A Population-Based Tumor-Volume Model for Head and Neck Cancer During Radiation Therapy With a Dynamic Oxygenated Compartment.

PURPOSE: With the coming era of digital medicine and healthcare technology, mathematical modeling of tumors has become a key step to optimize and realize precision radiation therapy. The purpose of this study was to develop a mathematical model for simulating the change of head and neck (HN) tumor volume during radiation therapy. METHODS AND MATERIALS: A formula was developed to describe the dynamic change of oxygenated compartment within a tumor, which was combined with the lethal lesions model to describe various cell processes during radiation therapy, including potentially lethal lesion repair and misrepair, cell proliferation/loss, and tumor reoxygenation. Parameter sensitivity analysis was performed to evaluate the impacts of lesion- and repair-related biological factors on radiation therapy outcomes. RESULTS: We tested our model on 14 available patients with HN cancer and compared the performance with 3 other models. The mean error of our model for the 12 good fit cases was 12.2%, which is considerably smaller than that of the linear quadratic model (19.7%), the generalized linear quadratic model (19.1%), and a 4-level cell population model (16.6%). Correlation analysis results revealed that for small tumors, there was a positive correlation (correlation coefficient r=0.9416) between hypoxic fraction (hf) and tumor volume, whereas the correlation became negative and not significant (r=-0.4365) for large tumors. It is demonstrated from sensitivity analysis that the production rate of lethal lesions (ηl) has a far greater impact on tumor volume than other parameters. The hf had an insignificant impact on tumor volume but had a notable influence on the volume of surviving cells. The final volume of surviving cells athf=0.5 was almost 8 ×102 times that of hf=0.01. The potentially lethal lesion-related parameters (the production rate of potentially lethal lessions per unit dose ηpl, the rate of correct repair per unit time εpl, and the rate of binary misrepair per unit time ε2pl) had rather small impacts (<1%) on both tumor volume and the volume of surviving cells, which indicates that the repaired and misrepaired sublethal cells only take up a small portion of the total cancer cell population. CONCLUSIONS: A population-based tumor-volume model for HN cancer during radiation therapy with a dynamic oxygenated compartment was developed in this study. Comprehensively considering the damage process of tumor cells caused by radiation therapy, the accurate prediction of the volume change of HN tumors during treatment was revealed. Meanwhile, various cell activities and their principles in the process of antitumor treatment were reflected, which has positive clinical reference significance for radiobiology.

Accurate and fast implementation of soybean pod counting and localization from high-resolution image.

Introduction: Soybean pod count is one of the crucial indicators of soybean yield. Nevertheless, due to the challenges associated with counting pods, such as crowded and uneven pod distribution, existing pod counting models prioritize accuracy over efficiency, which does not meet the requirements for lightweight and real-time tasks. Methods: To address this goal, we have designed a deep convolutional network called PodNet. It employs a lightweight encoder and an efficient decoder that effectively decodes both shallow and deep information, alleviating the indirect interactions caused by information loss and degradation between non-adjacent levels. Results: We utilized a high-resolution dataset of soybean pods from field harvesting to evaluate the model's generalization ability. Through experimental comparisons between manual counting and model yield estimation, we confirmed the effectiveness of the PodNet model. The experimental results indicate that PodNet achieves an R2 of 0.95 for the prediction of soybean pod quantities compared to ground truth, with only 2.48M parameters, which is an order of magnitude lower than the current SOTA model YOLO POD, and the FPS is much higher than YOLO POD. Discussion: Compared to advanced computer vision methods, PodNet significantly enhances efficiency with almost no sacrifice in accuracy. Its lightweight architecture and high FPS make it suitable for real-time applications, providing a new solution for counting and locating dense objects.

TasselLFANet: a novel lightweight multi-branch feature aggregation neural network for high-throughput image-based maize tassels detection and counting.

Accurately and rapidly counting the number of maize tassels is critical for maize breeding, management, and monitoring the growth stage of maize plants. With the advent of high-throughput phenotyping platforms and the availability of large-scale datasets, there is a pressing need to automate this task for genotype and phenotype analysis. Computer vision technology has been increasingly applied in plant science, offering a promising solution for automated monitoring of a large number of plants. However, the current state-of-the-art image algorithms are hindered by hardware limitations, which compromise the balance between algorithmic capacity, running speed, and overall performance, making it difficult to apply them in real-time sensing field environments. Thus, we propose a novel lightweight neural network, named TasselLFANet, with an efficient and powerful structure for accurately and efficiently detecting and counting maize tassels in high spatiotemporal image sequences. Our proposed approach improves the feature-learning ability of TasselLFANet by adopting a cross-stage fusion strategy that balances the variability of different layers. Additionally, TasselLFANet utilizes multiple receptive fields to capture diverse feature representations, and incorporates an innovative visual channel attention module to detect and capture features more flexibly and precisely. We conducted a series of comparative experiments on a new, highly informative dataset called MrMT, which demonstrate that TasselLFANet outperforms the latest batch of lightweight networks in terms of performance, flexibility, and adaptability, achieving an F1 measure value of 94.4%, a mAP.@5 value of 96.8%, and having only 6.0M parameters. Moreover, compared with the regression-based TasselNetV3-Seg† model, our proposed model achieves superior counting performance, with a mean absolute error (MAE) of 1.80, a root mean square error (RMSE) of 2.68, and a R2 of 0.99. The proposed model meets the accuracy and speed requirements of the vision system in maize tassel detection. Furthermore, our proposed method is reliable and unaffected by geographical changes, providing essential technical support for computerized counting in the field.

