RESUMEN

The real-time monitoring of corn yield by a combine harvester is a critical data source for constructing the yield histogram, which significantly benefits precision management and decision-making in modern precision agriculture. While widely used, the current photoelectric sensor-based yield monitoring method has limitations. It detects the corn height on each scraper and calculates the yield through a geometric formula. However, it neglects the noticeable difference in the corn stacking patterns affected by factors such as feeding volume, terrain, and driving speed. This oversight often results in low accuracy and poor stability in the prediction of corn yield, highlighting the need for a more advanced approach. To resolve this, we employ EDEM discrete element simulation to demonstrate the large difference of corn stacking patterns on the scraper of the elevator corresponding to feeding volume. Then, we develop a real-time monitoring system on our self-developed double elevator testing rig for carrying out a composite dataset for training three machine learning algorithm-based models, namely Deep Neural Networks (DNN), Gradient Boosting Machine (GBM), and Random Forest (RF). Importantly, these models have undergone rigorous validation under various feeding volumes, ensuring their robustness and reliability. The auxiliary elevator speed is meticulously set at 150r/min, 225r/min, and 450r/min, providing a comprehensive performance assessment. The results denote that the DNN model performs best and is stable, with a coefficient of determination (R2) of 0.998, root mean square error (RMSE) of 0.526, and mean absolute error (MAE) of 0.425. The paper also performs field experiments to test the proposed three prediction models and the system. The results also denote the DNN-based prediction model's best performance for the lowest relative error of 2.29% and the highest average accuracy of 97.85%. Consequently, the proposed real-time corn yield monitoring system achieves high accuracy and reliability for the combine harvester applications.

RESUMEN

BACKGROUND: Cancer remains a formidable global health challenge, currently affecting nearly 20 million individuals worldwide. Due to the absence of universally effective treatments, ongoing research explores diverse strategies to combat this disease. Recent efforts have concentrated on developing combined drug regimens and targeted therapeutic approaches. OBJECTIVE: This study aimed to investigate the anticancer efficacy of a conjugated drug system, consisting of doxorubicin and cisplatin (Dox-Cis), encapsulated within niosomes and modified with MUC-1 aptamers to enhance biocompatibility and target specific cancer cells. METHODS: The chemical structure of the Dox-Cis conjugate was characterized using Fourier Transform Infrared Spectroscopy (FTIR) and Liquid Chromatography Quadrupole Time-of-Flight Mass Spectrometry (LC-Q-TOF/MS). The zeta potential and morphological parameters of the niosomal vesicles were determined through Dynamic Light Scattering (DLS) and Transmission Electron Microscopy (TEM). In vitro assessments of cell viability and apoptosis were conducted on MUC-1 positive HeLa cells and MUC-1 negative U87 cells. RESULTS: The findings confirmed the successful conjugation of Dox and Cis within the niosomes. The Nio/Dox-Cis/MUC-1 formulation demonstrated enhanced efficacy compared to the individual drugs and their unencapsulated combination in both cell lines. Notably, the Nio/Dox-Cis/MUC-1 formulation exhibited greater effectiveness on HeLa cells (38.503 ± 1.407) than on U87 cells (46.653 ± 1.297). CONCLUSION: The study underscores the potential of the Dox-Cis conjugate as a promising strategy for cancer treatment, particularly through platforms that facilitate targeted drug delivery to cancer cells. This targeted approach could lead to more effective and personalized cancer therapies.

