RESUMEN

Purpose: To report the visual and refractive characteristics and the prevalence of amblyopia in patients with different types of Duane's Retraction Syndrome (DRS). Method: This retrospective study was performed on hospital records of 582 DRS patients at Farabi Hospital, Iran, from 2012 to March 2022. Results: The mean age of patients was 19.4 ± 11.9 (range, 3–70) years [335 (57.6 %) females and 247 (42.4 %) males (P < .001)]. DRS type I, II, III, and IV were presented in 347 (59.6 %), 148 (25.4 %), 82 (14.1 %), and 5 (0.9 %) patients, respectively. There were 530 (91.1 %) patients with unilateral and 52 (8.9 %) with bilateral involvement. In the unilateral patients, the DRS eyes' corrected distance visual acuity (CDVA) and astigmatism were significantly worse than the Non-DRS Eyes (P < .001). The mean amount of all refractive and visual parameters in bilateral patients' right or left eyes was significantly lower than in unilateral patients' non-DRS eyes (all P < .05). Anisometropia was observed in 75(12.9 %) of the patients. Amblyopia was observed in 18.5 % (98 patients) and 36.5 % (19 patients) of unilateral and bilateral DRS patients, respectively (P < .001). In unilateral patients, amblyopia was found in 57 (16.4 %) patients with Type I, 22 (14.9 %) patients with Type II, 16 (19.5 %) patients with Type III, and 3 (60 %) patients with Type IV. Forty-four (37.6 %) of patients with amblyopia had anisometropia. Conclusion: This large-scale study indicates that DRS types differ in terms of refractive error, visual acuity, and the prevalence of amblyopia and anisometropia. Clinicians should be aware of the clinical features associated with different types of DRS.(AU)

Asunto(s)

RESUMEN

Purpose: In children under 20 years, refractive development targets a cycloplegic refractive error of +0.5 to +1.5D, while presbyopes over 40 years generally have non-cycloplegic errors of ≥ +1D. Some papers suggest these periods are separated by a period of myopic refractive error (i.e., ≤ –0.50D), but this remains unclear. Hence, this work investigates the mean cycloplegic refractive error in adults aged between 20 – 40 years. Methods: In 2002 a cross-sectional study with stratified cluster sampling was performed on the population of Tehran, providing cycloplegic and non-cycloplegic refractive error data for the right eyes of 3,576 participants, aged 30.6 ± 18.6 years (range: 1–86 years). After grouping these data into age groups of 5 years, the refractive error histogram of each group was fitted to a Bigaussian function. The mean of the central, emmetropized peak was used to estimate the mean refractive error without the influence of myopia. Results: The mean cycloplegic refractive error at the emmetropized peak decreased from +1.10 ± 0.11D (95 % confidence interval) to +0.50 ± 0.04D before 20 years and remains stable at that value until the age of 50 years. The non-cycloplegic refractive error also sees a stable phase at 0.00 ± 0.04D between 15 – 45 years. After 45 – 50 years both cycloplegic and non-cycloplegic refractive error become more hypermetropic over time, +1.14 ± 0.12D at 75 years. Conclusions: The cycloplegic refractive error in adults is about +0.50D between 20 – 50 years, disproving the existence of the myopic period at those ages.(AU)

Asunto(s)

RESUMEN

The present study aimed to summarize and report data on errors related to treatment planning, which were collected by medical physicists. The following analyses were performed based on the 10-year error report data: (1) listing of high-risk errors that occurred and (2) the relationship between the number of treatments and error rates, (3) usefulness of the Automated Plan Checking System (APCS) with the Eclipse Scripting Application Programming Interface and (4) the relationship between human factors and error rates. Differences in error rates were observed before and after the use of APCS. APCS reduced the error rate by ~1% for high-risk errors and 3% for low-risk errors. The number of treatments was negatively correlated with error rates. Therefore, we examined the relationship between the workload of medical physicists and error occurrence and revealed that a very large workload may contribute to overlooking errors. Meanwhile, an increase in the number of medical physicists may lead to the detection of more errors. The number of errors was correlated with the number of physicians with less clinical experience; the error rates were higher when there were more physicians with less experience. This is likely due to the lack of training among clinically inexperienced physicians. An environment to provide adequate training is important, as inexperience in clinical practice can easily and directly lead to the occurrence of errors. In any environment, the need for additional plan checkers is an essential factor for eliminating errors.

