation of constraints in implementing robust AI colorectal polyp detection for sustainable healthcare system.

Colorectal cancer (CRC) is one of the significant threats to public health and the sustainable healthcare system during urbanization. As the primary method of screening, colonoscopy can effectively detect polyps before they evolve into cancerous growths. However, the current visual inspection by endoscopists is insufficient in providing consistently reliable polyp detection for colonoscopy videos and images in CRC screening. Artificial Intelligent (AI) based object detection is considered as a potent solution to overcome visual inspection limitations and mitigate human errors in colonoscopy. This study implemented a YOLOv5 object detection model to investigate the performance of mainstream one-stage approaches in colorectal polyp detection. Meanwhile, a variety of training datasets and model structure configurations are employed to identify the determinative factors in practical applications. The designed experiments show that the model yields acceptable results assisted by transfer learning, and highlight that the primary constraint in implementing deep learning polyp detection comes from the scarcity of training data. The model performance was improved by 15.6% in terms of average precision (AP) when the original training dataset was expanded. Furthermore, the experimental results were analysed from a clinical perspective to identify potential causes of false positives. Besides, the quality management framework is proposed for future dataset preparation and model development in AI-driven polyp detection tasks for smart healthcare solutions.

Geographical origin traceability of rice using a FTIR-based metabolomics approach.

Infrared spectroscopy is a crucial tool to achieve the origin traceability of rice, but it is constrained by data mining. In this study, a novel infrared spectroscopy-based metabolomics analytical method was proposed to discriminate rice products from 14 Chinese cities by seeking 'wave number markers'. Principal component analysis (PCA), cluster analysis and orthogonal partial least squares discriminant analysis (OPLS-DA) were employed to separate all rice groups. The S-plot, permutation test and variable importance in projection (VIP) are used to screen eligible 'markers', which were further verified by a pairwise t-test. There are 55-265 'markers' picked out from 14 rice groups, with their characteristic wave number bands to be 2935.658-3238.482, 3851.846-4000.364, 3329.136-3518.160, 1062.778-1213.225, 1161.147-1386.819, 3348.425-3560.594, 3115.038-3624.245, 2567.254-2872.007, 3334.923-3560.594, 3282.845-3543.235, 3338.780-3518.160, 3197.977-3560.594, 3163.258-3267.414 and 3292.489-3477.655 cm-1, respectively. All but No. 5 rice groups show significantly low absorbance on their 'marker' bands. A mixed rice containing congenial No. 5 and No. 6 rice (80 : 20, m/m) was employed to test the validity of the method, and found that the 'marker' band of the mixed rice is the range of 1170.791-1338.598 cm-1, implying the existence of considerable discrepancy between the mixed rice and other rice. The results indicate that infrared spectroscopy coupled with metabolomics analysis is competent for origin traceability of rice; thus, it provides a novel and workable approach for the accurate and rapid discrimination of rice from different geographical origins, and a distinctive perspective of metabolomics to explore infrared spectroscopy and beyond, especially not confined in the field of origin traceability.

Prediction and Construction of Energetic Materials Based on Machine Learning Methods.

Energetic materials (EMs) are the core materials of weapons and equipment. Achieving precise molecular design and efficient green synthesis of EMs has long been one of the primary concerns of researchers around the world. Traditionally, advanced materials were discovered through a trial-and-error processes, which required long research and development (R&D) cycles and high costs. In recent years, the machine learning (ML) method has matured into a tool that compliments and aids experimental studies for predicting and designing advanced EMs. This paper reviews the critical process of ML methods to discover and predict EMs, including data preparation, feature extraction, model construction, and model performance evaluation. The main ideas and basic steps of applying ML methods are analyzed and outlined. The state-of-the-art research about ML applications in property prediction and inverse material design of EMs is further summarized. Finally, the existing challenges and the strategies for coping with challenges in the further applications of the ML methods are proposed.

HLB-MCX-Based Solid-Phase Extraction Combined with Liquid Chromatography-Tandem Mass Spectrometry for the Simultaneous Determination of Four Agricultural Antibiotics (Kasugamycin, Validamycin A, Ningnanmycin, and Polyoxin B) Residues in Plant-Origin Foods.

An ultrahigh-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) method was established for the determination of four highly polar agricultural antibiotics kasugamycin, validamycin A, ningnanmycin, and polyoxin B in plant-derived foods. The samples were extracted with a 0.2% formic acid solution, purified by hydrophilic-lipophilic balance and mixed-mode cation-exchange solid-phase extraction, and then reconstituted for UPLC-MS/MS detection. The chromatographic analysis was performed on a BEH Amide column (100 mm × 2.1 mm, 1.7 µm) using gradient elution with a 0.1% formic acid solution and 0.1% formic acid acetonitrile as mobile phases. Method validation was performed on 15 matrices spiked at 0.02 (or 0.05), 0.5, and 2 mg/kg. The mean recovery rate ranged from 75 to 102% with relative standard deviations (RSD) was less than 20%. Good linearities (r > 0.99) in the range of 0.002-0.2 µg/mL were obtained. The limits of quantification (LOQs) were 0.02 and 0.05 mg/kg. Studies on the stability of the analytes in the stored kiwifruit samples showed that kasugamycin, validamycin A, and ningnanmycin were stable for at least 6 months, while polyoxin B was observed to be partially degraded (the degradation rate at 6 months was 31.3%). The method was demonstrated to be effective and reliable in real samples. In the kiwifruit samples treated after 7 days, no residues of ningnanmycin and polyoxin B were detected, while the residues of kasugamycin and validamycin A were 0.12 and 0.038 mg/kg, respectively.