Asunto(s)

RESUMEN

PURPOSE: To evaluate the results of a minimally invasive combined endoscopic and eyelid crease/medial suprabrow incision approach in collaboration with oculoplastic and sinus surgeons for the treatment of recurrent/recalcitrant sino-orbital mucoceles. METHODS: Eighteen cases of recurrent/recalcitrant sino-orbital mucoceles, treated in collaboration with oculoplastic and sinus surgeons at the University of Michigan, were retrospectively reviewed. The recurrence of mucocele, reduction in proptosis, and complications were evaluated. RESULTS: The mean age at the time of surgery was 49 years (range: 17-76 years). All cases had a history of previous sinus or orbital surgeries for mucoceles. Among 18 cases, eight were due to chronic sinus infections, six due to trauma, three due to Schneiderian papilloma, and one case was secondary to an inflammatory sinus disease. Thirteen cases (72%) presented with orbital or facial cellulitis, while five cases (38%) experienced periocular swelling and limited extraocular motility. Following a mean follow-up of 19 months (range: 1-76 months)), recurrence was observed in two cases (11%): one in a cystic fibrosis patient with chronic sinusitis, and the other in a case of Schneiderian papilloma. The mean pre-operative proptosis in the affected eye was 2.78 mm, with an average decrease of 2.33 mm after surgery. Complications occurred in two cases, including one case of hypoesthesia in the forehead and one case of post-operative strabismus. CONCLUSION: Our series of 18 cases of recurrent/recalcitrant mucoceles, with only two cases of recurrence, demonstrates that this minimally invasive approach can be successfully employed for advanced sino-orbital disease, with a low rate of adverse outcomes and aesthetically pleasing results.

RESUMEN

Purpose: The serum lipid level is strongly associated with atherosclerosis. However, research on the relationship between lipid-derived indices and acute ischemic stroke (AIS) occurrence in hemodialysis populations is limited. This study aimed to explore the predictive value of lipid-derived indices, including atherogenic index of plasma (AIP), Non- high density lipoprotein cholesterol (Non-HDL-C), Non-HDL-C/HDL-C, and lipoprotein combine index (LCI) in clinical practice for the occurrence and prognosis of AIS in hemodialysis patients. Methods: A total of 451 patients undergoing maintenance hemodialysis were screened and 350 were enrolled in this study. The lipid parameters exhibit a progressive increase across the tertiles, with values rising from Q1 through Q3. Enrolled patients were divided into three groups (Q1, Q2, and Q3) based on tertiles of AIP, Non-HDL-C, Non-HDL-C/HDL-C, and LCI values. Kaplan-Meier curves were performed to investigate the association between the AIP, Non-HDL-C, Non-HDL-C/HDL-C, LCI and AIS-free survival in hemodialysis patients. Chi-square analysis was used to explore the association between the AIP, Non-HDL-C, Non-HDL-C/HDL-C, LCI and AIS outcomes in hemodialysis patients. AIS outcomes were assessed using the modified Rankin Scale (mRS). Results: Kaplan-Meier analysis revealed that the AIS-free survival rates were significantly higher in the Q1 group compared to Q2 and Q3 groups for AIP, Non-HDL-C, Non-HDL-C/HDL-C, and LCI. Log rank tests showed statistically significant differences between the Q1 group and the Q2 and Q3 groups (p < 0.05 for all). The proportion of patients with a good outcome mRS was higher in the Q1 group compared to the Q2-Q3 groups (AIP: 0.818 vs 0.792; Non- HDL-C: 0.866 vs 0.767; Non- HDL-C/HDL-C: 0.867 vs 0.767; LCI: 0.938 vs 0.750). Conclusion: The four lipid-derived parameters are effective predictors of AIS in patients undergoing hemodialysis, and AIP has a strongest correlation with the risk of AIS. Hemodialysis patients with elevated levels of the four lipid-derived indices had a higher incidence of AIS and poorer functional outcomes compared to those with lower levels. Our conclusions may require confirmation by further research in the future.

Asunto(s)

