Global soil respiration estimation based on ecological big data and machine learning model.

Soil respiration (Rs) represents the greatest carbon dioxide flux from terrestrial ecosystems to the atmosphere. However, its environmental drivers are not fully understood, and there are still significant uncertainties in soil respiration model estimates. This study aimed to estimate the spatial distribution pattern and driving mechanism of global soil respiration by constructing a machine learning model method based on ecological big data. First, we constructed ecological big data containing five categories of 27-dimensional environmental factors. We then used four typical machine learning methods to develop the performance of machine learning models under four training strategies and explored the relationship between soil respiration and environmental factors. Finally, we used the RF machine learning algorithm to estimate the global Rs spatial distribution pattern in 2021, driven by multiple dimensions of environmental factors, and derived the annual soil respiration values. The results showed that RF performed better under the four training strategies, with a coefficient of determination R2 = 0.78216, root mean squared error (RMSE) = 285.8964 gCm-2y-1, and mean absolute error (MAE) = 180.4186 gCm-2y-1, which was more suitable for the estimation of large-scale soil respiration. In terms of the importance of environmental factors, unlike previous studies, we found that the influence of geographical location was greater than that of MAP. Another new finding was that enhanced vegetation index 2 (EVI2) had a higher contribution to soil respiration estimates than the enhanced vegetation index (EVI) and normalized vegetation index (NDVI). Our results confirm the potential of utilizing ecological big data for spatially large-scale Rs estimations. Ecological big data and machine learning algorithms can be considered to improve the spatial distribution patterns and driver analysis of Rs.

Association of survival with adjuvant chemotherapy in patients with stage IB gastric cancer: a multicentre, observational, cohort study.

Background: Recurrence following radical resection in patients with stage IB gastric cancer (GC) is not uncommon. However, whether postoperative adjuvant chemotherapy could reduce the risk of recurrence in stage IB GC remains contentious. Methods: We collected data on 2110 consecutive patients with pathologic stage IB (T1N1M0 or T2N0M0) GC who were admitted to 8 hospitals in China from 2009 to 2018. The survival of patients who received adjuvant chemotherapy was compared with that of postoperative observation patients using propensity score matching (PSM). Two survival prediction models were constructed to estimate the predicted net survival gain attributable to adjuvant chemotherapy. Findings: Of the 2110 patients, 1344 received adjuvant chemotherapy and 766 received postoperative observation. Following the 1-to-1 matching, PSM yielded 637 matched pairs. Among matched pairs, adjuvant chemotherapy was not associated with improved survival compared with postoperative observation (OS: hazard ratio [HR], 0.72; 95% CI, 0.52-1.00; DFS: HR, 0.91; 95% CI, 0.64-1.29). Interestingly, in the subgroup analysis, reduced mortality after adjuvant chemotherapy was observed in the subgroups with elevated serum CA19-9 (HR, 0.22; 95% CI, 0.08-0.57; P = 0.001 for multiplicative interaction), positive lymphovascular invasion (HR, 0.32; 95% CI, 0.17-0.62; P < 0.001 for multiplicative interaction), or positive lymph nodes (HR, 0.17; 95% CI, 0.07-0.38; P < 0.001 for multiplicative interaction). The survival prediction models mainly based on variables associated with chemotherapy benefits in the subgroup analysis demonstrated good calibration and discrimination, with relatively high C-indexes. The C-indexes for OS were 0.74 for patients treated with adjuvant chemotherapy and 0.70 for patients treated with postoperative observation. Two nomograms were built from the models that can calculate individualized estimates of expected net survival gain attributable to adjuvant chemotherapy. Interpretation: In this cohort study, pathologic stage IB alone was not associated with survival benefits from adjuvant chemotherapy compared with postoperative observation in patients with early-stage GC. High-risk clinicopathologic features should be considered simultaneously when evaluating patients with stage IB GC for adjuvant chemotherapy. Funding: National Natural Science Foundation of China; the National Key R&D Program of China.

An Improved EfficientNetV2 Model Based on Visual Attention Mechanism: Application to Identification of Cassava Disease.

With the characteristic of high recognition rate and strong network robustness, convolutional neural network has now become the most mainstream method in the field of crop disease recognition. Aiming at the problems with insufficient numbers of labeled samples, complex backgrounds of sample images, and difficult extraction of useful feature information, a novel algorithm is proposed in this study based on attention mechanisms and convolutional neural networks for cassava leaf recognition. Specifically, a combined data augmentation strategy for datasets is used to prevent single distribution of image datasets, and then the PDRNet (plant disease recognition network) combining channel attention mechanism and spatial attention mechanism is proposed. The algorithm is designed as follows. Firstly, an attention module embedded in the network layer is deployed to establish remote dependence on each feature layer, strengthen the key feature information, and suppress the interference feature information, such as background noise. Secondly, a stochastic depth learning strategy is formulated to accelerate the training and inference of the network. And finally, a transfer learning method is adopted to load the pretrained weights into the model proposed in this study, with the recognition accuracy of the model enhanced by means of detailed parameter adjustments and dynamic changes in the learning rate. A large number of comparative experiments demonstrate that the proposed algorithm can deliver a recognition accuracy of 99.56% on the cassava disease image dataset, reaching the state-of-the-art level among CNN-based methods in terms of accuracy.