RESUMEN

In this paper, a new method is designed to effectively determine the parameters of proton exchange membrane fuel cells (PEMFCs), i.e., ξ 1 , ξ 2 , ξ 3 , ξ 4 , R C , λ , and b . The fuel cells (FCs) involve multiple variable quantities with complex non-linear behaviours, demanding accurate modelling to ensure optimal operation. An accurate model of these FCs is essential to evaluate their performance accurately. Furthermore, the design of the FCs significantly impacts simulation studies, which are crucial for various technological applications. This study proposed an improved parameter estimation procedure for PEMFCs by using the GOOSE algorithm, which was inspired by the adaptive behaviours found in geese during their relaxing and foraging times. The orthogonal learning mechanism improves the performance of the original GOOSE algorithm. This FC model uses the root mean squared error as the objective function for optimizing the unknown parameters. In order to validate the proposed algorithm, a number of experiments using various datasets were conducted and compared the outcomes with different state-of-the-art algorithms. The outcomes indicate that the proposed GOOSE algorithm not only produced promising results but also exhibited superior performance in comparison to other similar algorithms. This approach demonstrates the ability of the GOOSE algorithm to simulate complex systems and enhances the robustness and adaptability of the simulation tool by integrating essential behaviours into the computational framework. The proposed strategy facilitates the development of more accurate and effective advancements in the utilization of FCs.

RESUMEN

PURPOSE: In the context of clinical research, there is an increasing need for new study designs that help to incorporate already available data. With the help of historical controls, the existing information can be utilized to support the new study design, but of course, inclusion also carries the risk of bias in the study results. METHODS: To combine historical and randomized controls we investigate the Fill-it-up-design, which in the first step checks the comparability of the historical and randomized controls performing an equivalence pre-test. If equivalence is confirmed, the historical control data will be included in the new RCT. If equivalence cannot be confirmed, the historical controls will not be considered at all and the randomization of the original study will be extended. We are investigating the performance of this study design in terms of type I error rate and power. RESULTS: We demonstrate how many patients need to be recruited in each of the two steps in the Fill-it-up-design and show that the family wise error rate of the design is kept at 5 % . The maximum sample size of the Fill-it-up-design is larger than that of the single-stage design without historical controls and increases as the heterogeneity between the historical controls and the concurrent controls increases. CONCLUSION: The two-stage Fill-it-up-design represents a frequentist method for including historical control data for various study designs. As the maximum sample size of the design is larger, a robust prior belief is essential for its use. The design should therefore be seen as a way out in exceptional situations where a hybrid design is considered necessary.

Asunto(s)

RESUMEN

Regulatory genetic toxicology testing is essential for identifying potentially mutagenic hazards. Duplex Sequencing (DS) is an error-corrected next-generation sequencing technology that provides substantial advantages for mutation analysis over conventional mutagenicity assays including: improved accuracy of mutation detection, ability to measure changes in mutation spectrum, and applicability across diverse biological models. To apply DS for regulatory toxicology testing, power analyses are required to determine suitable sample sizes and study designs. In this study, we explored study designs to achieve sufficient power for various effect sizes in chemical mutagenicity assessment. We collected data from MutaMouse bone marrow and liver samples that were analyzed by DS using TwinStrand's Mouse Mutagenesis Panel. Average duplex reads achieved in two separates studies on liver and bone marrow were 8.4 × 108 (± 7.4 × 107) and 9.5 × 108 (± 1.0 × 108), respectively. Baseline mean mutation frequencies (MF) were 4.6 × 10-8 (± 6.7 × 10-9) and 4.6 × 10-8 (± 1.1 × 10-8), with estimated standard deviations for the animal-to-animal random effect of 0.15 and 0.20, for liver and bone marrow, respectively. We conducted simulation analyses based on these empirically derived parameters. We found that a sample size of four animals per group is sufficient to obtain over 80% power to detect a two-fold change in MF relative to baseline. In addition, we estimated the minimal total number of informative duplex bases sequenced with different sample sizes required to retain power for various effect sizes. Our work provides foundational data for establishing suitable study designs for mutagenicity testing using DS.

