Research Articles (Statistics)

Permanent URI for this collectionhttp://hdl.handle.net/2263/1835

A collection containing some of the full text peer-reviewed/ refereed articles published by researchers from
the Department of Statistics

Browse

Now showing 1 - 20 of 641

Alternative priors for a contaminated exponential model evaluated in an insurance loss context
Krüger, Ingrid; Ferreira, Johan; Bekker, Andriette, 1958- (Springer, 2026-08)
This paper builds on the Bayesian framework for inference in the contaminated exponential model, with particular emphasis on the role of prior assumptions regarding the proportion of outliers. While a beta prior is standard in this context, we investigate alternative prior choices that offer greater flexibility in modelling atypical observations. Through simulation studies that span a wide range of parameter configurations, we demonstrate that these alternative priors can yield more insightful inference, particularly in small-sample settings where Bayesian methods are known to offer clear advantages. The practical relevance of the proposed approach is illustrated using a large insurance loss dataset and a smaller random subsample thereof. In both cases, the methodology enables identification of potential outliers by integrating full conditional posterior probabilities, which combine both data-driven evidence and prior assumptions. This unified perspective contrasts with the frequentist approach, where outlier detection depends solely on the observed data. To assess the sensitivity of inference to prior selection, we utilise the Wasserstein Impact Measure, which offers a clear measurement of the effect of prior specifications on posterior conclusions. Overall, the framework enhances both the theoretical understanding and practical implementation of outlier detection in contaminated exponential models, offering a valuable contribution to actuarial science and applied statistical modelling.
New distribution-free group runs Mann-Whitney U control charts for detecting mean shifts with application in mining ore process
Masilela, Busisiwe; Malela-Majika, Jean-Claude; Chaka, Lyson (Taylor and Francis, 2026)
Statistical process control (SPC) is a field of Statistics that uses statistical methods (or tools) to improve the quality of products and/or services. Among the tools used in SPC, control charts are the most popular. Most of the existing charts are designed based on the assumption of normality. These charts are known as parametric charts. When information about the type or nature of the underlying process distribution is not available, researchers recommend the use of nonparametric charts. This paper introduces new distribution-free group runs Shewhart, cumulative sum (CUSUM) and exponentially weighted moving average (EWMA) charts by combining the existing one-sided Shewhart, CUSUM or EWMA Mann-Whitney (MW) U sub-chart with an extended conforming run-length sub-chart. The performances of the resulting charts are investigated in terms of the characteristics of the distribution of the time to signal. The new charts are also compared to the existing basic and synthetic one-sided Shewhart, CUSUM and EWMA MW U charts. It was found that the new charts have very interesting properties as compared to the competing charts considered in this paper. A real-world silica concentrate data from a mining ore process is used to demonstrate the application and implementation of the new charts.
Lower limb clinical osteoarthritis and its association with joint pain and function, and severe injuries and surgeries, in women professional football players : a cross-sectional observational study
Ramagole, Dimakatso Althea; Janse van Rensburg, Dina Christina; Bennett, P.; Cowie, C.; Maas, M.; Mehta, R.; Ramkilawon, Gopika Devi; Rantanen, J.; Sallo, J.; Pluim, Babette M.; Kerkhoffs, G.; Gouttebarge, Vincent (South African Sports Medicine Association, 2025)
BACKGROUND : Athletes may sustain severe injuries and need to undergo surgery, potentially leading to osteoarthritis (OA). Early diagnosis and rehabilitation can significantly improve outcomes and extend an athlete's career. OBJECTIVES : To determine: 1) prevalence of clinical hip, knee, and ankle OA; 2) association with pain and function; 3) association with severe injuries and surgeries in female professional footballers. METHODS : This cross-sectional study involved female professional footballers who completed online questionnaires on joint pain and function, severe injuries, and surgeries. Their physicians evaluated them for clinical OA (cOA) of the hips, knees, and ankles. RESULTS : Among 74 participants, cOA prevalence of the hips, knees, and ankles was 2.7% (95% CI: 0-6.4), 5.0% (95% CI: 2.5-10.6), and 8.0% (95% CI: 1.9-14.3), respectively. Pain was present in the hips (p=0.615; Cramer's V=0.132), knees (p=0.556; Cramer's V=0.171), and ankles (p=0.028; Cramer's V=0.391). Joint function was impaired in the hips (p=0.379; Cramer's V=0.214), knees (p=0.738; Cramer's V=0.103), and worse in the ankles (p=0.0778; Cramer's V=0.255). A weak association existed between hip cOA and injuries (p=0.230; Cramer's V=0.193), with moderate associations in the knees (p=0.024; Cramer's V=0.290) and ankles (p=0.168; Cramer's V=0.206). Surgeries were associated with knee cOA (p=0.0578; Cramer's V=0.258) and not ankles (p=1.00; Cramer's V=0.027. CONCLUSION : The prevalence of clinical OA was low among female footballers. Pain was the primary symptom, accompanied by impaired joint function. Severe knee injuries and surgeries were linked to cOA. Further research is recommended in this population.