A Method for Medical Microscopic Images' Sharpness Evaluation Based on NSST and Variance by Combining Time and Frequency Domains.

An algorithm for a sharpness evaluation of microscopic images based on non-subsampled shearlet wave transform (NSST) and variance is proposed in the present study for the purpose of improving the noise immunity and accuracy of a microscope's image autofocus. First, images are decomposed with the NSST algorithm; then, the decomposed sub-band images are subjected to variance to obtain the energy of the sub-band coefficients; and finally, the evaluation value is obtained from the ratio of the energy of the high- and low-frequency sub-band coefficients. The experimental results show that the proposed algorithm delivers better noise immunity performance than other methods reviewed by this study while maintaining high sensitivity.

Long-term trend in vegetation gross primary production, phenology and their relationships inferred from the FLUXNET data.

Climate-induced changes in plant phenology and physiology plays an important role in control of carbon exchange between terrestrial ecosystems and the atmosphere. Based on dataset during 1997-2014 from 41 flux tower sites (440 site-years) across the northern hemisphere, relationships between long-term trends in start of growing season (SOS), end of growing season (EOS), length of growing season (LOS), maximal gross primary production (GPPmax), and seasonal and annual gross primary production (GPP) were analyzed. Statistical Models of Integrated Phenology and Physiology (SMIPP) were built for predicting the long-term trends in annual GPP. Results showed that SOS advanced and EOS delayed for forest sites, while both SOS and EOS for grassland (GRA) sites delayed. Long-term trends in SOS and EOS of evergreen needle-leaf forests (ENF) sites were greater than those of deciduous broadleaf forests (DBF) sites. Seasonal and annual GPP for forest sites increased, among which long-term trend in annual GPP of ENF sites was the largest. Spring GPP of GRA sites decreased, but annual GPP increased. Strong relationships between long-term trends in phenological and physiological indicators and seasonal GPP were found. Long-term trend in GPPmax had the highest relationship with long-term trend in annual GPP for forest sites, but long-term trend in SOS was the most related to long-term trend in annual GPP for GRA sites. Increases in spring and autumn GPP due to a one-day advance in SOS and delay in EOS for DBF sites were greater than ENF sites. Delay in EOS resulted in more carbon sequestration than advance in SOS for forest sites, while advance in SOS significantly increased spring GPP for GRA sites. The SMIPP model driven by long-term trends in LOS and GPPmax had stronger explanatory power for predicting long-term trend in annual GPP than the SMIPP model driven by long-term trends in SOS, EOS, and GPPmax. Long-term trend in annual GPP was accurately predicted by using the SMIPP model, while long-term trend in annual GPP for GRA sites was more difficult to be captured than the forest sites. Drought and disturbance effects on phenology and physiology were major factors for model uncertainty. This study is helpful to understand changes in phenology and carbon uptake and their differences among different vegetation types and provides a potential way for predicting annual rate of change in carbon uptake through vegetation photosynthesis at a global scale.

Improving Estimations of Spatial Distribution of Soil Respiration Using the Bayesian Maximum Entropy Algorithm and Soil Temperature as Auxiliary Data.

Soil respiration inherently shows strong spatial variability. It is difficult to obtain an accurate characterization of soil respiration with an insufficient number of monitoring points. However, it is expensive and cumbersome to deploy many sensors. To solve this problem, we proposed employing the Bayesian Maximum Entropy (BME) algorithm, using soil temperature as auxiliary information, to study the spatial distribution of soil respiration. The BME algorithm used the soft data (auxiliary information) effectively to improve the estimation accuracy of the spatiotemporal distribution of soil respiration. Based on the functional relationship between soil temperature and soil respiration, the BME algorithm satisfactorily integrated soil temperature data into said spatial distribution. As a means of comparison, we also applied the Ordinary Kriging (OK) and Co-Kriging (Co-OK) methods. The results indicated that the root mean squared errors (RMSEs) and absolute values of bias for both Day 1 and Day 2 were the lowest for the BME method, thus demonstrating its higher estimation accuracy. Further, we compared the performance of the BME algorithm coupled with auxiliary information, namely soil temperature data, and the OK method without auxiliary information in the same study area for 9, 21, and 37 sampled points. The results showed that the RMSEs for the BME algorithm (0.972 and 1.193) were less than those for the OK method (1.146 and 1.539) when the number of sampled points was 9 and 37, respectively. This indicates that the former method using auxiliary information could reduce the required number of sampling points for studying spatial distribution of soil respiration. Thus, the BME algorithm, coupled with soil temperature data, can not only improve the accuracy of soil respiration spatial interpolation but can also reduce the number of sampling points.