Chromatic aberration correction based on cross-channel information alignment in microscopy.

A microscope usually consists of dozens of complex lenses and requires careful assembly, alignment, and testing before use. Chromatic aberration correction is a significant step in the design of microscopes. Reducing chromatic aberration by improving optical design will inevitably increase the overall weight and size of the microscope, leading to more cost in manufacturing and maintenance. Nevertheless, the improvement in hardware can only achieve limited correction. In this paper, we propose an algorithm based on cross-channel information alignment to shift some of the correction tasks from optical design to post-processing. Additionally, a quantitative framework is established to evaluate the performance of the chromatic aberration algorithm. Our algorithm outperforms the other state-of-the-art methods in both visual appearance and objective assessments. The results indicate that the proposed algorithm can effectively obtain higher-quality images without changing the hardware or engaging the optical parameters.

Dark-field scattering image compression using a sparse matrix.

Dark-field scattering imaging is an imaging method with high contrast and high sensitivity. It has been widely employed in optical components evaluation, biomedical detection, semiconductor manufacturing, etc. However, useless background information causes data redundancy, which increases unnecessary time-space costs in processing. Furthermore, the problem is particularly serious in high-resolution imaging systems for large-aperture components. The dark-field scattering image compression (DFSIC) based on the compressed sparse row is proposed to solve this problem. The compression method realizes local data access for a sparse matrix. The result of the experiments shows that the average time-space consumption of the DFSIC is reduced to less than 2%, compared with the raw image structure, and is still kept below 68% in dense cases. This method provides a more efficient program implementation for the dark-field scattering imaging and exhibits potential in the application of the optical detection with large scale.

Pollution loads in the middle-lower Yangtze river by coupling water quality models with machine learning.

Pollution control and environmental protection of the Yangtze River have received major attention in China. However, modeling the river's pollution load remains challenging due to limited monitoring and unclear spatiotemporal distribution of pollution sources. Specifically, anthropogenic activities' contribution to the pollution have been underestimated in previous research. Here, we coupled a hydrodynamic-based water quality (HWQ) model with a machine learning (ML) model, namely attention-based Gated Recurrent Unit, to decipher the daily pollution loads (i.e., chemical oxygen demand, COD; total phosphorus, TP) and their sources in the Middle-Lower Yangtze River from 2014 to 2018. The coupled HWQ-ML model outperformed the standalone ML model with KGE values ranging 0.77-0.91 for COD and 0.47-0.64 for TP, while also reducing parameter uncertainty. When examining the relative contributions at the Middle Yangtze River Hankou cross-section, we observed that the main stream and tributaries, lateral anthropogenic discharges, and parameter uncertainty contributed 15, 66, and 19% to COD, and 58, 35, and 7% to TP, respectively. For the Lower Yangtze River Datong cross-section, the contributions were 6, 69, and 25% for COD and 41, 42, and 17% for TP. According to the attention weights of the coupled model, the primary drivers of lateral anthropogenic pollution sources, in descending order of importance, were temperature, date, and precipitation, reflecting seasonal pollution discharge, industrial effluent, and first flush effect and combined sewer overflows, respectively. This study emphasizes the synergy between physical modeling and machine learning, offering new insights into pollution load dynamics in the Yangtze River.

Water quality prediction based on sparse dataset using enhanced machine learning.

Water quality in surface bodies remains a pressing issue worldwide. While some regions have rich water quality data, less attention is given to areas that lack sufficient data. Therefore, it is crucial to explore novel ways of managing source-oriented surface water pollution in scenarios with infrequent data collection such as weekly or monthly. Here we showed sparse-dataset-based prediction of water pollution using machine learning. We investigated the efficacy of a traditional Recurrent Neural Network alongside three Long Short-Term Memory (LSTM) models, integrated with the Load Estimator (LOADEST). The research was conducted at a river-lake confluence, an area with intricate hydrological patterns. We found that the Self-Attentive LSTM (SA-LSTM) model outperformed the other three machine learning models in predicting water quality, achieving Nash-Sutcliffe Efficiency (NSE) scores of 0.71 for CODMn and 0.57 for NH3N when utilizing LOADEST-augmented water quality data (referred to as the SA-LSTM-LOADEST model). The SA-LSTM-LOADEST model improved upon the standalone SA-LSTM model by reducing the Root Mean Square Error (RMSE) by 24.6% for CODMn and 21.3% for NH3N. Furthermore, the model maintained its predictive accuracy when data collection intervals were extended from weekly to monthly. Additionally, the SA-LSTM-LOADEST model demonstrated the capability to forecast pollution loads up to ten days in advance. This study shows promise for improving water quality modeling in regions with limited monitoring capabilities.