Injury has a lower incidence but higher burden than illness in elite South African netball players : a prospective cohort study
Pistorius, Danielle; Jansen van Rensburg, Audrey; Janse van Rensburg, Dina Christina; Ramkilawon, Gopika Devi; Ramagole, Dimakatso Althea (South African Sports Medicine Association, 2025)
BACKGROUND : Netball poses significant injury risks with explosive movements and restricted court areas. Team sports with frequent contact predispose players to illness. OBJECTIVES : Over six months, the period prevalence, incidence, clinical characteristics, and severity of injuries and illnesses in elite South African (SA) netball players were investigated. METHODS : A six-month prospective cohort study followed 24 elite female national team players (age 25.7±4.8 years). Injury and illness data were self-reported via a two-weekly online survey. Main outcome variables included period prevalence (PP, %), incidence (I; injury/1000 player-hours; illness/1000 player-days), severity (time-loss in days), injury burden (InjB; days lost/1000 player exposure-hours) and illness burden (IllB; days lost/1000 player exposure-days). RESULTS : In total, 26 injuries were reported (PP=9%; I=5.4). Lower limb injuries (83%), specifically the knee (44%), involving joint sprains/ligament tears (28%), mainly occurred. Most injuries (55%) were sustained during match play (56%) and due to player contact (39%). Goal defenders suffered most injuries (22%). Injuries resulted in mild time-loss (1–7 days), with an InjB of 3.5 days lost/1000 player exposure-hours. In total 33 illnesses were reported (PP=11%, I=6.9), mainly involving the upper respiratory tract (85%). Illness severity was mostly minimal, with no time-loss (52%), resulting in an IllB of 17.3 days lost/1000 player exposure-days. CONCLUSION : This study provides important descriptive injury and illness information in elite netball players. InjB translates to 14.9 days lost, and IllB to 3.1 days lost during the six months. Player contact is a major cause of injury. Lower limb injuries, contact prevention and upper respiratory tract illnesses should receive attention in preventative strategies. More injury and illness data in elite netball is needed.
The impact of occupational therapy using Ayres Sensory Integration® on a child with cochlear implants : a case study
Kruger, Stefanie; Uys, Kitty; Van Niekerk, Karin; Botha, Tanita (Occupational Therapy Association of South Africa, 2025-08)
INTRODUCTION : A three-year-old child with bilateral cochlear implants (CIs) was referred to occupational therapy by the cochlear team due to suspected tactile over-reactivity. The child's behaviour fell outside of typical expectations of young children who receive CIs and had a ripple effect on her language and development e.g. removing and hiding her CIs after 45 - 60 min of usage. METHODS : A retrospective, longitudinal, experimental single case study was conducted. Occupational therapy using the Ayres Sensory Integration® approach (OT-SI) was provided to the child for 48 months in a private practice in South Africa. This study aimed to show that the child's sensory perception and regulation improved overtime as a result of intervention. Results of the Sensory Integration and Praxis Tests completed at 23 and 42 months after the onset of intervention were compared. OUTCOMES : After OT-SI intervention, the child tolerated and enjoyed wearing her CIs during waking hours every day. Test results and her increased participation in occupations confirmed that the child showed significant improvement. CONCLUSION : OT-SI made a meaningful difference for this child and should be considered as part of the intervention for children with CIs. The parents and cochlear team were satisfied with the positive outcomes of intervention. The study highlights the potential contribution of the occupational therapist to the cochlear team. IMPLICATION FOR PRACTICE • This case study provides valuable insights to promote early intervention to support occupational therapists working with young children with cochlear implants. • Clinical strategies are shared to stimulate professional reasoning and articulate goal setting for a child-directed intervention plan. • Knowledge is shared to empower occupational therapists to make a valuable contribution to the cochlear team and to fill the gap in available literature. • The article reflects the value of a support system for parents of children with cochlear implants. • Fostering continuous communication between team members can create an atmosphere of empathy and recognition of each family's unique journey.
