A machine learning-based algorithm for estimating the original corneal curvature based on corneal topography after orthokeratology.

OBJECTIVE: To estimate the original corneal curvature after orthokeratology by applying a machine learning-based algorithm. METHODS: A total of 497 right eyes of 497 patients undergoing overnight orthokeratology for myopia for more than 1 year were enrolled in this retrospective study. All patients were fitted with lenses from Paragon CRT. Corneal topography was obtained by a Sirius corneal topography system (CSO, Italy). Original flat K (K1) and original steep K (K2) were set as the targets of calculation. The importance of each variable was explored by Fisher's criterion. Two machine learning models were established to allow adaptation to more situations. Bagging Tree, Gaussian process, support vector machine (SVM), and decision tree were used for prediction. RESULTS: K2 after one year of orthokeratology (K2after) was most important in the prediction of K1 and K2. Bagging Tree performed best in both models 1 and 2 for K1 prediction (R = 0.812, RMSE = 0.855 in model 1 and R = 0.812, RMSE = 0.858 in model 2) and K2 prediction (R = 0.831, RMSE = 0.898 in model 1 and R = 0.837, RMSE = 0.888 in model 2). In model 1, the difference was 0.006 ± 1.34 D (p = 0.93) between the predictive value of K1 and the true value of K1 (K1before) and was 0.005 ± 1.51 D(p = 0.94) between the predictive value of K2 and the true value of K2 (K2before). In model 2, the difference was -0.056 ± 1.75 D (p = 0.59) between the predictive value of K1 and K1before and was 0.017 ± 2.01 D(p = 0.88) between the predictive value of K2 and K2before. CONCLUSION: Bagging Tree performed best in predicting K1 and K2. Machine learning can be applied to predict the corneal curvature for those who cannot provide the initial corneal parameters in the outpatient clinic, providing a relatively certain degree of reference for the refitting of the Ortho-k lenses.

An investigation into the causes of abnormal waste of Ortho-K lenses.

Purpose: To investigate the reasons for wasting orthokeratology (OK) lenses due to breakage or loss, provide more comprehensive guidelines for the clinical care of lenses and minimize time and costs for patients due to excessive broken and lost lenses. Methods: A survey was administered to clinic outpatients who had broken or lost their OK lenses before the regularly scheduled replacement cycle (1-1.5 years). The association between the frequency of OK lens breakage and daily care was assessed using Fisher's exact test and multivariable ordered logistic regression analysis. Results: A total of 306 valid questionnaires were collected. Among the subjects, 141 were male, and 165 were female, with a mean age of 10.57 ± 2.00 years (range: 6-18 years). In the investigation of the causes of OK lens waste, 81.4% of the patients reported lens breakage, 13.1% lost their lenses, and 5.6% of patients experienced both fragmentation and lens loss. More than half of the patients (52.90%) used incorrect lens cleaning techniques. In further analysis of the relationship between the frequency of OK lens fragmentation within a year and daily care habits, a significant difference was observed between the caregiver (P = 0.03) and whether the lenses were cleaned promptly after removal (P < 0.001). Mothers as daily caregivers of OK lenses had a lower frequency of fragmentation in a year compared to nanny or grandparents (P = 0.014, OR = 0.33, 95% CI = 0.13, 0.80). The failure to clean the lenses according to eye care practitioners' guidance was a risk factor for the frequent breakage of OK lenses (P < 0.001. OR = 5.29, 95% CI = 3.15, 8.89). Conclusions: The causes of OK lens waste were mainly attributed to caregivers, care practices and some unexpected situations that can be avoided through optometrists' reminders. Regardless of the reasons for noncompliant behavior leading to breakage or loss of OK lenses, all of the complications can probably be addressed by better and more frequent reinforcement of care procedures by practitioners. Better clinical guidance measures and more frequent reminders could prevent a large proportion of abnormal waste of OK lenses.

Weighted Zernike defocus coefficient of treatment zone is a meaningful indicator for myopia control efficacy of Ortho-K lenses.