RESUMEN

Purpose: The association between traditional lipid parameters and non-alcoholic fatty liver disease (NAFLD) has been extensively discussed. This study aims to evaluate and compare the lipoprotein combine index (LCI) and traditional lipid parameters [total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C)] to identify NAFLD. Patients and Methods: The analysis included 14,251 participants from the NAfld in the Gifu Area, Longitudinal Analysis (NAGALA). Logistic regression models were employed to calculate standardized odds ratios (ORs) and 95% confidence intervals (CIs) for assessing and comparing the association of LCI and traditional lipid parameters with NAFLD. Additionally, receiver operating characteristic (ROC) curves were used to calculate the area under the curve (AUC) for LCI and traditional lipid parameters in identifying NAFLD. Results: After adjusting for various confounders, we found that LCI was positively associated with NAFLD (OR=2.25, 95% CI 1.92-2.63), and this association was stronger than that of traditional lipid parameters [OR: TC1.23, TG1.73 LDL-C1.10]. Further subgroup analyses revealed that the association of LCI with NAFLD was stronger than other traditional lipid parameters in all subgroups, including men and women, overweight/obese [body mass index (BMI)≥25 kg/m2] and non-obese (BMI<25 kg/m2), and older (age≥45 years) and younger (age<45 years) participants. Additionally, ROC analysis indicated that LCI (AUC=0.8118) had significantly higher accuracy (All DeLong P<0.05) in identifying NAFLD compared to traditional lipid parameters (AUC: TC0.6309; TG0.7969; LDL-C0.6941); HDL-C0.7587). Sensitivity analysis further confirmed the robustness of the study findings. Conclusion: This study revealed for the first time a positive correlation between LCI and NAFLD. Compared to traditional lipid parameters, LCI has a higher correlation with NAFLD. Additionally, further ROC analysis demonstrated that LCI had higher accuracy in identifying NAFLD compared to traditional lipid parameters, suggesting that LCI may be a better marker for NAFLD identification than traditional lipid parameters.

RESUMEN

The aim of this study was to evaluate the impact of implementing a basketball-specific exercise program using Fitlight technology on the coordinative abilities (reactive coordination, reactive reaction time, and reactive movement combination capacity) of U14 and U16 junior basketball players. This study included 70 male basketball players, 36 subjects U14 and 34 subjects U16, divided into two equal groups for each age category: the experimental group (EG) and the control group (CG). This study included an initial and a final test, between which, training was conducted over a period of 18 weeks. For the EG, the program aimed to develop coordinative abilities through an experimental program that utilized Fitlight technology, while the control groups underwent an identically timed program, but their training did not include the use of technology. Four tests were adapted and applied: the Reactive Reaction Test, Choice Reactive Reaction Test, Reactive Hand-Eye Coordination Test, and a test to evaluate the reactive capacity for combining movements. The results of this study showed statistically significant progress between the initial and final tests for the experimental group, p < 0.05. The Cohen's d values for the experimental groups were above 0.8, indicating a very large effect size, while for the control group, these values were small to medium. The comparative analysis of the experimental groups U14 and U16 and control groups shows statistically significant differences in favor of the experimental groups U14 and U16. This study highlights the effectiveness of implementing specific training programs that use modern technologies in developing coordinative abilities in the training and evaluation process of junior basketball players.

Asunto(s)

RESUMEN

Both water management measures like damming and changes in precipitation as a result of anthropogenic induced climate change have exerted profound effects on the dynamics of streamwater-groundwater interaction (SGI). However, their compound effects on SGI have not been investigated so far. Taking the Fen River of China as an example, this study aims to examine the synergistic impacts of damming and precipitation anomalies on SGI dynamics. The sampling considered the seasonal and interannual variability of precipitation (May and September in 2019 representing a dry year; May and August in 2021 representing a wet year), and long-term daily observational data, including water levels and water discharge, were combined to elucidate the compound effects. Precipitation anomalies and damming exert significant individual and combined influences on SGI. Separately, dams and reservoirs reversed the SGI dynamics, significantly increasing the contributions of streamwater to groundwater from 0 to 29 % to 78 % in the dam-affected areas. Further, the groundwater discharge ratios behind the dam (about 60 %) were three times higher than those in front of the dam. Precipitation anomalies significantly amplified interannual variability in SGI patterns, and groundwater discharge ratios increased by 47 % during the dry period (2019) compared to flood period (2021). The combined influence of precipitation anomalies and dam regulation remarkably changed the lateral, vertical, and longitudinal water exchange dynamics. Precipitation anomalies affected the SGI dynamics at the whole watershed scale, whereas dam regulation regimes exhibited a stronger control at the local scale. The compound effects of dam regulation and precipitation anomalies can result in different SGI patterns under various climate scenarios. More attention should be paid to the interrelated feedback mechanisms between damming, extreme precipitation events, and their impact on the watershed-scale hydrological cycle.