Generating a combined analysis using linear mixed models to account for pre-history effects when a treatment factor comprises fixed and random levels
Chaka, Lyson; Njuho, Peter; Piepho, Hans-Peter; Malela-Majika, Jean-Claude (Cambridge University Press, 2026-04-01)
Experimental designs involving factors with a mix of fixed and random levels have been explored by few. These designs are useful when comparing a set of new treatments (fixed levels) to a population of established treatments (random levels). This approach enables partitioning variability and testing of both fixed and random effects, leading to improved estimates and more reliable inference. However, combining analyses of variance from partitioned data poses challenges, including data rearrangement, methodology, and coding complexities. Existing statistical software does not directly support combined analyses for such designs. The current study provides guidelines for conducting combined analysis of variance in linear mixed-effects models where factors have both fixed and random levels. The approach utilizes SAS PROC GLIMMIX and tools from the Comprehensive R Archive Network to compile the combined analysis of variance, followed by multiple comparisons of treatment means. The procedure is demonstrated using treatment structures in a completely randomized design, split-split-plot design, and repeated-measures design. The method can be extended to other experimental designs. This framework addresses a critical gap in existing literature by providing a practical and readily applicable analytical tool that enables integrated analysis across all factor levels within a single linear mixed-model structure, while accounting for pre-history effects associated with prior management practices in complex experimental systems.
Pooled analysis of PCV13 efficacy from controlled human infection trials in Malawi and the UK
Kudowa, Evaristar; Tembo, Godwin; Chirwa, Anthony E.; Chikaonda, Tarsizio; Muyaya, Alfred; Makhaza, Lumbani; Nsomba, Edna; Galafa, Bridgette; Thole, Faith; Ndaferankhande, John; Chimgoneko, Lorensio; Toto, Neema; Dula, Dingase; Morton, Ben; Pennington, Shaun H.; Hyder-Wright, Angela; Collins, Andrea M.; Mitsi, Elena; Ferreira, Daniela M.; Gordon, Stephen B.; Henrion, Marc Y.R.; Alderson, Mark; Antoine, Sandra; Bailey, Christopher R.; Bogaert, Debbie; Brown, Jeremy; Burr, Sarah; De Jonge, Marien L.; Doherty, Klara; Goldblatt, David; Gondwe, Joel; Gooding, Kate; Grigg, Jonathan; Harawa, Tina; Heyderman, Rob; Hinds, Jason; Jochems, Simon; Kapumba, Blessings; Kaudzu, Vella; Kneller, Robert; Malley, Richard; Mallewa, Jane; Manda-Taylor, Lucinda; Mangani, Edward; Mayuni, Mphatso; Mwandumba, Henry; Ngwira, Brian; Mwenechanya, Percy; Parker, Mike; Pollard, Andrew; Reuben, Modesta; Gomes, Gabriela; Weiser, Jeffrey; Jambo, Kondwani; Kenny-Nyazika, Tinashe; Mkutumula, Maureen; Chipwete, Maurice; Bailey, Christopher; Beall, Bernard; Bentley, Stephen; Bogaert, Debby; De Campo, Joe; Clarke, Stuart; Cleary, David; Finn, Adam; Gessner, Brad; Gordon, Stephen; Hales, Caz; Hamon, Melanie; Heyderman, Robert; Kiyono, Hiroshi; Leong, Samuel; Lipsitch, Marc; Mann, Alex; Mina, Michael; Miyaji, Eliane; Nakaya, Helder; Napolitani, Giorgio; Neill, Daniel; Netea, Mihai; Oggioni, Marco; Openshaw, Peter; Peterson, Fernanda (Nature Research, 2026-05-14)
We conducted the first pooled analysis of two randomised controlled vaccine trials on pneumococcal serotype 6B carriage, registered in Malawi (PACTR202008503507113) and the UK (ISRCTN45340436). This post-hoc exploratory study examined the sex-based differences in carriage, vaccine efficacy and vaccine-induced responses. PCV-13 reduced colonisation by 76% (p < 0.001) with non-significant interaction by sex (RR = 1.549, p = 0.413). Females showed a higher carriage rate than males (28% vs. 19%, p = 0.066). Baseline anti-6B Capsular Polysaccharide Immunoglobulin G (IgG) titres were higher in females, significantly in Malawi (2.62 µg/ml vs males 2.05 µg/ml, p = 0.015). Post-vaccination titres did not differ by sex. The pooled fold change in IgG pre-post vaccination, was higher in vaccinated females (5.47 vs 3.30, p = 0.053). This analysis demonstrates the utility and challenges of integrating CHIM data between diverse settings to evaluate vaccine efficacy, describe inter-setting differences, investigate biological and immunological factors influencing protection against pneumococcal carriage and ultimately inform future vaccine development strategies.