WheatLFANet: in-field detection and counting of wheat heads with high-real-time global regression network.

BACKGROUND: Detection and counting of wheat heads are of crucial importance in the field of plant science, as they can be used for crop field management, yield prediction, and phenotype analysis. With the widespread application of computer vision technology in plant science, monitoring of automated high-throughput plant phenotyping platforms has become possible. Currently, many innovative methods and new technologies have been proposed that have made significant progress in the accuracy and robustness of wheat head recognition. Nevertheless, these methods are often built on high-performance computing devices and lack practicality. In resource-limited situations, these methods may not be effectively applied and deployed, thereby failing to meet the needs of practical applications. RESULTS: In our recent research on maize tassels, we proposed TasselLFANet, the most advanced neural network for detecting and counting maize tassels. Building on this work, we have now developed a high-real-time lightweight neural network called WheatLFANet for wheat head detection. WheatLFANet features a more compact encoder-decoder structure and an effective multi-dimensional information mapping fusion strategy, allowing it to run efficiently on low-end devices while maintaining high accuracy and practicality. According to the evaluation report on the global wheat head detection dataset, WheatLFANet outperforms other state-of-the-art methods with an average precision AP of 0.900 and an R2 value of 0.949 between predicted values and ground truth values. Moreover, it runs significantly faster than all other methods by an order of magnitude (TasselLFANet: FPS: 61). CONCLUSIONS: Extensive experiments have shown that WheatLFANet exhibits better generalization ability than other state-of-the-art methods, and achieved a speed increase of an order of magnitude while maintaining accuracy. The success of this study demonstrates the feasibility of achieving real-time, lightweight detection of wheat heads on low-end devices, and also indicates the usefulness of simple yet powerful neural network designs.

Control of Salmonella on sprouting mung bean and alfalfa seeds by using a biocontrol preparation based on antagonistic bacteria and lytic bacteriophages.

The following reports on the application of a combination of antagonistic bacteria and lytic bacteriophages to control the growth of Salmonella on sprouting mung beans and alfalfa seeds. Antagonistic bacteria were isolated from mung bean sprouts and tomatoes by using the deferred plate assay to assess anti-Salmonella activity. From the isolates screened, an Enterobacter asburiae strain (labeled "JX1") exhibited stable antagonistic activity against a broad range of Salmonella serovars (Agona, Berta, Enteritidis, Hadar, Heidelberg, Javiana, Montevideo, Muenchen, Newport, Saint Paul, and Typhimurium). Lytic bacteriophages against Salmonella were isolated from pig or cattle manure effluent. A bacteriophage cocktail prepared from six isolates was coinoculated with E. asburiae JX1 along with Salmonella in broth culture. The combination of E. asburiae JX1 and bacteriophage cocktail reduced the levels of Salmonella by 5.7 to 6.4 log CFU/ml. Mung beans inoculated with Salmonella and sprouted over a 4-day period attained levels of 6.72 + or - 0.78 log CFU/g. In contrast, levels of Salmonella were reduced to 3.31 + or - 2.48 or 1.16 + or - 2.14 log CFU/g when the pathogen was coinoculated with bacteriophages or E. asburiae JX1, respectively. However, by using a combination of E. asburiae JX1 and bacteriophages, the levels of Salmonella associated with mung bean sprouts were only detected by enrichment. The biocontrol preparation was effective at controlling the growth of Salmonella under a range of sprouting temperatures (20 to 30 degrees Celsius) and was equally effective at suppressing the growth of Salmonella on sprouting alfalfa seeds. The combination of E. asburiae JX1 and bacteriophages represents a promising, chemical-free approach for controlling the growth of Salmonella on sprouting seeds.

Evaluation of a biocontrol preparation consisting of Enterobacter asburiae JX1 and a lytic bacteriophage cocktail to suppress the growth of Salmonella Javiana associated with tomatoes.

A biocontrol preparation based on a combination of Enterobacter asburiae JX1 and a cocktail of five lytic bacteriophages was evaluated for control of Salmonella Javiana within the rhizosphere of plants and in pre- and postharvest tomatoes. The biocontrol preparation introduced into the rhizosphere of growing tomato plants reduced the persistence of Salmonella, although no synergistic action was observed between E. asburiae JX1 or the bacteriophage cocktail when used in combination. When the biocontrol preparation was coinoculated with Salmonella onto the blossom of tomato plants, the prevalence of the enteric pathogen both on the surface and in internal tissues of the subsequent tomatoes was significantly reduced compared with controls. Tomatoes derived from plants inoculated with Salmonella alone had a prevalence of 92% surface contamination (22 of 24 tomato batches were positive for Salmonella) and 43% internal contamination (31 of 72 batches positive). This Salmonella prevalence was reduced to 0% (0 of 38 positive) and 2% (1 of 57 positive), respectively, when the biocontrol preparation was applied. Although bacteriophages reduced the prevalence of internalized Salmonella, the main growth suppressing effect was via the antagonistic activity of E. asburiae JX1. No bacteriophages were recovered from tomatoes despite being introduced at 6 log PFU onto the blossom of plants. The biocontrol preparation was not effective for controlling the growth of Salmonella introduced onto postharvest tomatoes that were stored for 7 days at 15 degrees C. The application of E. asburiae JX1 is a promising approach for controlling Salmonella encountered in tomato production, and there was no evidence to suggest that the antagonistic activity could be enhanced by the coinoculation of bacteriophages.