RESUMEN

Visual navigation is widely used in intelligent combine harvesters, but the existing algorithms do not have sufficiently high accuracy of the visual navigation line recognition under different sunlight conditions. To address this problem, this article proposes a sunlight-robust DeepLabV3+-based navigation line extraction method for combine harvesters. The navigation lines are extracted by constructing a new dataset and predicting the boundaries of the areas that have been and have not been cut. To address the problem that DeeplabV3+ is not sufficient light in the DCNN part, improvement is proposed by incorporating the MobileNetV2 module. In image segmentation, the prediction time is 22.5 ms, and the mean intersection over union (FMIOU) is 0.79. After image segmentation, the navigation lines are drawn using the line segment detection algorithm for the harvester. The proposed method is compared with other mainstream networks, and the prediction results are compared using the line segment detection method. The results show that this method can more quickly identify the navigation lines under different conditions of sunlight with less labeled data than the improved U-Net and DeeplabV3+, which uses Xception as the backbone. Compared to the traditional method and the improved U-Net, this method achieves good results and improves the recognition speed by 27 and 9 ms, respectively.

RESUMEN

Fast and accurate 3D scene perception is a crucial prerequisite for the autonomous navigation and harvesting of combine harvesters. However, crop field scenarios pose severe challenges for vision-based perception systems due to repetitive scenes, illumination changes and real-time constraints on embedded computing platforms. In this paper, we propose a feature-based, two-stage approach for real-time dense 3D mapping for combine harvesters. In the first stage, our approach constructs a sparse 3D map using reliable feature matching, which provides prior knowledge about the environment. In the second stage, our method formulates per-pixel disparity calculation as probabilistic inference. The key to our approach is the ability to compute dense 3D maps by combining Bayesian estimation with efficient and discriminative point cues from images, exhibiting tolerance against visual measurement uncertainties due to repetitive textures and uneven lighting in crop fields. We validate the performance of the proposed method using real crop field data, and the results demonstrate that our dense 3D maps provide detailed spatial metric information while maintaining a balance between accuracy and efficiency. This makes our approach highly valuable for online perception in combine harvesters operating with resource-limited systems.

RESUMEN

Background: Each year, there are billions of agricultural work accidents involving the operation of tractors, grain augers, harvest combines, power take-off devices, or balers and thrashers. Field accidents of this nature seem more common on afternoons, just as road accidents tend to skew toward nighttime. The lesions can vary widely and depends strictly on the operation of the machinery analyzed. Aims: This paper aims to present a peculiar case of decapitation by a combine harvester, showing how, in cases of injury due to agricultural machinery, it is fundamental a correct execution of a scene investigation, autoptic examination, and cooperation with a specialist in engineering. Case Report: A 54-year-old man was found decapitated on the header of a combine harvester; his extremities were also dismembered. At autopsy, a clean oblique cut across the first cervical vertebra had severed the head at the neck. Although the right arm remained intact, both lower extremities were mutilated, showing numerous exposed and open fractures. A bleeding, penetrating wound to the back was additionally noted. In the days that followed, missing parts (head and left leg) were discovered in other machine components (grain tank and straw walker, respectively). All observed injuries were compatible with the mechanics of the cochlea, its rotating movement inflicting the damages above. Collaboration between pathologists and engineers was fundamental to recreating the dynamics of this rare decapitation accident by a combine harvester.