Navigating multicollinearity in linear regression models : implications for big data analysis
Du Plessis, Salomi; Arashi, Mohammad; Millard, Salomon; Maribe, Gaonyalelwe (Universal Wiser Publisher, 2026-05)
The consequences of multicollinearity in regression analysis involving small, moderate, or high-dimensional datasets are well-established, and many notable solutions exist. However, the consequences of multicollinearity when considering big data, specifically data with a large number of observations, are not well established. In this paper, we determine the impact of multicollinearity on the linear regression model when applied to big data by numerically evaluating the bias, variance, and signs of the estimated regression coefficients. An extensive simulation study shows that multicollinearity does not substantially alter the statistical measures under consideration. Our analysis is also applied to a real-world dataset for method demonstration.
Explainable AI-driven comparative analysis of machine learning models for predicting HIV viral nonsuppression in Ugandan patients : retrospective cross-sectional study
Ngema, Francis; Whata, Albert; Olusanya, Micheal O.; Mhlongo, Siyabonga (JMIR Publications, 2026-01)
BACKGROUND : HIV viral suppression is essential for improving health outcomes and reducing transmission rates among people living with HIV. In Uganda, where HIV/AIDS is a major public health concern, machine learning (ML) models can predict viral suppression effectively. However, the limited use of explainable artificial intelligence (XAI) methods affects model transparency and clinical utility. OBJECTIVE : This study aimed to develop and compare ML models for predicting viral nonsuppression in Ugandan people living with HIV on antiretroviral therapy (ART), and then systematically apply comprehensive XAI techniques to the best-performing model to identify key predictors and demonstrate interpretability at both population and individual patient levels. METHODS : We retrospectively analyzed clinical and demographic data from 1101 Ugandan people living with HIV on ART at the HIV clinic in Muyembe Health Centre IV between June 2016 and April 2018, focusing on predicting viral nonsuppression (viral load >1000 copies per milliliter). The dataset was divided into model-building (training: 80%) and validation (test: 20%) sets. To address class imbalance, the synthetic minority over-sampling technique was applied. For global explanation, 8 ML algorithms-logistic regression, stacked ensemble, random forest, support vector machines, extreme gradient boosting (XGBoost), k-nearest neighbors, naïve Bayes, and artificial neural networks-were compared. Model performance was evaluated using metrics such as accuracy, precision, recall, F1-score, Cohen κ, and area under the curve (AUC). For local explanation, individual conditional expectation plots, Shapley Additive Explanations (SHAP), breakdown, and SHAP force plots were used to provide insights into predictions for individual patients. RESULTS : The XGBoost ensemble model demonstrated superior performance with an accuracy of 0.89, precision of 0.59, recall of 0.65, and AUC of 0.80. The model achieved high specificity (0.93) and moderate sensitivity, yielding a Cohen κ of 0.55 and F1-score of 0.62, indicating good discriminative ability for viral nonsuppression prediction. SHAP feature importance analysis identified adherence assessment over the preceding 3 months as the most influential predictor of viral nonsuppression, followed by age group, urban residence, and duration on ART. Local SHAP consistently demonstrated that poor adherence was the primary driver of both correctly identified nonsuppressed cases and false positive predictions, reinforcing adherence as the critical determinant of treatment outcomes. CONCLUSIONS : The XGBoost model demonstrated optimal performance for predicting viral nonsuppression among Ugandan people living with HIV on ART, achieving an AUC of 0.80. Comprehensive XAI analysis identified adherence assessment as the primary predictor, followed by age group, residence type, and ART duration. XAI methods provided transparent interpretation of model predictions at both population and individual patient levels, enabling identification of key risk factors for targeted clinical interventions in resource-limited settings.
Pseudo partial likelihood method for proportional hazards models when time origin is missing for control group with applications to SARS-CoV-2 seroprevalence study
Chung, Yunro; Murugan, Vel; Beyene, Kassu Mehari; Chen, Ding-Geng (Din) (Renmin University of China, 2025-10-07)
Time-to-event data analysis without a well-defined time origin commonly occurs in observational studies that retrospectively collect survival endpoints. For instance, after enrolling participants who have or have not received a specific treatment, an event status can be observed for all participants; however, the start date of treatment is only observable for the treatment group. The corresponding time origin does not exist for the control group, resulting in missing survival time data. Complete-case analysis is often considered the standard approach, but it disregards information from all participants in the control group and does not allow us to compare their survival distributions. To address this challenge, we propose a novel semiparametric proportional hazards model by regarding these missing time origins as nuisance parameters. We approximate the risk sets as cumulative normal distributions to deal with these nuisance parameters and develop estimation and inference procedures for our proposed estimator. We study the asymptotic properties of this model and conduct the simulation studies to validate its finite sample property. Analysis of data from a recent SARS-CoV-2 seroprevaluence study illustrates the applicability of our methods. The proposed methods are implemented in the R package coxphm.
