Browse Articles

Early identification of the efficacy of 0.125% atropine treatment for children with Myopia: A prospective pilot study

Purpose This study aimed to investigate whether early axial length (AL) changes in the short term after 0.125% atropine treatment could predict long-term axial elongation in children with myopia. Methods This was a prospective cohort study involving children aged 5–15 years with myopia who were treated with 0.125% atropine for myopia control. AL was measured 1–2 months after starting treatment and then every 3 months for follow-up visits. Regression analysis was used to develop a model of AL changes with time. A generalized estimating equation (GEE) model was then used to identify correlations between the early AL changes and long-term AL changes. Results Eighty eyes of 40 patients (mean age 8.4 years) were included in the final analysis. The estimation curve of AL changes with time indicated that the AL decreased at 67 days (the turning point in the regression model) after 0.125% atropine treatment and then increased gradually with time. Univariate GEE showed that a larger AL elongation in the initial 4 months was significantly associated with AL changes at 6 months (β = 0.354, P = 0.020, 6 ~ 12 months period from baseline) and 12 months (β = 0.560, P = 0.045, 6 ~ 18 months period from baseline) after that period in all myopic eyes. Conclusions The magnitude of AL elongation in the initial 4 months of 0.125% atropine treatment correlated positively with the further half-year and one-year AL changes. Identifying these changes may be useful for controlling refractory myopia in children.

Study on the effect of light distribution on the greenhouse environment in Chinese solar greenhouse

Solar greenhouse is a primary agricultural facility in northern China during winter, providing a certain level of security for the demand for vegetables and melons in the northern regions. However, there remains a lack of uniformity between crop requirements and the light and thermal environment within the planting area of the greenhouse, resulting in non-uniform growth and development of crops. The present study set out with the objective of investigating the impact of the light environment on the internal conditions of a solar greenhouse. To this end, experimental measurements were employed in conjunction with deep learning models. The results showed that rates of change in air temperature and light intensity were significantly higher in the vertical than the horizontal direction, especially below 1,800 metres, where significant differenced in temperature and light distribution existIn the horizontal direction, the impact of light distribution on soil temperature was significant within a range of less than 4,500 mm from the southern base of the greenhouse. By contrast, the impact was less pronounced within a range of 4,500 to 9,000 mm, In the temporal dimension, light variation significantly affected soil temperatures within 150 mm of the surface, but had no significant effect on temperatures within the 300–600 mm range. Similarly, light variation significantly affected temperatures within 200 mm of the inner wall surface, but had no significant effect on temperatures within the 400–800 mm range.Furthermore, vertical differences in light intensity significantly affected temperatures within the 800 mm height range from the indoor ground level, whereas the impact at other heights was less pronounced. The LSTM prediction model was highly accurate, and this study provided the necessary data and theoretical basis for regulating the light and temperature environments in solar greenhouse.

Improving pediatric care in Uganda with a digital platform and quality improvement initiative: A retrospective review of Smart Triage + QI

Objective This is a retrospective review of the feasibility study and implementation of the Smart Triage and Quality Improvement (QI) initiative at Holy Innocents Children’s Hospital (HICH), a dedicated pediatric hospital in Mbarara, Uganda, over a 5-year period. The aim of this QI initiative was to improve triaging rates and the time-to-antimicrobials in HICH’s outpatient department (OPD). Methods Smart Triage is a risk prediction algorithm and digital platform that enables healthcare workers to triage patients and track treatments effectively. Following the feasibility study, the QI program was implemented in September 2021 using three Plan-Do-Study-Act cycles: 1) Standardize Training, 2) Adjust Workflows, and 3) QI Team Communication. Data sources were triage and hospital reports. Monthly run charts of OPD attendance, acuity of illness, triaging rates, median-time-to-antimicrobials, and mortality rates of admitted patients were created. The trajectories of the variables were assessed using linear regression with time as the explanatory variable. Results 121,521 children attended HICH OPD from November 2018 to October 2023. The OPD triaging rate increased to 91% by October 2023, with a sustained plateau above 90% since July 2022. There was a significant reduction in the median time-to-antimicrobials during the 5-year period, from 77.6 to 53.6 minutes, with a slope of −0.4 minutes per month (CI: −0.73 to −0.04, p-value: 0.029). The inpatient mortality rate decreased from 5.1% in August 2018 to 2.6% in October 2023, with a significant increase in the number of cases with comparable illness severity. Conclusion The impact of Smart Triage was sustained beyond the end of the feasibility trial and showed sustained improvements in processes such as treatment times and clinical outcomes including a reduction in mortality. HICH’s leadership integrated a culture of QI across disciplines and departments, contributing to this initiative’s sustainability and impact.