BACKGROUND: The goal of this study was to reproduce a three-dimensional representation of corneal defocus characteristics after orthokeratology (Ortho-K) treatment via an indicator defined as the weighted Zernike defocus coefficient of the treatment zone (Cweighted defocus). This could be used to predict the effectiveness of Ortho-K treatment quantitatively in a timely manner after the one-month visit. METHODS: Seventy myopic children with axial length (AL) elongation after Ortho-K treatment (group A) and 63 myopic children with AL shortening after Ortho-K treatment (group B) were included in this one-year retrospective study. The proposed indicator was calculated by a customized MATLAB program. Multivariate binomial logistic regression and multivariate linear regression analyses were used to explore the association between AL change and the Cweighted defocus, age, sex, and other ocular biometric parameters. RESULTS: The 12-month AL change, age, pupil diameter, and vertical decentration of the Ortho-K lens were significantly different between the two groups. Multivariate logistic regression analysis showed that a larger Cweighted defocus (≥ 0.35 D/mm2) (OR: 0.224; 95% CI: 0.078-0.646; P = 0.006) was correlated with the emergence of AL shortening after orthokeratology treatment. A multivariate linear regression model showed that a greater Cweighted defocus was associated with slower 12-month AL elongation (ß = - 0.51, P = 0.001). CONCLUSIONS: The Cweighted defocus is an effective predictive indicator of myopia control, and a larger Cweighted defocus may lead to slower elongation of AL. This meaningful indicator may help in the evaluation and adjustment of Ortho-K lens parameters in a timely manner and minimize the cost of clinical trial and error.

7-Methylxanthine Influences the Behavior of ADORA2A-DRD2 Heterodimers in Human Retinal Pigment Epithelial Cells.

INTRODUCTION: The goal of this study was to investigate the presence of ADORA2A-DRD2 heterodimers in human retinal pigment epithelial (RPE) cells; determine if 7-methylxanthine (7-MX), a nonselective adenosine receptor antagonist which was used to control myopia progression, can influence the behavior of RPE cells through the ADORA2A-DRD2 receptor pathway; and assess the changes in the expression of signaling molecules during cellular signal transduction. METHODS: Human RPE cells were cultured in vitro in the presence or absence of 7-MX. Cell proliferation was evaluated with the CCK-8 assay. Apoptosis and necrosis rates were determined by annexin V-FITC/propidium iodide staining and flow cytometry. Immunofluorescence and coimmunoprecipitation were used to examine the protein expression and colocalization of ADORA2A and DRD2 in RPE cells. ADORA2A and DRD2 were knocked down with small interfering RNAs (siRNAs). Changes in the protein expression of ERK1/2 and phospho-ERK1/2 (pERK 1/2), which are signaling molecules downstream of dopamine receptors, were evaluated by Western blot analysis. RESULTS: Immunofluorescence and coimmunoprecipitation showed that ADORA2A and DRD2 were colocalized in RPE cells. The expression of ADORA2A in RPE cells was inhibited by treatment with 50 µmol/L 7-MX for 48 h, and the expression of DRD2, ERK1/2, and pERK1/2 was increased after treatment with 50 µmol/L 7-MX for 48 h. After siRNA-mediated knockdown of DRD2 in RPE cells and further treatment with 50 µmol/L 7-MX for 48 h, the expression of DRD2 was nearly restored to the level observed in the native control. At the experimental concentrations, 7-MX and siRNAs did not affect the proliferation or apoptosis of human RPE cells. CONCLUSIONS: ADORA2A and DRD2 heterodimers were present in RPE cells. 7-MX may affect the behaviors of RPE cells through the ADORA2A-DRD2 receptor pathway. 7-MX is an inhibitor of ADORA2A receptors that can prevent inhibition of the DRD2 receptor pathway and increase DRD2 receptor pathway activity. This phenomenon may explain the mechanism of action through which 7-MX can control myopia progression.

Machine learning algorithm improves accuracy of ortho-K lens fitting in vision shaping treatment.

PURPOSE: To construct a machine learning (ML)-based model for estimating the alignment curve (AC) curvature in orthokeratology lens fitting for vision shaping treatment (VST), which can minimize the number of lens trials, improving efficiency while maintaining accuracy, with regards to its improvement over a previous calculation method. METHODS: Data were retrospectively collected from the clinical case files of 1271 myopic subjects (1271 right eyes). The AC curvatures calculated with a previously published algorithm were used as the target data sets. Four kinds of machine learning algorithms were implemented in the experimental analyses to predict the targeted AC curvatures: robust linear regression models, support vector machine (SVM) regression models with linear kernel functions, bagging decision trees, and Gaussian processes. The previously published calculation method and the novel machine learning method were then compared to assess the final parameters of ordered lenses. RESULTS: The linear SVM and Gaussian process machine learning models achieved the best performance. The input variables included sex, age, horizontal visible iris diameter (HVID), spherical refraction (SER), cylindrical refraction, eccentricity value (e value), flat K (K1) and steep K (K2) readings, anterior chamber depth (ACD), and axial length (AL). The R-squared values for the output AC1K1, AC1K2 and AC2K1 values were 0.91, 0.84, and 0.73, respectively. The previous calculation method and machine learning methods displayed excellent consistency, and the proposed methods performed best based on flat K reading and e values. CONCLUSIONS: The ML model can provide practitioners with an efficient method for estimating the AC curvatures of VST lenses and reducing the probability of cross-infection originating from trial lenses, which is especially useful during pandemics, such as that for COVID-19.