Robust median regression for count data with general lower truncation using a contaminated discrete Weibull model
Burger, Divan A.; Van Niekerk, Janet; Lesaffre, Emmanuel (Walter de Gruyter, 2026-06)
In right-skewed count data, the mean is disproportionately affected by a long upper tail, whereas the median remains a more representative measure of central tendency. Discrete Weibull (DW) regression links covariates to a shifted median, which in turn induces the exact integer median; however, a single DW component can fit poorly when the observed count distribution has a markedly heavier upper tail than a single-component model can accommodate. We propose a contaminated DW (cDW) regression that augments the baseline DW distribution with a more dispersed secondary component within a finite mixture while retaining a single shifted-median link. This mixture accommodates extreme counts more effectively, thereby stabilizing the median-based regression coefficients. The model accommodates general lower truncation at an arbitrary threshold c, including c = 1 for strictly positive outcomes and c = 0 for nonnegative counts, and is estimated using a straightforward Bayesian Markov chain Monte Carlo algorithm implemented in JAGS; R code accompanies the paper. Applied to hospital length-of-stay data, the cDW regression reduces the influence of outliers and achieves superior predictive performance relative to a single-component DW model, as demonstrated by leave-one-out cross-validation and a Kullback-Leibler influence diagnostic. Simulation experiments show that, under strongly heavy-tailed mixture settings, the cTDW model accurately recovers the regression coefficients and improves on the single-component TDW model. Because the added tail component can increase probability mass at both extremes, we further recommend embedding the cDW in a hurdle framework when structural zeros are present: the zero probability is modeled separately, and the heavy-tail mixture is applied only to positive counts. The cDW regression model provides a robust, median-centered alternative for analyzing skewed, possibly truncated count outcomes.
Evaluating the safe application of the Bridging Infix method for anterior pelvic fixation in relation to neighbouring anatomical structures
Van Schalkwyk, Jerolize; Keough, Natalie; Naicker, Jade; Strydom, Sven; Snyckers, Christian H.; Masenge, Andries; Mogale, Nkhensani (Elsevier, 2026-04)
BACKGROUND : Numerous techniques for anterior pelvic fixation exist, however, a novel minimally invasive subcutaneous technique, the Bridging Infix, has been proposed. Although, little is known regarding the risk of injury to anatomical structures in close proximity to the modified implant site. The current study aimed to assess the implant’s proximity to adjacent anatomical structures and palpable bony landmarks to ensure secure and safe fixation. METHODS : An anatomical study was conducted using 50 formalin-fixed cadavers and two fresh-frozen specimens. Dissections were performed to identify and measure the distances between the Bridging Infix implant and critical anatomical structures, including the iliohypogastric nerve, ilioinguinal nerve, superficial epigastric vessels, superficial circumflex iliac vessels, spermatic cord, and the femoral neurovascular bundle. Measurements were also taken between these anatomical structures and relevant bony landmarks. RESULTS : When the Bridging Infix was applied following accepted technical guidelines, no surrounding anatomical structures were compromised. The ilioinguinal and iliohypogastric nerves were shielded by the abdominal muscles and located, on average, 48.96 (SD ± 20.65, range: 10.21–85.26) mm and 31.55 (SD ± 14.45, range:7.42–72.89) mm from the anterior superior iliac spine (ASIS), respectively. The spermatic cord showed no signs of compression by the implant. The superficial epigastric vessels occasionally passed through the medial window field area, while the superficial circumflex iliac vessels were found near the ASIS and lateral window. Nevertheless, vessels can be easily cauterized or ligated when necessary, underscoring the importance of surgeons being familiar with these anatomical relations. In this case, the femoral vessels were positioned deep to the implant, with no gross evidence of direct contact or compression. CONCLUSION : Anatomical structures initially considered to be at risk during the placement of the Bridging Infix were observed to lie at an adequate distance from the fixation point, without macroscopic evidence of implant-related compression. Further studies are required to clinically validate these findings. HIGHLIGHTS • No anatomical structures were injured during the Bridging Infix proving it is a minimally invasive technique. • Requires surgical expertise, layer-by-layer dissection, full musculature retraction to avoid IHN and IIN injury. • Compression of the spermatic cord and femoral vessels was not observed. • Further quantitative analysis should be conducted to confirm findings and assess potential risks under varying conditions.
