Effect of an artificial intelligence-assisted system on endoscopic diagnosis of superficial oesophageal squamous cell carcinoma and precancerous lesions: a multicentre, tandem, double-blind, randomised controlled trial.

BACKGROUND: Despite the usefulness of white light endoscopy (WLE) and non-magnified narrow-band imaging (NBI) for screening for superficial oesophageal squamous cell carcinoma and precancerous lesions, these lesions might be missed due to their subtle features and interpretation variations among endoscopists. Our team has developed an artificial intelligence (AI) system to detect superficial oesophageal squamous cell carcinoma and precancerous lesions using WLE and non-magnified NBI. We aimed to evaluate the auxiliary diagnostic performance of the AI system in a real clinical setting. METHODS: We did a multicentre, tandem, double-blind, randomised controlled trial at 12 hospitals in China. Eligible patients were aged 18 years or older and underwent sedated upper gastrointestinal endoscopy for screening, investigation of gastrointestinal symptoms, or surveillance. Patients were randomly assigned (1:1) to either the AI-first group or the routine-first group using a computerised random number generator. Patients, pathologists, and statistical analysts were masked to group assignment, whereas endoscopists and research assistants were not. The same endoscopist at each centre did tandem upper gastrointestinal endoscopy for each eligible patient on the same day. In the AI-first group, the endoscopist did the first examination with the assistance of the AI system and the second examination without it. In the routine-first group, the order of examinations was reversed. The primary outcome was the miss rate of superficial oesophageal squamous cell carcinoma and precancerous lesions, calculated on a per-lesion and per-patient basis. All analyses were done on a per-protocol basis. This trial is registered with the Chinese Clinical Trial Registry (ChiCTR2100052116) and is completed. FINDINGS: Between Oct 19, 2021, and June 8, 2022, 5934 patients were randomly assigned to the AI-first group and 5912 to the routine-first group, of whom 5865 and 5850 were eligible for analysis. Per-lesion miss rates were 1·7% (2/118; 95% CI 0·0-4·0) in the AI-first group versus 6·7% (6/90; 1·5-11·8) in the routine-first group (risk ratio 0·25, 95% CI 0·06-1·08; p=0·079). Per-patient miss rates were 1·9% (2/106; 0·0-4·5) in AI-first group versus 5·1% (4/79; 0·2-9·9) in the routine-first group (0·37, 0·08-1·71; p=0·40). Bleeding after biopsy of oesophageal lesions was observed in 13 (0·2%) patients in the AI-first group and 11 (0·2%) patients in the routine-first group. No serious adverse events were reported by patients in either group. INTERPRETATION: The observed effect of AI-assisted endoscopy on the per-lesion and per-patient miss rates of superficial oesophageal squamous cell carcinoma and precancerous lesions under WLE and non-magnified NBI was consistent with substantial benefit through to a neutral or small negative effect. The effectiveness and cost-benefit of this AI system in real-world clinical settings remain to be further assessed. FUNDING: National Natural Science Foundation of China, 1·3·5 project for disciplines of excellence, West China Hospital, Sichuan University, and Chengdu Science and Technology Project. TRANSLATION: For the Chinese translation of the abstract see Supplementary Materials section.

[Tracing the sources of sedimentary organic matter in Nanyue small watershed based on 13C, 15N and C/N]. / 基于13C、15N和C/N识别南岳小流域沉积有机质的来源.

Losses of organic matter in agricultural watersheds result in eutrophication and land degra-dation, which not only threaten water quality and food security, but also lead to environmental problems such as the greenhouse gases emission. We used 13C, 15N and C/N as fingerprint markers to trace the sources of sedimentary organic matter at the outlet in the Nanyue small watershed. We analyzed the spatial distribution in watershed sedimentary organic matter and soils of typical land use types, including forest, paddy field, and vegetable fields. The Bayesian stable isotope mixing model was used to quantitatively estimate the contribution of different sources. The results showed that there was significant spatial variation of δ13C. The δ13C of sediment organic matter (-22.6‰±0.53‰) and forest soil (-23.13‰±1.71‰) was significantly higher than that of paddy soil (-25.24‰±1.4‰). The differences of δ15N among the sources were not significant, with sediment having the maximum (4.37±0.83)‰ and forest soil having the minimum (2.38±1.97)‰. Forest soil had the highest C/N of 16.66±7.18, while paddy soil had the lowest C/N of 11.95±0.92. The results of the Bayesian stable isotope mixture model showed that the contribution rates of forest land, paddy fields and vegetable fields to the organic matter deposited at the outlet in the watershed were 19.6%, 15.7%, and 64.7%, respectively. Paddy filed and vegetable field had a combined contribution rate of 80.4%. It was concluded that, soils of agricultural land were the main sources of organic matter deposited in the Nanyue small watershed, and that nutrient loss in the watershed would be effectively controlled by optimizing farmland management.