Asunto(s)

RESUMEN

Because of the influence of harsh and variable working environments, the vibration signals of rolling bearings for combine harvesters usually show obvious characteristics of strong non-stationarity and nonlinearity. Accomplishing accurate fault diagnosis using these signals for rolling bearings is a challenging subject. In this paper, a novel fault diagnosis method based on composite-scale-variable dispersion entropy (CSvDE) and self-optimization variational mode decomposition (SoVMD) is proposed, systematically combining the nonstationary signal analysis approach and machine learning technology. Firstly, an improved SoVMD algorithm is developed to realize adaptive parameter optimization and to further extract multiscale frequency components from original signals. Subsequently, a CSvDE-based feature learning model is established to generate the multiscale fault feature space (MsFFS) of frequency components for the improvement of fault feature learning ability. Finally, the generated MsFFS can serve as the inputs of the Softmax classifier for fault category identification. Extensive experiments on the vibration datasets collected from rolling bearings of combine harvesters are conducted, and the experimental results demonstrate the more superior and robust fault diagnosis performance of the proposed method compared to other existing approaches.

RESUMEN

Objective: Dyslipidemia is a key risk factor for coronary artery disease (CAD). This study aimed to investigate the correlation between the atherogenic index of plasma (AIP) and the severity of CAD. Methods: 2,491 patients were enrolled in this study and analyzed retrospectively, including 665 non-CAD patients as the control group and 1,826 CAD patients. The CAD patients were classified into three subgroups according to tertiles of SYNTAX score (SS). Non-high-density lipoprotein cholesterol (Non-HDL-C) was defined as serum total cholesterol (TC) minus serum high-density lipoprotein cholesterol (Non-HDL-C), atherogenic index (AI) was defined as the ratio of non-HDL-C to HDL-C; AIP was defined as the logarithm of the ratio of the concentration of triglyceride (TG) to HDL-C; lipoprotein combine index (LCI) was defined as the ratio of TC∗TG∗ low-density lipoprotein cholesterol (LDL)to HDL-C; Castelli Risk Index I (CRI I) was defined as the ratio of TC to HDL-C; Castelli Risk Index II (CRI II) was defined as the ratio of LDL-C to HDL-C. Results: The levels of AIP (P < 0.001), AI (P < 0.001), and LCI (P = 0.013) were higher in the CAD group compared with the non-CAD group. The Spearman correlation analysis showed that AIP (r = 0.075, P < 0.001), AI (r = 0.132, P < 0.001), and LCI (r = 0.072, P = 0.001) were positively correlated with SS. The multivariate logistic regression model showed CRI I (OR: 1.11, 95% CI: 1.03-1.19, P = 0.005), CRI II (OR: 1.26, 95% CI: 1.15-1.39, P < 0.001), AI (OR: 1.28, 95% CI: 1.17-1.40, P < 0.001), AIP (OR: 2.06, 95% CI: 1.38-3.07, P < 0.001), and LCI (OR: 1.01, 95% CI: 1.01-1.02, P < 0.001) were independent predictors of severity of CAD After adjusting various confounders. Conclusion: CRI I, CRI II, AIP, AI, and LCI were independent predictors of the severity of CAD, which could be used as a biomarker for the evaluation of the severity of CAD.

RESUMEN

Research on autonomous driving technology is actively underway to solve the facing problems in the agricultural field. Combine harvesters used in East Asian countries, including Korea, are tracked-type vehicles. The steering control system of the tracked vehicle has different characteristics from the wheeled vehicle used in the agricultural tractor. In this paper, a dual GPS antenna-based autonomous driving system and path tracking algorithm were developed for a robot combine harvester. An α-turn-type work path generation algorithm and a path tracking algorithm were developed. The developed system and algorithm were verified through experiments using actual combine harvesters. The experiment consisted of an experiment with harvesting work and an experiment without harvesting work. In the experiment without harvesting work, an error of 0.052 m occurred during working driving and 0.207 m during turning driving. In the experiment where the harvesting work was carried out, an error of 0.038 m occurred during work driving and 0.195 m during turning driving. As a result of comparing the non-work area and driving time to the results of manual driving, the self-driving experiment with harvesting work showed an efficiency of 76.7%.