[Measurement of 11 benzophenone ultraviolet-filters in cosmetics by high performance liquid chromatography].

A sample preparation and analytical method with liquid-liquid extraction (LLE) and high performance liquid chromatography (HPLC) was developed to detect 11 benzophenone ultraviolet-filters in cosmetics. The target compounds were extracted by the mixed solutions of tetrahydrofuran (TH)/methanol/water or dichloromethane/water at proper ratios. The extracts were centrifuged and filtered to remove matrix compounds, and then analyzed by HPLC. The separation of analytes was carried out on a Diamonsil-C18 column (150 mm x 4.6 mm, 5 µm) with 0.1% (v/v) formic acid aqueous solution (containing 10 mmol/L ammonium acetate) as mobile phase A and methanol containing 0.1% (v/v) formic acid as mobile phase B. The spiked recoveries of the method (n = 7) were 93.4%-103.8% with the relative standard deviations of 0.1%-4.2%. The limits of detection (LODs) were in the range of 4.0-30 µg/g and the limits of quantitation (LOQs) ranged from 15 to 100 µg/g. The method was applied to the determination of 42 cosmetic samples randomly purchased from the supermarket in Dalian. Five benzophenone series were always detected, in which the content of benzophenone-3 in sunscreen cream and the content of benzophenone-2 in perfume were very high and reached 2 785 µg/g and 2 106 µg/g, respectively. The results showed that the developed method is efficient, reliable and sensitive, which can be applied to the determination of benzophenones in cosmetics.

[Chemical constituents from seeds of Sophora alopecuroides].

OBJECTIVE: To study the chemical constituents of Sophora alopecuroides seeds. METHODS: Five compounds were isolated and purified from this plant by various chromatographic methods. Structures were elucidated on the basis of spectral analysis. RESULTS: The structures were identified as 5"-methoxy-6-C-beta-D-glucopyranosyl-6"-C-beta-D-xylopyranosyl cupressuflavone(1),5,6-dihydroxy-3,7, 3',4'-tetramethoxyflavone(2),butein(3),beta-daucosterol(4) and 3'-methoxyluteolin(5). CONCLUSION: Compounds 2,4 and 5 are isolated from Sophora alopecuroides for the first time.

[Determination of 17 ultraviolet-filters in cosmetics by high performance liquid chromatography].

An optimum HPLC method was developed for the simultaneous determination of 17 ultraviolet (UV)-filters in cosmetics, in which two UV-filters were included more than the 15 UV-filters specified in the Hygenic Standard for Cosmetics (HSC). Our research focused on the optimization of sample processing, chromatographic column, and mobile phase, in order to avoid the use of highly corrosive solvents as tetrahydrofuran (THF) and perchloric acid, which are used in the HSC method. The cosmetic sample was extracted by the mixed solution of THF/methanol/water. The target chemicals were separated on a Phenomenex PFP column (250 mm x 4.6 mm, 5 microm) in gradient elution mode using water solution (containing 0.1% formic acid) and 20% methanol solution of isopropyl alcohol (containing 0.1% formic acid) as mobile phases. The detection wavelengths were set at 311 nm and 280 nm. The results of recovery tests and samples analysis indicated that the developed method can be used to simultaneously determine the 17 UV-filters in cosmetics, accurately and reliably. It is a more universal analytical method than the HSC method because common HPLC pump without improvement to adapt to corrosive THF, can be used. At the same time, the quantitative accuracy can be improved because baseline separation of all target chemicals was achieved

Inclusion complex of butachlor with beta-cyclodextrin: characterization, solubility, and speciation-dependent adsorption.

Due to soil adsorption, higher amounts of the herbicide butachlor are necessary to achieve its herbicidal activity, hence increasing its environmental risks. In this study, the effects of beta-cyclodextrin (beta-CD) on solubility and soil adsorption of butachlor were investigated. Formation of a 1:1 stoichiometric inclusion complex between them with an apparent stability constant of 443 L mol(-1) was confirmed in the solution. Fourier transform infrared spectroscopy showed that the (N-CO) amide bond and alkyl ether moiety of butachlor molecule could enter into the cavity of beta-CD, but the double-substituted aromatic ring was excluded because it was larger size than the cavity. Significant enhancing dissolution of butachlor in the inclusion complex occurred in comparison to the free herbicide. The adsorption of butachlor on soil was reduced with an increase of beta-CD concentration because of the formation of the inclusion complex with low adsorption potency. Although the sorption distribution coefficient of complexed butachlor (i.e., butachlor/beta-cyclodextrin inclusion complex) (K(d,c) = 6.14) was about 14% of that of the free herbicide (K(d,f) = 44.54), the proportion of the adsorbed amount of complexed butachlor to the total adsorbed amount rose with the increase of beta-CD concentration. Thus, the adsorption of inclusion complex cannot be neglected in the presence of high concentrations cyclodextrins, although its water solubility was much higher than that of the free herbicide. These results indicate that beta-CD may be used as a formation additive to improve the solubility of butachlor, reduce its adsorption on soil, and increase the availability of butachlor for weeds.