Health-related quality of life and QALY loss under COVID-19 lockdown: The case of Spain

Objectives The COVID-19 pandemic forced many countries to implement confinement measures to limit the spread of the virus. Measuring the loss in terms of quality-adjusted life-years (QALYs) may provide a commensurable basis for comparing the impact of COVID-19. The aim of this research was to explore the impact of the first 21 days of COVID-19 lockdown on health-related quality of life (HRQoL) and associated QALY loss of the Spanish general population. Methods A quota-based online survey was conducted in four waves with 500 general population respondents each: one conducted shortly before the lockdown (baseline) and 3 follow-ups conducted weekly. HRQoL data were collected using EQ-5D-5L. For comparison with pre-covid responses, data from the 2011–2012 National Health Survey was taken as reference. Data were analyzed using frequency analysis and logistic regression. QALY loss was estimated over the follow-up period and for the entire duration of the lockdown. Results Comparing the baseline results to the follow-up results shows little change with respect to the distributions of reported problems in any of the 5 dimensions during the follow-up period. However, results for anxiety/depression show a 32% increase in the proportion of reported problems. The Spanish population was estimated to accrue a total of 1,994,216 QALYs over the study period. Based on the reference data, the population should have accrued 2,054,737 QALYs, leading to a loss of 60,520 QALYs over 21 days. For the entire lockdown, the corresponding loss would be 285,310 QALYs. Conclusions A population under a lockdown situation reported higher rates of anxiety/depression problems than in a regular situation. On a country-wide scope, this may lead to a substantial loss in terms of QALYs, especially over longer periods of time. This is the first study to directly assess the impact of the lockdown in terms of QALY loss on a country-wide level.

Data anomalies and the economic commitment of climate change

Robust skeletal motion tracking using temporal and spatial synchronization of two video streams

Accurate and reliable skeletal motion tracking is essential for rehabilitation monitoring, enabling objective assessment of patient progress and facilitating telerehabilitation applications. Traditional marker-based motion capture systems, while highly accurate, are costly and impractical for home rehabilitation, whereas marker-less methods often suffer from depth estimation errors and occlusions. Recent studies have explored various computer vision and deep learning approaches for human pose estimation, yet challenges remain in ensuring robust depth accuracy and tracking under occlusion conditions. This study proposes a three-dimensional human skeleton tracking system for upper limb activities that integrates temporal and spatial synchronization to improve depth estimation accuracy for rehabilitation exercises. The proposed system combines a 90° secondary camera to compensate for the depth prediction inaccuracies inherent in single-camera systems, reducing error margins by up to 0.4 m. In addition, a linear regression-based depth error correction model is implemented to refine depth coordinates, further improving tracking precision. The Kalman filtering framework is employed to enhance temporal consistency, allowing real-time interpolation of missing joint positions. Experimental results demonstrate that the proposed method significantly reduces depth estimation errors of the elbow and wrist joint (p < 0.001) compared to single camera setups, particularly in scenarios involving occlusions and non-frontal perspectives. This study provides a cost-effective and scalable solution for remote patient monitoring and motor function evaluation.

Reply to: Inconclusive proof of ferroelectricity in peptide-VDF ribbons

Effects of adding N2-fixing Rhodopseudomonas palustris to stimulate the growth and yield of canary melon (Cucumis melo L.)

This study evaluated the effects of a mixture of four N2-fixing strains of Rhodopseudomonas palustris-VNW64, VNS89, TLS06, and VNS02-(PNSB) on soil properties, nitrogen (N) uptake, plant growth, and yield of canary melon cultivated in alluvial soil. A greenhouse experiment was conducted using a completely randomized block design with eight treatments: (i) 100% N of recommended fertilizer formula (RFF), (ii) 85% N of RFF, (iii) 70% N of RFF, (iv) 100% N of RFF + PNSB, (v) 85% N of RFF + PNSB, (vi) 70% N of RFF + PNSB, (vii) PNSB only, and (viii) no fertilization. The application of PNSB improved soil pH and available N concentrations. The highest N uptake (33.9 kg N ha ⁻ ¹) was recorded in the 100% RFF + PNSB treatment. Notably, the 70% RFF + PNSB treatment achieved comparable N uptake (27.7 kg N ha ⁻ ¹) to the 100% RFF treatment (28.6 kg N ha ⁻ ¹). The 85% RFF + PNSB treatment maintained plant height and yield equivalent to the 100% RFF treatment. These results suggest that supplementing with PNSB can reduce N fertilizer application by up to 15% without compromising crop performance. The PNSB mixture should be further tested under a field trial.