RESUMEN

The study has been conducted to assess the farm mechanization level and tried to identify the determinants. The research utilizes primary data, collected through personal interview of 397 farm households located in the four districts of central and southeast Oromia region of Ethiopia. Multi-stage, purposive and random sampling procedures were used to select the respondents using probability proportional to size from each district. Mechanization Index (MI) based on the matrix use of animate and mechanical energy inputs that incorporate cost factors was used to estimate farm mechanization level, while Tobit model was employed to analyze factors that determine the farm mechanization level. Household's sex, educational background, experience in farming, family labor availability and social capital, location of household, access to all-weather roads and distance to farm mechanization service providers centers, participation in market and off-farm activities, landholding, land fragmentation and size of livestock owned (TLU) are statistically significant in determining level of farm mechanization. Land consolidation, availing infrastructural facilities and facilitating adult education and short-term trainings are important recommendations to enhance farm mechanization level in the study area.

RESUMEN

The wastewater generated from citric acid production has a high organic loading content. The treatment and reuse of citric acid wastewater with high organic loading become extremely important. In this study, the performance of calcium hydroxide (Ca(OH)2) precipitation as a low-cost and environmentally friendly pre-treatment method and aerobic membrane bioreactor (MBR) combined treatment system was investigated for the treatment of citric acid (CA) wastewater. At the first step, optimization parameters such as agitation speed (100, 150, 200 rpm), temperature (30, 50, 70 °C), and reaction time (2, 4, 6 h) for Ca(OH)2 precipitation as a pre-treatment method were investigated using response surface methodology (RSM) to achieve maximum chemical oxygen demand (COD) removal. Experimental sets were designed using Box-Behnken Design. As a result of pre-treatment with Ca(OH)2 precipitation, a COD removal efficiency of 97.3% was obtained. Then, pre-treated CA wastewater was fed continuously to the MBR process for 10 days, which was the second stage of the combined process. As a result of the MBR process, 92.0% COD removal efficiency was obtained for 24 h HRT and 10 days SRT. In total, 99.8% COD removal efficiency was obtained when combined process was used and COD concentration decreased from 52,000-114 mg/L. For the treatment and reuse of wastewater from citric acid production, Ca(OH)2 precipitation and MBR combined treatment systems demonstrated an effective strategy.

Asunto(s)

RESUMEN

Cutaneous melanoma is an aggressive and highly metastatic skin cancer. In recent years, immunotherapy and targeted small-molecule inhibitors have improved the overall survival of patients. Unfortunately, most patients in advanced stages of disease exhibit either intrinsically resistant or rapidly acquire resistance to these approved treatments. However, combination treatments have emerged to overcome resistance, and novel treatments based on radiotherapy (RT) and targeted radionuclide therapy (TRT) have been developed to treat melanoma in the preclinical mouse model, raising the question of whether synergy in combination therapies may motivate and increase their use as primary treatments for melanoma. To help clarify this question, we reviewed the studies in preclinical mouse models where they evaluated RT and TRT in combination with other approved and unapproved therapies from 2016 onwards, focusing on the type of melanoma model used (primary tumor and or metastatic model). PubMed® was the database in which the search was performed using mesh search algorithms resulting in 41 studies that comply with the inclusion rules of screening. Studies reviewed showed that synergy with RT or TRT had strong antitumor effects, such as tumor growth inhibition and fewer metastases, also exhibiting systemic protection. In addition, most studies were carried out on antitumor response for the implanted primary tumor, demonstrating that more studies are needed to evaluate these combined treatments in metastatic models on long-term protocols.

Asunto(s)