RESUMEN

Background: Atrial fibrillation (AFib) detection via mobile ECG devices is promising, but algorithms often struggle to generalize across diverse datasets and platforms, limiting their real-world applicability. Objective: This study aims to develop a robust, generalizable AFib detection approach for mobile ECG devices using crowdsourced algorithms. Methods: We developed a voting algorithm using random forest, integrating six open-source AFib detection algorithms from the PhysioNet Challenge. The algorithm was trained on an AliveCor dataset and tested on two disjoint AliveCor datasets and one Apple Watch dataset. Results: The voting algorithm outperformed the base algorithms across all metrics: the average of sensitivity (0.884), specificity (0.988), PPV (0.917), NPV (0.985), and F1-score (0.943) on all datasets. It also demonstrated the least variability among datasets, signifying its highest robustness and effectiveness in diverse data environments. Moreover, it surpassed Apple's algorithm on all metrics and showed higher specificity but lower sensitivity than AliveCor's Kardia algorithm. Conclusions: This study demonstrates the potential of crowdsourced, multi-algorithmic strategies in enhancing AFib detection. Our approach shows robust cross-platform performance, addressing key generalization challenges in AI-enabled cardiac monitoring and underlining the potential for collaborative algorithms in wearable monitoring devices.

Asunto(s)

RESUMEN

When analyzing GPS time series, common mode errors (CME) often obscure the actual crustal movement signals, leading to deviations in the velocity estimates of station coordinates. Therefore, mitigating the impact of CME on station positioning accuracy is crucial to ensuring the precision and reliability of GNSS time series. The current approach to separating CME mainly uses signal filtering methods to decompose the residuals of the observation network into multiple signals, from which the signals corresponding to CME are identified and separated. However, this method overlooks the spatial correlation of the stations. In this paper, we improved the Independent Component Analysis (ICA) method by introducing correlation coefficients as weighting factors, allowing for more accurate emphasis or attenuation of the contributions of the GNSS network's spatial distribution during the ICA process. The results show that the improved Weighted Independent Component Analysis (WICA) method can reduce the root mean square (RMS) of the coordinate time series by an average of 27.96%, 15.23%, and 28.33% in the E, N, and U components, respectively. Compared to the ICA method, considering the spatial distribution correlation of stations, the improved WICA method shows enhancements of 12.53%, 3.70%, and 8.97% in the E, N, and U directions, respectively. This demonstrates the effectiveness of the WICA method in separating CMEs and provides a new algorithmic approach for CME separation methods.

RESUMEN

While interest in using wearable sensors to measure infant leg movement is increasing, attention should be paid to the characteristics of the sensors. Specifically, offset error in the measurement of gravitational acceleration (g) is common among commercially available sensors. In this brief report, we demonstrate how we measured the offset and other errors in three different off-the-shelf wearable sensors available to professionals and how they affected a threshold-based movement detection algorithm for the quantification of infant leg movement. We describe how to calibrate and correct for these offsets and how conducting this improves the reproducibility of results across sensors.

Asunto(s)

RESUMEN

INTRODUCTION: Amblyopia is a common cause of preventable visual impairment in children, affecting 1% to 6% globally. This study assesses amblyopia prevalence and risk factors among school children in rural Telangana, India, using the Spot Vision Screener (Welch Allyn, Inc., Skaneateles Falls, New York, USA), a portable, noninvasive device recommended for automated vision screening. METHODS: A cross-sectional study was conducted on 714 schoolchildren aged 5-10 years. Screening was performed using the Spot Vision Screener, evaluating refractive errors, ocular alignment, and other amblyopia risk factors. Children identified with potential amblyogenic factors were referred for comprehensive ophthalmological evaluation to confirm diagnosis. RESULTS: Out of 714 children screened, 84 were referred by the Spot Vision Screener for further evaluation. Subsequent examination by ophthalmologists confirmed amblyopia in 65 children, resulting in a prevalence of 9.10%. Myopic refractive error was the most prevalent (69.23%), followed by astigmatism (21.53%) and hypermetropia (9.23%) among amblyopia cases. CONCLUSION: The Spot Vision Screener proved to be a reliable tool for identifying amblyopia risk factors in school children, facilitating early detection and referral for appropriate management. This study underscores the importance of implementing effective vision screening programs in rural settings to mitigate preventable childhood blindness.

RESUMEN

The topic of diagnostic imaging error and the tools and strategies for error mitigation are poorly investigated in veterinary medicine. The increasing popularity of diagnostic imaging and the high demand for teleradiology make mitigating diagnostic imaging errors paramount in high-quality services. The different sources of error have been thoroughly investigated in human medicine, and the use of AI-based products is advocated as one of the most promising strategies for error mitigation. At present, AI is still an emerging technology in veterinary medicine and, as such, is raising increasing interest among in board-certified radiologists and general practitioners alike. In this perspective article, the role of AI in mitigating different types of errors, as classified in the human literature, is presented and discussed. Furthermore, some of the weaknesses specific to the veterinary world, such as the absence of a regulatory agency for admitting medical devices to the market, are also discussed.