Comparison of negative pressure wound therapy with conventional wound care in the treatment of sternal wound infection after cardiac surgery: A meta-analysis with trial sequential analysis

Background Negative pressure wound therapy (NPWT) has become a popular treatment option for sternal wound infection (SWI). However, it remains uncertain whether the therapeutic benefits of NPWT are superior to conventional wound care. This study aimed to systematically evaluate the therapeutic effects of NPWT on SWI compared to conventional wound care through meta-analysis. Methods A comprehensive search of PubMed, Web of Science, Embase, and the Cochrane Library databases was conducted from inception to April 29, 2024 for all potential studies. The pooling of dichotomous outcome data was achieved using relative risk (RR), with results presented within a 95% confidence interval (CI). We utilized the standard mean difference (SMD) and 95% CI for continuous outcomes. Heterogeneity test, publication bias assessment, sensitivity analysis, and trial sequential analysis (TSA) were conducted. Publication bias was detected through the Begg’s and Egger’s tests. Software R 4.3.1, Stata 12.0, and TSA v0.9.5.10 Beta software were utilized for all analyses. Results Out of 1832 articles identified, 10 were included in this study. The overall results revealed that NPWT significantly decreased the sternal wound reinfection (SWRI) rate (RR [95% CI] = 0.179 [0.099 to 0.323], 95% prediction interval [PI]: 0.082 to 0.442), in-hospital mortality (RR [95% CI] = 0.242 [0.149 to 0.394], 95% PI: 0.144 to 0.461), and shortened the length of intensive care unit (ICU) stay (SMD [95% CI] = −0.601 [−0.820 to −0.382], 95% PI: −1.317 to 0.128) compared with conventional wound care. There was no significant difference in length of hospital stay (SMD [95% CI] = −0.402 [−0.815 to 0.012], 95% PI: −1.801 to 0.998) and treatment duration (SMD [95% CI] = −0.398 [−1.646 to 0.849], 95% PI: −16.340 to 15.543) between the NPWT group and control group. Further subgroup analysis demonstrated the benefits of NPWT in shortening hospitalization length in the European population (p < 0.05). Conclusion The present evidence corroborates that the application of NPWT in the treatment of SWI after cardiac surgery effectively reduces the SWRI incidence and in-hospital mortality while shortening the length of ICU stay.

Improving the prediction of potato yield gaps: Solanum-model parameterization and evaluation in southwestern China

Sustainable agriculture has made significant contributions to both food security and global economic growth. Development of new cultivars and identifying and utilizing the yield potential of existing gemplasms are two key options for sustainable crop production. In this study, field trials and a new plant growth stimulator model developed by the International Potato Center called “Solanum” was used to assess the yield potential and ‘yield gaps’ of three potato cultivars in spring, autumn, and early spring seasons in southwestern China. The results showed that the average potential yield of potato crops in spring, early spring, and autumn seasons was 125.6 t/ha, 56.40 t/ha, and 45.30 t/ha with a gap from the potential yield of 107.30 t/ha, 36.70 t/ha, and 32.10t/ha, respectively. Further analysis revealed that the late blight disease was the main cause of large yield gap in spring season, whereas inadequate rain fall was the the major factor impacting the actual yield of potato crops in autumn and early spring seasons. Therefore, we report for the first time, that the spring potato in Yunnan Province, southwestern China has the highest potential yield in the world, and that extending growing season coupled with managing late blight can increase actual yield by 115%. The high yield potential of spring potato could be very beneficial to the local economy and add an enormous pelf and prosperity to the region.