Machine learning based strategy surpasses the traditional method for selecting the first trial Lens parameters for corneal refractive therapy in Chinese adolescents with myopia.

PURPOSE: Return zone depth (RZD) and landing zone angle (LZA) are important parameters of corneal refractive therapy (CRT) lenses. A new machine learning algorithm is proposed for prescribing CRT lens parameters in Chinese adolescents with myopia. METHODS: This is a retrospective study. In total, 1037 Chinese adolescents with myopia (1037 right eyes) were enrolled. A calculation model based on corneal elevation maps was constructed to calculate RZD and LZA for the four quadrants. Furthermore, multiple linear regression and optimized machine learning models were established to predict RZD and LZA values for different combinations of age, sex, and ocular parameters. The four methods (sliding card, linear regression, calculation and optimized machine learning) were then compared to the parameters of the final ordered lens. RESULTS: The optimized machine learning pipeline achieved the best performance. Age, sex, horizontal visible iris diameter (HVID), spherical equivalent refraction degree (SER), eccentricity (e), keratometric (K) readings, corneal astigmatism (CA), axial length (AL), AL/corneal curvature ratio (AL/MK), and anterior chamber depth (ACD) were significant to the machine learning model. The R values for the nasal, temporal, superior and inferior LZA based on machine learning were 0.843, 0.693, 0.866 and 0.762, respectively, and those for the RZD were 0.970, 0.964, 0.975 and 0.964, respectively. CONCLUSIONS: The feasibility and efficiency of an optimized machine learning method to predict LZA and RZD parameters has been demonstrated. The advantage of the proposed method is that it is more accurate, easier to use and faster to implement than the traditional sliding card method.

A machine learning-based algorithm used to estimate the physiological elongation of ocular axial length in myopic children.

BACKGROUND: Axial myopia is the most common type of myopia. However, due to the high incidence of myopia in Chinese children, few studies estimating the physiological elongation of the ocular axial length (AL), which does not cause myopia progression and differs from the non-physiological elongation of AL, have been conducted. The purpose of our study was to construct a machine learning (ML)-based model for estimating the physiological elongation of AL in a sample of Chinese school-aged myopic children. METHODS: In total, 1011 myopic children aged 6 to 18 years participated in this study. Cross-sectional datasets were used to optimize the ML algorithms. The input variables included age, sex, central corneal thickness (CCT), spherical equivalent refractive error (SER), mean K reading (K-mean), and white-to-white corneal diameter (WTW). The output variable was AL. A 5-fold cross-validation scheme was used to randomly divide all data into 5 groups, including 4 groups used as training data and one group used as validation data. Six types of ML algorithms were implemented in our models. The best-performing algorithm was applied to predict AL, and estimates of the physiological elongation of AL were obtained as the partial derivatives of AL predicted -age curves based on an unchanged SER value with increasing age. RESULTS: Among the six algorithms, the robust linear regression model was the best model for predicting AL, with a R 2 value of 0.87 and relatively minimal averaged errors between the predicted AL and true AL. Based on the partial derivatives of the AL predicted -age curves, the estimated physiological AL elongation varied from 0.010 to 0.116 mm/year in male subjects and 0.003 to 0.110 mm/year in female subjects and was influenced by age, SER and K-mean. According to the model, the physiological elongation of AL linearly decreased with increasing age and was negatively correlated with the SER and the K-mean. CONCLUSIONS: The physiological elongation of the AL is rarely recorded in clinical data in China. In cases of unavailable clinical data, an ML algorithm could provide practitioners a reasonable model that can be used to estimate the physiological elongation of AL, which is especially useful when monitoring myopia progression in orthokeratology lens wearers.

The contribution of miR-122 to the innate immunity by regulating toll-like receptor 4 in hepatoma cells.