Navigating adversity : non-completion and evacuation of trail runners at the 2022 Magoebaskloof Forest Marathon
Maleka, Thabo; Jansen van Rensburg, Audrey; Viljoen, Carel Thomas; Ramkilawon, Gopika Devi; Ramagole, Dimakatso Althea; Nair, Anoop; Janse van Rensburg, Dina Christina (South African Sports Medicine Association, 2026-04-22)
BACKGROUND : Trail running occurs on natural, uneven terrain. The literature is limited in exploring factors that contribute to trail runners not completing a race. OBJECTIVES : This study firstly investigated self-reported reasons for runners who did not finish (DNF) the 2022 Magoebaskloof Forest Marathon across 10km, 20km, 34km, and 50km distances, and secondly assessed runners' evacuation methods. METHODS: A descriptive cross-sectional study of consenting trail runners (n=429) who completed the post-race DNF survey (10km, 20km, 34km and 50km distances). Within two weeks post-race, runners reported whether they finished, reasons for discontinuing and evacuation methods. We report numbers (n) and point prevalence (%). RESULTS : Of the 228 runners who completed the post-race survey, 16 (7%) DNF. DNF runners entered the 34 km (19%) and 50 km distance (81%) and were aged 31–50 years (94%). No DNFs occurred in the 10km and 20km races. Approximately 27% did not meet cut-off times, while 19% got lost despite trail directions. Other contributing factors included pre-existing and new injuries (23%), illnesses (15%), and inadequate training (8%). While 50% of the DNF group had >5 years total running experience, only 19% had >5 years of trail running experience. Runners who finished the trail race had significantly more trail running experience than those who DNF (p=0.0082). The primary evacuation method was hiking back to the nearest aid station (63%), with 19% subsequently transported by shuttle. CONCLUSION : Multiple factors may have contributed to the race non-completion. Many DNF runners started with injuries or illnesses, inadequate training, or limited trail-running experience. Event planners should improve measures to prevent runners from getting lost. These findings emphasise the importance of preparation, navigation skills, and race management in ensuring trail running safety and success.
Effects of acupuncture on stride, speed, and heart rate variability in thoroughbred racehorses
Hartwigsen, Roselle; Grant, C.C. (Catharina Cornelia); Botha, Tanita; Van Vollenhoven, Elize (Medpharm Publications, 2026-03)
Thoroughbred racehorses are at increased risk of injury during race preparations and trials, primarily due to intensive workloads that predispose them to musculoskeletal injuries. Evaluating clinical interventions to improve health and prevent injury is essential for addressing animal welfare concerns. This study investigated the medium-term effects of acupuncture on stride, speed and cardiac autonomic regulation (assessed via heart rate variability [HRV]), in Thoroughbred flat racehorses. This prospective, randomised, parallel-group study included an intervention and a control group. Thirty Thoroughbred racehorses in full training (racing age 3 years; range 3.0–3.9), from a single professional training yard, and at the same stage of their racing careers, were maintained under similar nutritional, training, and environmental conditions. The intervention group received three acupuncture treatments, over a three-week period. HRV and 800-m gallop stride parameters (SP) were measured pre- and post-acupuncture period in both groups. Pre-post changes or delta (Δ) in each indicator were calculated for both groups and compared using non-parametric Mann-Whitney U tests. Treatment was associated with non-significant reductions in parasympathetic HRV indicators namely root mean square of successive differences in beat-to-beat RR-interval (ΔRMSSD: treatment vs control: - 2.19 vs. 1.47 ms), high frequency components (ΔHF: - 37.19 vs. 19.42 ms2), and standard deviation of short-term variability (ΔSD1: - 1.56 vs. 1.04 ms), suggesting lower vagal tone. Maximum stride count (Max SC) increased significantly in the control group (p = 0.004). Overall, the treatment group showed decreased stride count and vagal activity indicating that a short recovery period may be required after acupuncture.