Instantaneous center of rotation, the first step to build up the digital laboratory of complex motions

Calculating instantaneous centers of rotation to describe combined rotational and translational motions has a long history in many fields of applied science and basic rigid body kinematics. However, only some theoretical studies have explored the fundamental characteristics of this system. This study used digital three-dimensional modeling and computing methods to examine the system’s operation in a controlled in vitro-like environment. The effects of inaccurate registrations on the resulting motion were also analyzed. We registered 28.65, 14.33 and 9.55 EcD_ratios for 2°, 4° and 6° of closure respectively, and described a structured, predictable framework based on a solid mathematical background. Our findings align with previous publications, indicating that the longstanding debate over the pure rotation of the temporomandibular joint arises from misinterpretations of scientific findings due to a lack of fundamental knowledge of the basic characteristics of the system. Our simplified geometrical approach significantly reduces the complexity of the existing complex kinematic model, making it more accessible for practical applications, easier to understand, and potentially more applicable in orthopedics or temporomandibular joint radiology. We identified five fundamental characteristics of the system as we described the effects of the acting translational component in the complex motion. We also presented a detailed model concerning the effects of inaccurate rotation axis registration on the resulting compromised transformation, improving our understanding of the error tolerance level of articulation systems. Our results show that the system might tolerate errors as great as 3–4 cm in some settings in the parallel error direction, while in case of circular and perpendicular error types an approximately 2 mm axis registration error would exceed the clinically desirable 0.1 mm occlusal error level. Our experimental modeling strategy might provide extensive data for machine learning and for analyzing and comprehending the fundamental characteristics of different complex motion systems in the future.

Antibiotic self-medication in Otuke District, Northern Uganda: Prevalence and associated factors

Antibiotic self-medication is a growing public health concern, particularly in low- and middle-income countries where access to healthcare is limited. The practice contributes to antimicrobial resistance which increases health care costs, morbidity, and mortality in the population. This study aimed to investigate the prevalence of antibiotic self-medication and its associated factors in the Otuke District, Northern Uganda. A community-based cross-sectional study was conducted in Otuke. Data was collected among adults aged 18 years and above using semi-structured questionnaire. Collected data was coded and double-entered into SPSS Software version 26 and exported to STATA 14 for analysis of frequencies and percentages. Modified Poisson regression was used to run analysis of the association at a P-value of 0.05. Out of the 385 participants, 261 (67.79%) reported having self-medicated with antibiotics in the past six months. The most commonly self-medicated antibiotics were amoxicillin 134 (51.3%), ampiclox 87 (33.3%) and metronidazole 57 (21.9%). Participants with previous successful treatment were 2.33 times more likely to self-medicate (PR = 2.33, 95% CI: 1.89–2.87, P < 0.001) while poor staff attitude increased the likelihood by 1.53 times (PR = 1.53, 95% CI: 1.38–1.71, P < 0.001). Knowledge about antibiotics was negatively associated with ASM in that those who had knowledge on antibiotics were 25% less likely to self-medicate with antibiotics (PR = 0.75, 95% CI: 0.65–0.86, P < 0.001). The practice of antibiotic self-medication was highly prevalent in Otuke district due to previous successful treatments and poor health care systems. The commonly self-medicated antibiotics were amoxicillin, ampiclox and metronidazole. We recommend public health interventions such as community education on antimicrobial resistance regulation of antibiotic use in the country.

Designing and evaluation of the effect of community-based intervention on breast self-examination among reproductive-aged women in Ethiopia: A Cluster Randomized Controlled Trial

Background Though early intervention saves many lives worldwide, breast cancer remains a leading cause of cancer among women in Ethiopia. This study, therefore, aimed to evaluate community-based interventions promoting breast self-examination using the Health Belief Model. Methods A cluster randomized controlled trial followed by a cross-sectional study lasting six months was used to evaluate the effectiveness of the community-based educational intervention on breast self-examination among reproductive-aged women in Ethiopia. A total of 810 participants were randomly assigned in a 1:1 ratio and assessed at baseline, three months, and six months post-intervention. A general linear model for repeated measures was used to examine the mean differences in study variables across time points. Non-parametric tests (Cochran’s Q) were employed to analyze dichotomous variables related to breast self-examination practices. Path analysis was conducted to examine the interactions among the constructs of the Health Belief Model. Results A total of 810 reproductive-aged women participated in the study, yielding a 100% response rate at baseline. The mean age of participants was 33.2 ± 7.7 years in the intervention group and 33.5 ± 8.1 years in the control group. The proportion of women performing breast self-examinations increased from 33.3% at baseline to 59.9% at the end of the intervention. And the Comprehensive knowledge about breast self-examination rose from 11.7% to 69.1% over the same period. Perceived susceptibility, perceived severity, knowledge, and health motivations had a statistically significant mean difference between the intervention and control groups (p < 0.0001). We registered PACTR database (https://pactr.samrc.ac.za/): “PACTR201802002902886”. Conclusions The study found that there is a strong interplay between the likelihood of performing breast self-examination and perceived susceptibility, perceived severity, knowledge, and health motivations. Field specialists should figure out the problem related to perception and awareness through intensive health promotion interventions. Trial registration Registered in the Pan African Clinical Trial Registry (www.pactr.org) database, and the unique identification number for the registry is PACTR201802002902886.