RESUMEN

Soil salinity and water deficit often occur concurrently, but understanding their combined effects on plants' ion regulation is limited. With aim to identify if introducing drought with salinity alleviates salt stress's ionic effects, Panicum antidotale - a halophytic grass- was grown in the presence of single and combined stressors, i.e., drought and salt (low and high). Regulation of cations and anions along with the antioxidant capacity and modifications in leaf anatomy were investigated. Results showed a combination of low salt and drought minimally affected plant (dry) mass by improving the selective ions absorption and nutrient use efficiencies. The lowest ratio for efficiency of photosystem II and carbon assimilation (ΦPSII/ΦCO2) suggested less generation of reactive oxygen species, which were probably detoxified with constitutively performing antioxidant enzymes. In contrast, the combination of high salinity and drought escalated the adverse effects caused due to individual stressors. The selective ion absorption increased, but the non-selective ions transport caused an ionic imbalance indicating the highest ratio of Na+/K+. Although the area of mesophyll increased, a reduction in epidermis (cell number and area) predicted a mechanical injury prone to water loss in these plants. The compromised activity of antioxidant enzymes also suggested treatment-induced oxidative damage. Yet, the synergistic interaction between high salinity and drought was not detrimental to the survival of P. antidotale. Therefore, we suggest planting this grass in habitats with harsh environmental conditions to meet the increasing fodder demands without compromising agricultural lands' productivity.

RESUMEN

Photodynamic therapy (PDT) is a promising anticancer noninvasive technique that relies on the generation of reactive oxygen species (ROS). Unfortunately, PDT still has many limitations, including the resistance developed by cancer cells to the cytotoxic effect of ROS. Autophagy, which is a stress response mechanism, has been reported as a cellular pathway that reduces cell death following PDT. Recent studies have demonstrated that PDT in combination with other therapies can eliminate anticancer resistance. However, combination therapy is usually challenged by the differences in the pharmacokinetics of the drugs. Nanomaterials are excellent delivery systems for the efficient codelivery of two or more therapeutic agents. In this work, we report on the use of polysilsesquioxane (PSilQ) nanoparticles for the codelivery of chlorin-e6 (Ce6) and an autophagy inhibitor for early- or late-stage autophagy. Our results, obtained from a reactive oxygen species (ROS) generation assay and apoptosis and autophagy flux analyses, demonstrate that the reduced autophagy flux mediated by the combination approach afforded an increase in the phototherapeutic efficacy of Ce6-PSilQ nanoparticles. We envision that the promising results in the use of multimodal Ce6-PSilQ material as a codelivery system against cancer pave the way for its future application with other clinically relevant combinations.

RESUMEN

While scientific disciplines revere reproducibility, many studies - experimental and computational alike - fall short of this ideal and cannot be reproduced or even repeated when the model is shared. For computational modeling of biochemical networks, there is a dearth of formal training and resources available describing how to practically implement reproducible methods, despite a wealth of existing tools and formats which could be used to support reproducibility. This chapter points the reader to useful software tools and standardized formats that support reproducible modeling of biochemical networks and provides suggestions on how to implement reproducible methods in practice. Many of the suggestions encourage readers to use best practices from the software development community in order to automate, test, and version control their model components. A Jupyter Notebook demonstrating several of the key steps in building a reproducible biochemical network model is included to supplement the recommendations in the text.

Asunto(s)

RESUMEN

Strategies such as ensemble learning and averaging techniques try to reduce the variance of single deep neural networks. The focus of this study is on ensemble averaging techniques, fusing the results of differently initialized and trained networks. Thereby, using micrograph cell segmentation as an application example, various ensembles have been initialized and formed during network training, whereby the following methods have been applied: (a) random seeds, (b) L 1-norm pruning, (c) variable numbers of training examples, and (d) a combination of the latter 2 items. Furthermore, different averaging methods are in common use and were evaluated in this study. As averaging methods, the mean, the median, and the location parameter of an alpha-stable distribution, fit to the histograms of class membership probabilities (CMPs), as well as a majority vote of the members of an ensemble were considered. The performance of these methods is demonstrated and evaluated on a micrograph cell segmentation use case, employing a common state-of-the art deep convolutional neural network (DCNN) architecture exploiting the principle of the common VGG-architecture. The study demonstrates that for this data set, the choice of the ensemble averaging method only has a marginal influence on the evaluation metrics (accuracy and Dice coefficient) used to measure the segmentation performance. Nevertheless, for practical applications, a simple and fast estimate of the mean of the distribution is highly competitive with respect to the most sophisticated representation of the CMP distributions by an alpha-stable distribution, and hence seems the most proper ensemble averaging method to be used for this application.