[Responses of net assimilation rate to elevated atmospheric CO2and temperature at different growth stages in a double rice cropping system]. / 双季稻不同生育期净同化速率对大气CO2浓度和温度升高的响应.

Effects of elevated atmospheric CO2 concentration and temperature on rice dry matter accumulation vary in planting regions and cropping systems. It remains unclear how dry matter productivity responds to factorial combination of elevated CO2 and temperature in the double rice cropping system of China. Field experiments were conducted using open-top chambers (OTC) to simulate different scenarios of elevated CO2 and/or temperature for three rotations of double rice in Jingzhou, Hubei Province. Liangyou 287 and Xiangfengyou 9 were used as rice cultivar for early rice and late rice, respectively. There were five treatments: UC, paddy field without OTC covering; CK, OTC with the similar temperature and CO2 concentration to field environment; ET, OTC with 2 ℃ temperature elevation; EC, OTC with 60 µmol·mol-1 CO2 elevation; ETEC, OTC with simu-ltaneous 2 ℃ temperature elevation and 60 µmol·mol-1 CO2 elevation. We measured aboveground biomass, leaf area index (LAI) and net assimilation rate (NAR) of dry matter under different treatments. Our results showed that elevated CO2 and/or temperature had no significant effects on NAR from transplanting to jointing, increased NAR from jointing to heading, but decreased NAR from heading to maturity (except for EC treatment in early rice). Elevated CO2 and/or temperature promoted leaf area development at all growth stages, with ETEC showing the highest increase in LAI except at maturity. Warming and CO2 enrichment jointly promoted dry matter accumulation at heading, with ETEC increasing aboveground biomass by 10.3%-39.8% and 23.6%-34.4% compared with CK in early rice and late rice, respectively. At maturity of early rice, elevated temperature partly offset the positive effects of elevated CO2 on aboveground biomass, as shown by a reduction of 3.2%-14.1% under ETEC compared with EC. Contrarily at maturity of late rice, co-elevation of CO2 and temperature further increased aboveground biomass, showing a synergistic interaction. Results from regression analysis showed that warming and CO2 enrichment had positive effects on NAR at vegetative stages of double rice, while warming showed negative effects on NAR at reproductive stages. Considering the dissimilarities in growth characteristics, growing periods and ambient temperature, elevated CO2 and temperature might increase dry matter production in the Chinese double rice cropping system.

[Key Production Process of Nitrous Oxide and Nitrogen Sources in Tuojia River].

The nitrogen (N) pollution of water is a common global problem. To understand the key production process of N2 O and identify the dominant N sources, Tuojia River, a typical agricultural watershed in a subtropical area, was investigated. To analyze the characteristics of dual nitrate isotopes (δ15N-NO3-,δ18O-NO3-) in water, and N isotope (δ15Norg) and carbon-nitrogen ratio (C/N) in sediment organic matter from four reaches(S1-S4), the stable isotopes method was used. The results showed that the sources of nitrate varied significantly among river segments and were affected by agricultural production and human habitation on the land surface. The average δ15N-NO3- in reaches S1, S2, S3, and S4 were 1.72‰, 2.62‰, 4.10‰, and -1.28‰, respectively, while the average δ18O-NO3- were 2.60‰,-0.06‰, 0.85‰, and -0.62‰. The N in terrestrial soil made a large contribution to nitrate sources in reach S1, while soil N, ammonium N fertilizer, and manure played a main role in reaches S2 and S3. Most of the nitrate came from ammonium N fertilizer in reach S4. We also found that δ15Norg in sediment organic matter ranged from -0.69‰ to 11.21‰, and C/N was between 7.30 and 12.02. The mean δ15Norg in reaches S1-S4 were 1.91‰, 2.96‰, 4.72‰, and 3.23‰, respectively, and the mean C/N values were 10.62, 8.63, 9.05, and 9.22, respectively. Although there were some differences in δ15Norg among reaches S2-S4, the dominant N source was sewage in those reaches. However, soil organic matter was the main N source in the sediments of reach S1. The mean δ18O-NO3- in reaches S1-S4 were -7.01‰,-0.17‰,-0.28‰, and -0.60‰, respectively, indicating that nitrification was the key N2 O production process in these reaches. The ratios of δ15N-NO3- and δ18O-NO3- were 0.66,-41.01,-30.23, and 9.39 in reaches S1-S4, respectively. Finally, we found that there was a positive correlation between NO3--N and δ15N-NO3-. To summarize, the N transformation and N2 O production could be dominated by the nitrification process in Tuojia River.