Robust PCA with machine learning for retinal image enhancement and diabetes classification
Likassa, Habte Tadesse; Chen, Ding-Geng (Din) (Springer Nature, 2026-06-23)
Non-mydriatic retinal images, akin to neuroimaging, often suffer from artifacts and blurring that significantly inhibit the accurate diagnosis and early detection of neurodegenerative diseases. Current image enhancement methods, such as Robust Principal Component Analysis (RPCA), are commonly employed for early disease detection and medical diagnosis. However, these methods typically assume uniform singular value weights with existing nuclear norms, an assumption that may not hold because of the inherent variations in noisy image data. In addition, RPCA approaches primarily focus on global enhancement, failing to capture the fine details critical for precise analysis of retinal images. This limitation underscores the need for advanced and more adaptive method that can address these challenges and improve the accuracy of diagnostic tools in the biosciences, particularly for the early detection of diseases. To address these drawbacks, we propose a novel method that combines RPCA, the truncated weighted nuclear norm (TWNN), and the equalization of adaptive histograms (AHE). Unlike existing methods, the proposed approach enhances degraded retinal images by further applying Gaussian filtering to correct the edges and AHE used to detail the retinal images, preserve key structures, and improve contrast. The method is formulated as an optimization technique, incorporating Histogram Oriented Gradients (HOG) features, with the enhanced images used for diabetes prediction via machine learning, and the parameters are updated iteratively using ADMM. We conducted ablation studies to select the best model that predicts diabetes from the machine learning algorithms, including Support Vector Machine, Logistic Regression, and K-Nearest Neighbors, on retinal fundus images across all severity levels using the proposed method. Among these, SVM demonstrated the highest performance in predicting diabetes, from which we considered it for the classification purpose in this study. .One of the interesting aspect of this study is that, unlike the existing methods, we use the enhanced images produced by our approach for diabetes classification.The results of the study show that the proposed method substantially enhanced the quality of the images and improved prediction accuracy based on the public databases.
Socioeconomic and regional predictors of antenatal care contact frequency among ever-married somali women with recent pregnancies
Ali, Mahad Ibrahim; Belay, Denekew Bitew; Mohamed, Jama; Jama, Umalkhayr Abdi; Dessie, Zelalem G.; Chen, Ding-Geng (Din) (Springer Nature, 2026-07-01)
BACKGROUND : Antenatal care (ANC) is critical for preventing complications and reducing maternal and child mortality. Apart from essential services, such as hypertension treatment and HIV testing, ANC provides women with preventive healthcare, including tetanus vaccinations, advice on healthy pregnancy practices, warning signs, and social and emotional support. This study evaluated the factors associated with the number of ANC contacts among ever-married women with recent pregnancies in Somalia. METHODS : This cross-sectional study used data from the 2020 Somalia Health and Demographic Survey. This study included 8,909 women aged 15–49 years who met the inclusion criteria. Count regression models were used to examine factors associated with the number of ANC contacts among ever-married women with recent pregnancies in Somalia, and Incidence Rate Ratio (IRR) with 95% CI were reported. RESULTS : In this study, 67.96% of women had no contact with ANC during their last pregnancy. Zero-inflated Poisson (ZIP) regression model was obtained to be the best fit. According to this model, women with secondary education (IRR = 1.21, 95% CI 1.03–1.42) and higher education (IRR = 1.27, 95% CI 1.08–1.48) had more ANC contacts than women with no formal education. Being in the highest wealth quintile (IRR = 1.27, 95% CI 1.09–1.49) was also associated with increased ANC contacts. Living in certain regions was associated with the number of ANC contacts. In the zero-inflated section, maternal age, education, region, media exposure, and wealth index emerged as key predictors of ANC utilization in Somalia. CONCLUSIONS : Many ever-married women with recent pregnancies did not receive any ANC, and socio-demographic factors such as educational level, maternal age, wealth index, place of residence, media exposure, and region were significantly associated with the number of ANC contacts in Somalia. Implementing initiatives based on these findings and developing public health strategies on promoting a culture of increased ANC utilization could improve maternal and child health outcomes.
Bayesian analysis of the prevalence and determinants of viral load suppression in South Africa
Yilema, Seyifemickael Amare; Belay, Alebachew Taye; Rad, Najmeh Nakhaei; Chen, Ding-Geng (Din) (Springer Nature, 2026-06-07)
Lowering the rate of onset, progression, mortality, and morbidity remains a key goal in the management of Human Immunodeficiency Virus (HIV). Antiretroviral therapy (ART) is working effectively to enhance prognosis and quality of life among individuals living with HIV. Viral load suppression is a viral load of less than 1000 copies/ml in an individual receiving antiretroviral therapy (ART), indicating that ART is working effectively to stop the virus from multiplying. Despite significant progress in HIV treatment, achieving and sustaining viral load suppression remains an ongoing public health concern. Therefore, ART leads to a high chance of suppressing HIV viral load and undetectable levels of the virus to improve individual health and well-being from the disease. However, the prevalence of HIV remains high despite notable advances in increasing access to ART. The results of this study revealed that approximately 64.8% of adults with HIV patients had viral suppression in South Africa. Individuals aged between 25 and 34 years (AOR = 1.44, 95% CrI: 1.007, 2.082), aged between 35 and 44 years (AOR = 1.55, 95% CrI: 1.085, 2.218), older than 54 years (AOR = 2.30, 95% CrI: 1.445, 3.699), and of the female sex (AOR = 1.68, 95% CrI: 1.311, 2.156) were significantly associated with higher odds of viral load suppression. However, non-African races (AOR = 0.344, 95% CrI: 0.238, 0.500), missed ART treatment (AOR = 0.478, 95% CrI: 0.369, 0.621), and married individuals (AOR = 0.690, 95% CrI: 0.515, 0.920) had negative significant effect on viral load suppression. The study is far from WHO and UNAIDS’s ambitious 95-95-95 plan to reach viral load suppression to 95% by 2025. Based on the findings of this study, the concerned body should strengthen the regular monitoring of peripheral health facilities and enhance efforts to achieve and maintain a suppressed HIV viral load.