RESUMEN

Cognitive impairment has been widely accepted as a disease progression measure prior to the onset of Huntington's disease. We propose a sophisticated measurement error correction method that can handle potentially correlated measurement errors in longitudinally collected exposures and multiple outcomes. The asymptotic theory for the proposed method is developed. A simulation study is conducted to demonstrate the satisfactory performance of the proposed two-stage fitting method and shows that the independent working correlation structure outperforms other alternatives. We conduct a comprehensive longitudinal analysis to assess how brain striatal atrophy affects impairment in various cognitive domains for Huntington's disease.

RESUMEN

Purpose: This study aimed to investigate the prevalence of refractive error (RE) and risk factors for myopia among older adults in the Han and various minority ethnic groups across seven provinces in China. Methods: This cross-sectional study forms a part of the ophthalmic dataset of the China National Health Survey (CNHS). Face-to-face interviews and ophthalmic examinations were conducted in seven provinces located in western and northern China. The age- and sex-adjusted prevalence of RE among Han and seven other ethnic groups aged 50-80 years were compared. A mixed-effects model was used to identify the risk factors associated with RE. Results: A total of 12,902 participants, including 8800 Han and 4102 from ethnic minorities, were included in the study. The age- and sex-adjusted prevalence of myopia, high myopia, hyperopia, and astigmatism ranged from 15.3 % (Manchu) to 22.9 % (Han), 0.2 % (Yugur) to 2.8 % (Han), 21.6 % (Tibetan) to 48.9 % (Uyghur), and 38.7 % (Yi) to 57.5 % (Manchu) across different ethnicities, respectively. Compared to the Han population, the Mongolian (odds ratios (OR) 0.62, 95 % confidence interval (CI) 0.46-0.84, p = 0.002), Tibetan (OR 0.66, 95 % CI 0.52-0.85, p = 0.001), Uyghur (OR 0.63, 95 % CI 0.49-0.80, p < 0.001), Yi (OR 0.65, 95 % CI 0.46-0.92, p = 0.014), and Yugur (OR 0.65, 95 % CI 0.50-0.85, p = 0.001) ethnicities were less likely to have myopia. There was no significant difference in the prevalence of myopia between the Manchu, Korean, and Han ethnic groups. Factors associated with a lower prevalence of myopia included rural residence (p < 0.001), a body mass index (BMI) > 18.5 kg/m2 (all p < 0.001), residence in higher latitude areas (p = 0.020), and a history of smoking (p = 0.002 in the past smoking group, p = 0.031 in the current smoking group). The Mongolian (p = 0.006) and Yugur (p = 0.007) populations, participants living in rural areas (p = 0.012), and those with a BMI >24 kg/m2 (p = 0.038 in the >24.0 ≤ 27.0 kg/m2 group or p = 0.041 in the >27.0 kg/m2 group) were less likely to have high myopia. Factors associated with a higher prevalence of hyperopia included older age (all p < 0.001), rural residence (p = 0.039), higher latitude areas (p = 0.031), smoking history (p = 0.040), and Mongolian (p = 0.001), Uyghur (p < 0.001), Yi (p < 0.001), and Yugur (p = 0.002) ethnicities. Conversely, the Manchu population (p = 0.004) and individuals with higher education levels than illiteracy (p = 0.024 or p < 0.001) were less likely to have hyperopia. Conclusions: Myopia affected more than one-fifth of the older adults in the Han population in this survey. Significant differences in the prevalence of RE were observed between minority ethnicities and Han individuals, except for the Manchu and Korean groups.