Virus and viroid diversity in hops, investigating the German hop virome

Germany is worldwide one of the largest hop (Humulus lupulus L.) producers, an essential crop for the brewing industry. However, infections caused by viruses and viroids can severely impact hop yield and quality. In 2019, citrus bark cracking viroid (CBCVd) – a highly aggressive pathogen in hop – was first reported in Germany, raising concerns about its spread and prompting a broader investigation of the German hop virome.To investigate the viro-diversity in German hops, we started with a pilot study in 2021 targeting three hopyards in the Hallertau region (Bavaria), where CBCVd was previously detected. This study was expanded in 2022 and 2023 to include other main hop growing regions of Tettnang (Baden-Wuerttemberg) and Elbe-Saale (Saxony, Saxony-Anhalt, Thuringia). Leaf samples were collected from hop as well as non-hop plants inside and outside the hopyard, pooled, and proceeded for double-stranded RNAs extraction. High-throughput sequencing (HTS) was used as a diagnostic tool, followed by RT-PCR confirmation. Our analysis identified four viruses infecting hops; hop latent virus (HpLV), hop mosaic virus (HpMV), apple mosaic virus (ApMV), arabis mosaic virus (ArMV) – and two viroids; hop latent viroid (HLVd) and CBCVd. HpLV, HpMV, and HLVd were consistently found across all targeted hopyards, while CBCVd was confined to the Hallertau region. ArMV was only detected in one hopyard at one sampling timepoint. ApMV was the only virus detected in both hop and non-hop plants. Additional analysis of hop pool datasets revealed the presence of other potential hop pathogens, i.e., fungi and bacteria. The results showed a low diversity of viruses and viroids infecting hops. However, this study provides a comprehensive overview on the major viruses and viroids in German hopyards. The results may serve as a useful resource for the development of disease management strategies in hop cultivation and highlight the valuable implementation of HTS in plant pathogen surveillance.

Incidence of dry eye symptoms and behavioural-cultural risk factors among university students population in Jordan

Purpose To estimate the incidence of dry eye (DE) symptoms among university students in Jordan and to examine the relationship between behavioral and cultural risk factors and DE symptom severity. Methods A cross-sectional study involving 788 university students was conducted in Jordan. Participants’ mean age was 21.87 years (SD = 3.824; range: 18–45 years). The incidence and severity of DE symptom were assessed using the validated Arabic version of the Ocular Surface Disease Index (ARB-OSDI) questionnaire, administered through Google Forms. The survey included demographic questions and behavioral-cultural risk factors (smoking and eye cosmetic use). One-Way ANOVA and multi-regression analyses were used to investigate the association between OSDI mean scores and behavioral-cultural risk factors. Results The incidence of DE symptoms, defined as an OSDI score ≥ 13, was 74.2% among university students. Higher DE symptom severity was statistically associated with females’ gender (p < 0.001), older age (≥27 years) (p = 0.032), contact lens use (p = 0.001), frequent use of eye cosmetics (p < 0.001), and a history of DED (p < 0.001). Smoking habits, including the use of Dokha or Ajami, smoking in enclosed spaces, and daily smoking, were also associated with increased DE symptom severity (all p < 0.001). Contributing factors to the high incidence and severity of DE symptom included long-term use of eye cosmetics (particularly mascara and internal eyeliner) and sleeping while wearing contact lenses. Conclusion Dry eye symptoms are highly prevalent among university students in Jordan and are significantly associated with factors such as age, gender, contact lens use, cosmetic application, and tobacco consumption.