BACKGROUND: Hepatocellular carcinoma (HCC) is a kind of malignancies to impact human health. It has been reported that aberrant toll-like receptor (TLR) signaling may contribute to the development and progression of HCC, especially TLR4. MiR-122, which extensively involved in hepatitis virus infection and the apoptosis of hepatoma cells, might be decreased in HCC patients livers. The hypothesis of this study was whether miR-122 plays a role in inflammatory pathways through regulating TLR4 expression in hepatoma cells. METHODS: The expression of miR-122 in the tissues of HCC patients compared to controls in TCGA datasets was analyzed. The relationship between miR-122 and TLR4 was detected in HCC cell lines by increasing/decreasing miR-122 expression. The target of miR-122 on TLR4 was confirmed by luciferase reporter assays. The proliferation of HCC cells and production of proinflammatory cytokines were measured with miR-122 upregulation and inhibition. RESULTS: We found that the expression of miR-122 was decreased in HCC tissues and showed the diagnostic capacity for HCC in TCGA datasets. MiR-122 and TLR4 expression have negative correlation in normal liver cells and HCC cells. Upregulation of miR-122 significantly inhibited TLR4 expression in hepatoma cells, including in hepatoma cells with the induction of LPS, while knocking down miR-122 increased TLR4 expression. By screening potential miR-122 targets among TLR4, we found that there was a putative miR-122 target in TLR4 3'UTR. Mutations in the nt1603-nt1609 region of TLR4 3'UTR abandoned the impact of miR-122 on TLR4 expression. Over-expression/down-expression of miR-122 could influence the proliferation and the expression of natural immune factors. CONCLUSIONS: MiR-122 might target TLR4 and regulate host innate immunity in hepatoma cells, which revealed a new molecular mechanism of miR-122 on the regulation of innate immunity.

Use of antibiotics by urban and rural residents in Heilongjiang Province, China: cross-sectional study.

OBJECTIVE: To identify and compare the factors affecting the knowledge of, attitude towards and use of antibiotics between urban and rural residents in China. METHODS: A total of 3631 urban and rural residents in Heilongjiang Province, China, were selected using random sampling. Questionnaires recorded demographic characteristics and participant knowledge of, attitude towards and use of antibiotics. The responses of rural and urban residents were compared, and logistic regression analysis was applied to identify the factors that may contribute to the knowledge of, attitude towards and use of antibiotics. RESULTS: The majority of the participants (>60%) were aware that antibiotics could be used to treat bacterial infections and that bacteria could be resistant to antibiotics. However, only roughly half (40-60%) of the participants were aware that bacterial resistance to antibiotics had become a problem in China. Urban participants reported a more adequate knowledge of, attitude towards and use of antibiotics than rural participants. Logistic regression analysis indicated that urban residency, female gender and level of education were associated with knowledge of, attitude towards and use of antibiotics. CONCLUSIONS: Within our sample in Heilongjiang Province, the knowledge of, attitude towards and use of antibiotics were suboptimal in roughly half of all urban and rural residents, but better in urban than in rural residents. Targeted interventions to educate rural residents in particular may reduce the misuse of antibiotics.

Local injection of M-CH combined with i.p. hyperthermic hypo-osmolar infusion is an effective therapy in advanced gastric cancer.

Treatment failure of surgically treated gastric cancer is attributed to the spread of gastric cancer cells into the abdominal cavity and lymphatic or hematogenic canals. In the present study, local injection of mitomycin C bound to activated carbon (M-CH) combined with i.p. hyperthermic hypo-osmolar infusion (IPHHOI) was intraoperatively administered to prevent lymph node recurrence and peritoneal recurrence of gastric cancer. Between April 1998 and August 1999, 79 patients with advanced gastric cancer were allocated randomly to two groups. Forty patients underwent M-CH plus IPHHOI combined with surgery (M-CH1+IPHHOI group) and the remaining 39 underwent surgery alone (control group). Lymph node and peritoneal recurrence were significantly decreased in the M-CH1+IPHHOI group compared to that in the control group (p<0.05). The 1- and 2-year survival rates for the M-CH1+IPHHOI group were 91.2 and 72.1%, and those for the control group were 78.9 and 45.5%. The M-CH1IPHHOI group reaped a significant survival benefit (p=0.0352) compared to the control group. Although this study was conducted randomly for a small number of patients and short time, compared with the control group, the M-CH1+IPHHOI group had a beneficial effect in preventing lymph node recurrence and peritoneal recurrence after curative gastrectomy for advanced gastric cancer.