RESUMEN

INTRODUCTION:  Previous studies have shown that isolated eyelid myokymia (EM) is usually caused by stress, fatigue, and caffeine consumption. The purpose of this study was to evaluate the association between EM and digital screen time, uncorrected refractive error, intraocular pressure (IOP), and blood electrolyte levels. METHODS: Between February 2023 and June 2024, 103 eyes of 103 patients who applied to the ophthalmology outpatient clinic with complaints of eyelid twitching lasting for more than two weeks and 103 eyes of 103 healthy individuals as a control group were included in the study. All participants were asked to record their daily time spent with digital screens for two weeks. Cycloplegic refractive error, IOP, optic nerve head cup/disc (C/D) ratio, and blood calcium, sodium, potassium, and magnesium levels were recorded and compared between the two groups. RESULTS: Mean digital screen time was 4.84±1.74 hours in the control group and 6.88±2.01 hours in the EM group. It was found that digital screen time was significantly higher in the EM group compared to the control group (p<0.001). There was a strong positive correlation between the duration of eyelid twitching and the time spent in front of digital screens (p<0.001, r=0.670). There was no significant difference in cycloplegic refractive error, IOP, C/D ratio, and blood electrolyte levels between the two groups (p>0.05). CONCLUSION: Prolonged digital screen time might play a role in the development of EM. On the other hand, no relationship was found between eyelid twitching and uncorrected refractive error, glaucoma, or blood electrolyte levels.

RESUMEN

This technical note extends a recent discussion in this journal of the role of validation study data in rational decision making. One argument that has been made in this context, using elements of Bayesian decision theory, is that further aggregation of validation study data into error rates involves a loss of information that compromises rational inference and decision making and should therefore be discouraged. This technical note seeks to explain that this argument can be developed at different levels of detail, depending on the definition of the propositions of interest, the forensic findings to be evaluated (and hence the form of the likelihood ratio), and the characterization of the relative desirability of decision consequences. The analyses proposed here reveal the cascade of abstractions and assumptions into which discussions about the use of validation study results in forensic science have fallen. This reinforces the conclusion that further aggregation of validation study data into error rates is problematic. It also suggests that even if a definition of error rate(s) could be agreed upon and defensively quantified in a given application, we should rethink and possibly adjust our expectations about what exactly error rates can practically contribute to rational modes of reasoning and decision making in legal contexts.

RESUMEN

We consider evaluating biomarkers for treatment selection under assay modification. Survival outcome, treatment, and Affymetrix gene expression data were attained from cancer patients. Consider migrating a gene expression biomarker to the Illumina platform. A recent novel approach allows a quick evaluation of the migrated biomarker with only a reproducibility study needed to compare the two platforms, achieved by treating the original biomarker as an error-contaminated observation of the migrated biomarker. However, its assumptions of a classical measurement error model and a linear predictor for the outcome may not hold. Ignoring such model deviations may lead to sub-optimal treatment selection or failure to identify effective biomarkers. To overcome such limitations, we adopt a nonparametric logistic regression to model the relationship between the event rate and the biomarker, and the deduced marker-based treatment selection is optimal. We further assume a nonparametric relationship between the migrated and original biomarkers and show that the error-contaminated biomarker leads to sub-optimal treatment selection compared to the error-free biomarker. We obtain the estimation via B-spline approximation. The approach is assessed by simulation studies and demonstrated through application to lung cancer data.

RESUMEN

INTRODUCTION: Reliable peripheral quantitative computed tomography (pQCT) assessment is essential to the accurate longitudinal reporting of bone and muscle quality. However, the between-day reliability of pQCT and the influence of age on outcome reliability is currently unknown. OBJECTIVE: To quantify the same- and between-day reliability of morphological pQCT at proximal and distal segments of the forearm, shank, and thigh, and explore the influence of participant body size, age, and sex on outcome reliability. METHODS: Men and women (49 % female, 18-85 years, n=72-86) completed two consecutive-day pQCT testing sessions, where repeat measurements were conducted on day-one for technical error, and between-day for biological error quantification. Testing was undertaken following best practice body composition testing guidance, including standardized presentation and consistent time-of-day. RESULTS: All measurements of bone were classified as having 'good' to 'excellent' reliability [intraclass correlation coefficient (r=0.786- 0.999], as were measurements of muscle area (ICC r=0.991-0.999) and total fat (r=0.996-0.999). However, between- and same-day muscle density measurements at the thigh and forearm were classified as 'poor' (r=0.476) and 'moderate' (r=0.622), respectively. Likewise, intramuscular fat area at the thigh was classified as 'moderate' (r=0.737) for between-day measurement. Biological error was inflated compared to technical error by an average of 0.4 % for most measurements. Error values tended to increase proportionally with the amount of tissue quantified and males had significantly greater biological error for measurement of distal tibial bone (p<0.002) and trabecular area (p<0.002). Biological error was inflated among older adults for measurement of forearm muscle density (p<0.002). CONCLUSIONS: Most pQCT outcomes can be implemented with confidence, especially outcomes that assess bone area and density at any of the radial, tibial, and femoral sites investigated herein. However, it is important to account for the influence of biological measurement error in further studies, especially for muscle and intramuscular fat outcomes derived by pQCT.