AI learns from nature to design super-adhesive gels that work underwater

GNN-RMNet: Leveraging graph neural networks and GPS analytics for driver behavior and route optimization in logistics

Logistics networks are becoming increasingly complex and rely more heavily on real-time vehicle data, necessitating intelligent systems to monitor driver behavior and identify route anomalies. Traditional techniques struggle to capture the dynamic spatiotemporal relationships that define driver actions, route deviations, and operational inefficiencies in big fleets. This paper introduces GNN-RMNet, a hybrid deep learning system that combines GNN, ResNet, and MobileNet for interpretable, scalable, and efficient driver behavior profiling and route anomaly detection. GNN-RMNet utilizes spatiotemporal GPS trajectories and vehicle sensor streams to learn contextual and relational patterns from structured driving data in real time, thereby identifying dangerous driving and route violations. On a real-world GPS-vehicle sensor dataset, the proposed model achieves 98% accuracy, 97% recall, an F1-score of 97.5%, and domain-specific measures like Anomaly Detection Precision (96%) and Route Deviation Sensitivity (95%). Modular design offloads ResNet-GNN analytics to edge nodes while preserving MobileNet components for on-vehicle inference, resulting in reduced inference latency (32 ms). Comparing GNN-RMNet against baseline, ensemble, and hybrid models shows its accuracy, efficiency, and generalization advantages. Computational feasibility, anomaly scoring interpretability, and future deployment concerns, including cybersecurity, data privacy, and multimodal sensor integration, are all covered. For real-time fleet safety management and secure, intelligent, and context-aware logistics, GNN-RMNet seems promising. The framework incorporates multimodal, privacy-aware, and scalable driver analytics, enabling its use in intelligent transportation systems and urban logistics infrastructures.

Is your AI benchmark lying to you?

BBDetector: Intelligent border binary detection in IoT device firmware based on a multidimensional feature model

In the field of firmware security analysis for Internet of Things (IoT) devices, border binary detection has become an important research focus. However, the existing methods for border binary detection have problems such as insufficient feature characterization, high false-negative rates, and low intelligence levels. To mitigate these issues, we introduce BBDetector, a border binary detection method based on a multidimensional feature model. First, we constructed the first known set of border binaries at a certain scale by collecting and analyzing a diverse set of real-world firmware. To characterize the features of border binaries comprehensively, we proposed a multidimensional feature model (MDFM). Next, we extracted the feature vectors of binaries via the MDFM and designed a novel stacking method to achieve border binary detection. This method involves ensemble learning, combining extreme gradient boosting, light gradient boosting machine, and categorical boosting as base learners with random forest as the meta-learner. Finally, a border binary detection model (XLC-R) was obtained by training with feature vectors. We tested and evaluated BBDetector on two datasets. The experimental results showed that XLC-R achieved a precision of 94.98%, a recall of 91.02%, and an F1 score of 92.84% for the constructed representative Dataset I. Additionally, BBDetector detected 3.25 times and 2.23 times more border binaries in Dataset II than did the state-of-the-art tools Karonte and SaTC, respectively. BBDetector provides an accurate method for border binary detection in IoT firmware security analysis, significantly enhancing the pertinence of vulnerability detection, dramatically reducing the complexity of firmware security analysis, and providing essential technical support for improving IoT device security.

Evaluation of collapsible deformation of foundation under rectangular load based on the improved binary medium model

The increasing frequency of extreme weather events and climate change can substantially impact the collapse phenomenon and other challenges associated with the deformation of foundation soils. These can also affect soil moisture regimes, particularly soil suction. The global engineering and geotechnical hazards related to the deformation of foundation soil collapsibility require immediate attention from engineers. The differential equations of the collapsible consolidation deformation of a collapsible loess foundation under concentrated force are formulated using an improved two-dimensional medium model in conjunction with the Biot consolidation theory, fracture mechanics, and continuum theory. The equations are solved using the mathematical and physical methodologies of the Laplace transform and the Hankel transform, and boundary conditions are introduced. The mathematical models of lateral displacement, vertical displacement, and pore water pressure of a collapsible loess foundation with vertical depth, radial distance, and saturation under rectangular load are provided. The proposed model was validated through a series of numerical calculations and analyses. It was demonstrated that the deformation of the collapsible loess foundation under the improved binary medium rectangular load is exceedingly similar to the corresponding engineering deformation. The results of the investigation significantly impact the theoretical research of collapsible loess foundations.