Dechlorination of chloroacetanilide herbicides by plant growth regulator sodium bisulfite.

Chloroacetanilide herbicides are frequently detected in groundwater and surface waters, and pose high risks to aquatic biota. In this study, sodium bisulfite (NaHSO(3)), a plant growth regulator used in China, was used to remove three chloroacetanilide herbicides including alachlor, acetochlor and S-metolachlor. These herbicides were rapidly dechlorinated by NaHSO(3) in neutral conditions. The dechlorination was accelerated with increasing pH, temperature and NaHSO(3) concentrations. Kinetic analysis and mass spectrum identification revealed that the reaction followed S(N)2 nucleophilic substitution, in which the chlorine was replaced by the reactive specie sulfite. Alachlor and its isomer acetochlor had similar reaction rates, whereas they were more readily transformed than S-metolachlor that had larger steric hindrance and weaker electrophilicity. The transformation products were chloroacetanilide ethane sulfonic acids (ESAs), which were also encountered as major metabolites of these herbicides in natural environment via common metabolic pathways and were less toxic to green algae compared to the parent herbicides. These results indicate that NaHSO(3) can accelerate transformation of chloroacetanilide herbicides to the less toxic transformation products by nucleophilic substitution and dechlorination in aquatic environment. NaHSO(3) can be potentially used for the removal of chloroacetanilide herbicides from wastewater effluent, spill sites and accidental discharge.

Distribution of PAHs in pine (Pinus thunbergii) needles and soils correlates with their gas-particle partitioning.

Pine (Pinus thunbergii) needles and surface soils were simultaneously sampled at 35 sites across Liaoning province, China, to investigate the distribution of polycyclic aromatic hydrocarbons (PAHs) in the two media. We hypothesized that the distribution of PAHs in soils and pine needles was related to the subcooled liquid vapor pressure (p(o)L) and the gas-particle partition coefficient (Kp), since soils accumulate PAHs mainly through dry/wet deposition of particles and pine needles sequester PAHs mainly from the gas phase, and the same physicochemical properties (e.g., p(o)L) determine the characteristics of PAHs deposition to soils and to needles. To verify the hypothesis, a soil-pine needle quotient (Q(SP)) was defined, which is a dimensionless ratio of PAH concentrations in soils and pine needles. A significant relationship between logQ(SP) and log p(o)L was observed (r = 0.94), and the variation of the regression parameters of log Q(SP)-log p(o)L and log Kp-log p(o)L relationships was similar. An adjusted soil-pine needle quotient (Q'(SP)) was defined by deducting the contributions of particle PAHs to pine needles and the vapor PAHs to soils. Log Q'(SP) correlated with log p(o)L and logarithm of the particle to gas ratio (log Cp/C(A)) more evidently than log Q(SP). In addition, Q(SP) (and Q'(SP)) could be used to characterize the removal factors of PAHs during atmospheric transport. All the observations proved that Q(SP) (and Q'(SP)) carry the information of gas-particlepartitioning and correlate with p(o)L.

Combined experimental and theoretical study on photoinduced toxicity of an anthraquinone dye intermediate to Daphnia magna.

The toxicity of chemicals can be enhanced by light through two photochemical pathways: Photomodification to more toxic substances and photosensitization. In the present study, the reactive oxygen species (ROS) mechanism for photoinduced acute toxicity of 1-amino-2,4-dibromoanthraquinone (ADBAQ) to Daphnia magna was clarified by experiment and theoretical calculation. The results of the present study show that ADBAQ exhibited high toxicity to D. magna under simulated solar radiation (SSR), with a median effective concentration of 1.23 +/- 0.19 nM (mean +/- standard deviation). The photomodified ADBAQ (mixtures of ADBAQ and its photoproducts) was less phototoxic than the intact ADBAQ. The SSR-only or ADBAQ-only treatments did not affect the ROS level in D. magna, whereas increased ROS levels were observed in the presence of SSR and ADBAQ. The ROS in vivo were determined by measuring the fluorescence of 2',7'-dichlorofluorescein, which is a useful technique to assess toxicity of chemicals to aquatic organisms. The antioxidants, including vitamin C, vitamin E, and beta-carotene, decreased the photoinduced oxidative damage to D. magna, probably by scavenging ROS. These experimental results demonstrate that photosensitization is the potential mechanism of photoinduced toxicity of ADBAQ to D. magna. Proposed phototoxic pathways of ADBAQ were elucidated by means of time-dependent density functional theory. The theoretical calculation indicates that superoxide anion and singlet oxygen are able to be generated through electron transfer or energy transfer in the photosensitization reactions.