RESUMEN

A central mechanism of human action control is the prompt binding between actions and the stimuli provoking them. Perceiving the same stimuli again retrieves any bound responses, facilitating their execution. An open question is whether such binding and retrieval only emerges when stimulus-response rules are known upon taking action or also when agents are forced to guess and receive feedback about whether they were successful or not afterward. In two experiments, we tested the hypothesis that knowing rules before responding would boost binding between stimuli and responses during action-taking relative to guessing situations. Second, we assessed whether the content of the feedback matters for binding in that agents might use feedback to build correct stimulus-response bindings even for wrong guesses. We used a sequential prime-probe design to induce stimulus-response binding for prime responses that were either rule-based or guesses, and to measure retrieval of these bindings in response times and errors in the probe. Results indicate that binding and retrieval emerge for successful but not for wrong guesses. Binding effects for correct guesses were consistently small in effect size, suggesting that pre-established stimulus-response bindings from instructed rules might indeed boost binding when taking action.

RESUMEN

BACKGROUND: Elevated level of double-negative T (DNT) cells is a historical hallmark of autoimmune lymphoproliferative syndrome (ALPS) diagnosis. However, the peripheral blood level of DNT cells might also be compromised in autoimmune lymphoproliferative immunodeficiencies (ALPID) other than ALPS, inattention to which would increase the delay in diagnosis of the underlying genetic defect and hinder disease-specific treatment. MATERIALS AND METHODS: This cross-sectional study recruited patients suffering from ALPID (exclusion of ALPS) with established genetic diagnosis. Following thorough history taking, immunophenotyping for lymphocyte subsets was performed using BD FACS CaliburTM flowcytometry. RESULTS: Fifteen non-ALPS ALPID patients (60% male and 40% female) at a median (interquartile range: IQR) age of 14.0 (7.6-21.8) years were enrolled. Parental consanguinity and family history of immunodeficiency were present in 8 (53.3%) patients. The median (IQR) age at first presentation, clinical and molecular diagnosis were 18 (4-36) months, 8.0 (4.0-17.0) years, and 9.5 (5.0-20.9) years, respectively. Molecular defects were observed in these genes: LRBA (3, 20%), CTLA-4 (2, 13.3%), BACH2 (2, 13.3%), AIRE (2, 13.3%), and FOXP3, IL2Rß, DEF6, RASGRP1, PIK3CD, and PIK3R1 each in one patient (6.7%). The most common manifestations were infections (14, 93.3%), autoimmunity (12, 80%), and lymphoproliferation (10, 66.7%). The median (IQR) count of white blood cells (WBCs) and lymphocytes were 7160 (3690-12,600) and 3266 (2257-5370) cells/mm3, respectively. The median (IQR) absolute counts of CD3+ T lymphocytes and DNTs were 2085 (1487-4222) and 18 (11-36) cells/mm3, respectively. Low lymphocytes and low CD3+ T cells were observed in 3 (20%) patients compared to normal age ranges. Only one patient with FOXP3 mutation had DNT cells higher than the normal range for age. CONCLUSIONS: Most non-ALPS ALPID patients manifested normal DNT cell count. For a small subgroup of patients with high DNT cells, defects in other IEI genes may explain the phenotype and should be included in the diagnostic genetic panel.

Asunto(s)

RESUMEN

Only very limited information is available on why some nonsynonymous variants severely alter gene function while others have no effect. To identify the characteristic features of mutations that strongly influence gene function, this study focused on SRK which encodes a highly polymorphic receptor kinase expressed in stigma papillary cells that underlies a female determinant of self-incompatibility in Brassicaceae. A set of 300 Arabidopsis thaliana transformants expressing mutated SRKb from A. lyrata was constructed using error-prone PCR and the genotype and self-incompatibility phenotype of each transformant were determined. Almost all the transformants showing the self-incompatibility defect contained mutations in AlSRKb that altered localization to the plasma membrane. The observed mutations occurred in amino acid residues that were highly conserved across S haplotypes and whose predicted locations were in the interior of the protein. Our findings suggested that mutations causing the self-incompatibility defect were more likely to result from changes to AlSRKb biosynthesis than from loss of AlSRKb function. In addition, we examined whether the RandomForest and Extreme Gradient Boosting methods could predict the self-incompatibility phenotypes of SRK mutants.