Spatial heterogeneity in sediment and carbon accretion rates within a seagrass meadow correlated with the hydrodynamic intensity.

The majority of the carbon stored in seagrass sediments originates outside the meadow, such that the carbon storage capacity within a meadow is strongly dependent on hydrodynamic conditions that favor deposition and retention of fine organic matter within the meadow. By extension, if hydrodynamic conditions vary across a meadow, they may give rise to spatial gradients in carbon. This study considered whether the spatial gradients in sediment and carbon accretion rates correlated with the spatial variation in hydrodynamic intensity within a single meadow. Field measurements were conducted in three depth zones across a Zostera marina L. (eelgrass) meadow in Nahant Harbor, Massachusetts. Four sediment cores were collected in each zone, including one outside the meadow (control) and three within the meadow at increasing distances from the nearest meadow edge. Sedimentation and carbon accretion rates were estimated by combining the measurements of dry bulk density, organic carbon fraction (%OC), 210Pb, and 226Ra. Tilt current meters measured wave velocities within each zone, which were used to estimate turbulent kinetic energy (TKE). Both sediment and carbon accretion rates exhibited spatial heterogeneity across the meadow, which were correlated with the spatial variation in near-bed TKE. Specifically, both accretion rates increased with decreasing TKE, which was consistent with diminished resuspension associated with lower TKE. A method is proposed for using spatial gradients in hydrodynamic intensity to improve the estimation of total meadow accretion rates.

Value of layer-specific speckle tracking echocardiography for early detection of myocardial injury caused by chemotherapy in breast cancer patients with cardiovascular risk.

The probability of toxicity-related myocardial injury event with anthracyclines is controversial, which could be related to the underlying cardiac status before chemotherapy. Our study sought to investigate the influence of cardiovascular risk factors on myocardial motion and cardiac function using layer-specific speckle tracking echocardiography (STE) during chemotherapy with epirubicin. Female patients with first-diagnosed breast cancer were prospectively enrolled in our study and received 4 chemotherapeutic cycles with epirubicin in each cycle of 21 days. All patients underwent echocardiography for layer-specific STE analysis before and after all chemotherapy. Clinical data including cardiovascular risk factors were collected. According to the Framingham score, patients with cardiovascular risk factors were divided into groups with low, medium, and high risk. 134 patients existed in the final analysis. The accumulated dose of epirubicin for were 560.0 ± 103.8 mg. 97 (72.4%) patients had cardiovascular risk factors. According to the Framingham score, 57 (42.5%) patients categorized in high risk. Endocardial layer strain after chemotherapy were lower than those at baseline (p < 0.05, all), especially for patients with high risk. The changes of endocardial longitudinal strain during chemotherapy were associated with cardiovascular risks at baseline with correlation coefficient of 0.627. Our study found that layer-specific STE is valuable for early detection of toxicity-related myocardial injury for patients with breast cancer after epirubicin chemotherapy and cardiovascular risk factors have greatly influenced on cardiac function during chemotherapy. The endocardial layer strain is sensitive to evaluate early-stage toxicity-related myocardial injury after epirubicin chemotherapy.

Prognostic value of myocardial contrast echocardiography in acute anterior wall ST-segment elevation myocardial infarction with successful epicardial recanalization.