[Key pathway of methane production and characteristics of stable carbon isotope of the Tuojia River waterbody.] / 脱甲河水系CH4å ³é”®äº§ç”Ÿé€”å¾„åŠå ¶ç¨³å®šç¢³åŒä½ç´ ç‰¹å¾.

This study aimed at exploring the key pathway of methane production and clarifying the composition and distribution of carbon (C) isotopes in the Tuojia River waterbody in Hunan Pro-vince. We estimated CH4 concentrations and fluxes of four reaches (S1, S2, S3 and S4) by a two-layer diffusion model and gas chromatography. The spatial and temporal distribution of CH4 flux and its relationship with environmental factors were examined. The key pathway of CH4 production was investigated by stable C isotope method to analyze the distribution characteristics of 13C isotope (δ13C) of water dissolved CH4 and seston/benthic organic matter. There was significant seasonal variability in water pH, with mean value of (7.27±0.03). The concentration of dissolved oxygen (DO) showed strong seasonal and spatial variations, with the range of 0.43-13.99 mg·L-1. The maximum value of DO occurred in S1 and differed significantly in summer and autumin. In addition, DO differed significantly in winter and other seasons in S2, S3 and S4. The concentration of dissolved organic carbon (DOC) showed a gradual increasing trend from source to estuary. The highest concentration of DOC (8.32 mg·L-1) was found in S2, while the lowest was observed in S1 (0.34 mg·L-1). The electrical conductivity (EC) and oxidation-reduction potential (ORP) of water ranged from 17 to 436 µS·cm-1 and from -52.30 to 674.10 mV, respectively, which were significantly different among the four reaches (P＜0.05). Water ammonium nitrogen (NH4+-N) and nitrate nitrogen (NO3--N) concentrations were in the ranges of 0.30-1.35 (averaged 0.90±0.10) mg·L-1 and 0.82-2.45 (averaged 1.62±0.16) mg·L-1, respectively. The dissolved concentration and diffusion flux of CH4 ranged from 0 to 5.28 µmol·L-1 and from -0.34 to 619.72 µg C·m-2·h-1, respectively, with significant temporal and spatial variations. They showed a similar trend among reaches. Their values were highest in spring, followed by in winter and lowest in summer and autumn. Spatially, the CH4 concentration and flux followed the order of S2>S3>S4>S1. The correlation analysis showed that CH4 flux was positively correlated with NH4+-N and DOC. The pathway of CH4 production of all reaches was dominated by acetic acid fermentation, while there were obvious differences among the four reaches. The contribution of CH4 from acetic acid fermentation was greatest (87%) in S1, followed by S4(81%), S2(78%) and S3(76%). The mean value of the δ13C for dissolved CH4, seston organic matter and benthic organic matter was -41.64‰±1.91‰, -14.07‰±1.06‰ and -26.20‰±1.02‰, respectively. There was a positive correlation between the δ13C of dissolved CH4 and benthic organic matter, whereas the δ13C value of dissolved CH4 was negatively correlated with CH4 flux.

[Temporal and spatial variations of hydrogen and oxygen isotopes in Tuojia River and its influencing factors.] / è„±ç”²æ²³æ°¢æ°§åŒä½ç´ ç»„åˆ†æ—¶ç©ºåˆ†å¸ƒç‰¹å¾åŠå ¶å½±å“å› ç´ .