Penalizing complexity priors for Bayesian inference of circular models
Ye, Xiang; Van Niekerk, Janet; Rue, Havard (Sage, 2026)
Advancements in computational power and methodologies have enabled research on massive datasets. However, tools for analyzing data with directional or periodic characteristics, such as wind directions and customers’ arrival time in 24-hour clock, remain underdeveloped. While statisticians have proposed circular distributions for such analyses, significant challenges persist in constructing circular statistical models, particularly in the context of Bayesian methods. These challenges stem from limited theoretical development and a lack of historical studies on prior selection for circular distribution parameters. In this article, we propose a framework for selecting hyperpriors that contracts to a simpler model in circular scenarios, especially when there is insufficient information to guide prior selection. We introduce well-examined Penalized Complexity (PC) priors for the most widely used circular distributions. Comprehensive comparisons with existing hyperpriors in the literature are conducted through simulation studies and a practical case study. Final, we discuss the contributions and implications of our work, providing a foundation for further advancements in constructing Bayesian circular statistical models.
Hyperbaric oxygen therapy versus intravenous thrombolysis in the treatment of central retinal artery occlusion : a systematic review and meta-analysis
Bakdalieh, Anas; Siddiqui, Dawood; Chen, Ding-Geng (Din); Yu, Minzhong (MDPI, 2026-04)
BACKGROUND : Central retinal artery occlusion (CRAO) causes sudden, often profound monocular vision loss. We compared the efficacy and safety of hyperbaric oxygen therapy (HBOT) versus intravenous thrombolysis (IVT) using a systematic review and meta-analysis. METHODS : Following PRISMA 2020 guidance, we searched PubMed, Cochrane Library, Web of Science, and VHL from inception to 27 May 2025, screened records in duplicate, and synthesized visual outcomes (best-corrected visual acuity [BCVA], logMAR), onset-to-treatment time, and adverse events using random-effects models. RESULTS : Twenty-five observational studies (781 patients; 557 HBOT eyes, 225 IVT eyes) met inclusion. Both HBOT and IVT were associated with significant improvements in BCVA (HBOT: MD −0.57 logMAR; IVT: MD −0.53 logMAR). Clinically meaningful improvement occurred in 45.8% after HBOT and 42.0% after IVT. Adverse event rates were similar (HBOT 11.3%; IVT 10.2%) but differed in type (ear barotrauma with HBOT; hemorrhagic events with IVT). CONCLUSIONS : HBOT and IVT both improve visual outcomes in CRAO. Differences in adverse event profiles and substantial heterogeneity among IVT studies underscore the need for adequately powered comparative trials and standardized treatment pathways. Registration: Not prospectively registered; PRISMA 2020 checklist is provided.
Virtual escape rooms as a game-based learning strategy through the lens of the ARCS model
Snyman, Ciska; Van Wyk, Mari; Moodley, Kimera; Botha, Tanita (University of Limpopo, 2025)
Generation Z students, who have grown up immersed in digital technologies, bring distinct expectations to their learning experiences. Yet, their motivation in online and hybrid learning environments remains a significant challenge. The study examines virtual escape rooms as a game-based learning strategy to enhance student motivation, aligned with a mixed-methods approach, combining quantitative surveys and qualitative interviews to measure motivation in a Microsoft Excel virtual escape room. Drawing on 461 voluntary survey responses and five in-depth interviews, the authors designed and facilitated the intervention within a compulsory university first-year digital literacy module. Results indicated that the immersive, gamified design of escape rooms boosted student motivation substantially, promoted collaborative learning, and enhanced critical thinking. Students reported greater motivation through attention-grabbing game elements, real-world applications, confidence-building challenges, and satisfaction from task completion. The study concludes that virtual escape rooms can serve as an effective, innovative strategy to promote active learning and student motivation. While international literature often generalises Generation Z as digitally fluent, this study acknowledges that South African students may have varied technological experiences due to differing access and educational backgrounds. Educators in Southern Africa could implement a similar digital strategy, like virtual escape rooms, to improve student motivation.

Browse

Recent Submissions