Although myocardial contrast echocardiography (MCE) can evaluate microvascular perfusion abnormalities, its prognostic value is uncertain in acute anterior wall ST-Segment elevation myocardial infarction (STEMI) with successful epicardial recanalization. Therefore, the study aims to investigate the prognostic role of qualitative and quantitative MCE in acute anterior wall STEMI with successful epicardial recanalization. 153 STEMI patients were assessed by MCE within 7 days after successful epicardial recanalization. Qualitative perfusion parameters (microvascular perfusion score index, MPSI) and quantitative perfusion parameters (A, ß, and Aß) were acquired using a 17-segment model. And corrected A and Aß were calculated. Patients were all followed for major adverse cardiovascular events (MACEs). During median follow-up of 27 (4) months, 39 (25.49%) patients experienced MACEs, while 114 (74.51%) were free from MACEs. Patients with MACEs had higher MPSI (1.65 ± 0.13 vs. No-MACEs 1.35 ± 0.20, P < 0.001), lower ß (1.09 ± 0.19 s-1 vs. No-MACEs 1.34 ± 0.30 s-1, P < 0.001), corrected A (0.17 ± 0.03 dB vs. No-MACEs 0.19 ± 0.04 dB, P = 0.039) and lower corrected Aß (0.19 ± 0.06 dB/s vs. No-MACEs 0.25 ± 0.08 dB/s, P < 0.001). MPSI of 1.44 provided an area under the curve (AUC) of 0.872, while ß of 1.18 s-1 and corrected Aß of 0.22 dB/s provided AUCs of 0.759 and 0.724, respectively. The combination of MPSI, ß and corrected Aß provided an increased AUC of 0.964 (all P < 0.05). Time-dependent ROC analysis showed that the AUCs of the MPSI, ß, corrected Aß and the combination at 1, 1.5 and 2 years indicated a strong predictive power for MACEs (AUC = 0.900/0.894/0.881 for MPSI, 0.648/0.704/0.732 for ß, 0.674/0.686/0.722 for corrected Aß, and 0.947/0.962/0.967 for the combination, respectively). Patients with MPSI < 1.44, ß > 1.18 s-1, or corrected Aß > 0.22 dB/s had lower event rate (all Log Rank P ≤ 0.001). MPSI, ß, corrected Aß, GLS and WBC were independent predictors of MACEs with adjusted hazard ratio of 34.41 (8.18-144.87), P < 0.001 for MPSI; 39.29 (27.46-65.44), P < 0.001 for ß; 8.93 (1.46-54.55), P = 0.018 for corrected Aß; 10.88 (2.83-41.86), P = 0.001 for GLS; and 1.43 (1.16-1.75), P = 0.001 for WBC. Qualitative and quantitative MCE can accurately predict MACEs in acute anterior wall STEMI with successful epicardial recanalization, and their combined predictive value is higher.

Layer-Specific Strain for Long-Term Outcome Prediction After First-Onset Myocardial Infarction.

Many studies have reported the prognostic value of global strain obtained with speckle tracking echocardiography (STE) in patients with acute myocardial infarction (AMI). However, as a novel method derived from STE, layer-specific strain has seldom been evaluated with respect to prediction of AMI outcomes. We sought to investigate the predictive value of layer-specific strain and whether it has incremental value compared with conventional parameters, such as left ventricular ejection fraction and wall motion score index, and STE parameters. Our study was prospective. Ninety-two patients with first-onset AMI were enrolled and underwent echocardiography before coronary intervention for analysis of global and layer-specific strain. Cox proportional hazard ratio (HR) and receiver operating characteristic curve analyses were performed for the prediction of cardiac events and cardiac death. Fifty-three patients have had cardiac events during follow-up. Endocardial longitudinal strain has received relatively higher HRs for risk predictions of both cardiac events (HR = 1.69) and cardiac death (HR = 3.21) adjusted with clinical data. The areas under the receiver operating characteristic curves of the longitudinal strain at the endocardial layer from layer-specific strain were higher than those of global strain and conventional parameters for cardiac event prediction (p ˂ 0.05, all). Layer-specific strain is valuable for cardiac risk prediction after infarction and has incremental values in addition to conventional and global STE parameters. Myocardial damage at the endocardial layer was closely related to outcomes of AMI patients at long-term follow-ups.