The characteristics of hydrogen and oxygen stable isotopes in river is important for regional hydrologic cycle research. To uncover water supply sources in subtropical agricultural basin from a perspective of stable isotopes, field measurements were conducted in four reaches (S1, S2, S3 and S4) of Tuojia River from April to August 2017. We analyzed the spatial and temporal variations in hydrogen and oxygen isotopes and deuterium excess parameters and their relationship with precipitation, altitude and water quality. Results showed that hydrogen and oxygen isotopes and deuterium excess values ranged from -43.17‰ to -26.43‰ (-35.50‰±5.44‰), -7.94‰ to -5.70‰ (-6.86‰±0.74‰), and 16.77‰ to 23.49‰ (19.39‰±1.95‰), respectively. Under the influence of monsoon circulation, hydrogen and oxygen isotopes showed substantial seasonal variation, with spring (δD: -29.88‰±3.31‰; δ18O: -6.18‰±0.57‰) > summer (δD: -39.25‰±2.65‰; δ18O: -7.32‰±0.42‰). The spatial distribution of hydrogen and oxygen isotopes values increased fluctuantly with the position from the sampling site to the river's source, with δD: S1＜S4＜S3＜S2, and δ18O: S1＜S3＜S4＜S2. The deuterium excess values had no significant temporal variation, while it spatially increased gradually with the river levels. The slope and intercept of water line in this river were smaller than that of the local meteoric water line, suggesting that precipitation was the primary water source for this river. At the seasonal scale, both δD and δ18O were significantly negatively correlated with water temperature (δD: r=-0.92; δ18O: r=-0.88) and δ18O was negatively correlated with altitude (r=-0.96). At spatial scale, δ18O had a significantly positive correlation with water temperature. The δD and δ18O had negative correlation with precipitation, but being not statistically significant.

[Observation for CH4 and N2O emissions under different rates of nitrogen and phosphate fertilization in double rice fields].

Two non-CO2 greenhouse gas emissions (methane and nitrous oxide) and related environmental factors were measured within rice growing season under five treatments including non-fertilization (CK), balanced fertilization (BF), decreased nitrogen and phosphate 1 (DNP1), decreased nitrogen and phosphate 2 (DNP2) and increased nitrogen and phosphate 1 (INP) in double rice fields of red clay soil in 2009, using the method of static chamber-gas chromatograph techniques. The results showed that the average CH4 emission fluxes for treatments of BF, DNP1, DNP2 and INP were 4.57, 5.42, 4.70 and 4.65 mg x (m2 x h)(-1) during early rice growing period, which increased by 39%, 49%, 41% and 40% compared with non-fertilizer treatment, respectively. The average CH4 emission fluxes in late rice growing season was higher than preseason's. Compared to CK, CH4 emission increased by 11%, 1%, 26% and - 4% in treatments of BF, DNP1, DNP2 and INP within late rice growing season. Applying nitrogen and phosphate enhanced CH4 emission in turning green period for early and late rice. No significant difference was observed between the CH4 emissions of five treatments during early and late rice growing season (p > 0.05). N2O emission was very little during mid-seasonal drainage period. In contrast, N2O emission peaks were observed in period of alternation of wetting and drying after mid-seasonal drainage in this experiment. N2O emission was, on average, equivalent to 0.18% of the nitrogen applied in double rice growing season. Statistically, air temperature, soil Eh and soil moisture (water-filled pore space, WFPS) at 0-10cm depth significantly affected the fluctuations of the seasonal CH4 flux, but no significant correlationship has been found between N2O flux and related environmental factors. CH4 was the dominated greenhouse gas in double rice fields which contributed approximately 90% for the integrated global warming potential of CH4 and N2O released during the rice growing season. Therefore, the mitigation options should focus on how to reduce CH4 emission in local area. The result indicates that BF is a recommended fertilization method for early rice production, and a optimum fertilization for late season can increase rates of nitrogen and phosphate fertilizers on the basis of BF treatment slightly by considering total global warming potential and grain yield. The rates of BF treatment were 150-90-90 kg x hm(-2) N-P2O5-K2O for early rice, and 180-90-135 kg x hm(-2) N-P2O5-K2O for late rice, respectively.