Absolute Value Of Prediction Error Research Articles

Refractive outcomes after immediate primary phacoemulsification for acute primary angle closure

This study investigated the refractive outcomes of 64 eyes overall including 32 immediate primary phacoemulsification in acute primary angle closure (APAC) eyes and 32 of their fellow eyes. We investigated best-corrected visual acuity, intraocular pressure (IOP), average keratometric diopter (K), spherical equivalent, axial length (AL), central corneal thickness, and anterior chamber depth (ACD) at preoperative examination (Pre) and more than 1-month post-phacoemulsification (1 m), and changes in values. Using SRK/T, Barrett Universal II (Barrett), Hill-Radial Basis Function Version 3.0 (RBF 3.0), and Kane formulas, we calculated and compared refractive prediction error (PE), absolute value of PE (AE), and changes in K, AL, and ACD from Pre to 1 m between APAC and fellow eyes. From Pre to 1 m, K remained similar in APAC and fellow eyes (p = 0.069 and p = 0.082); AL significantly decreased in APAC and in fellow eyes (both p < 0.001); and ACD significantly increased in APAC and in fellow eyes (both p < 0.001). The change in AL differed significantly between the two groups (p = 0.007). Compared to the fellow eyes, PE with SRK/T and Barret formulas (p = 0.0496 and p = 0.039) and AE with Barrett and RBF 3.0 formula (p = 0.001 and p = 0.024) were significantly larger in the APAC eyes. Thus, attention should be paid to refractive prediction error in immediate primary phacoemulsification for APAC eyes caused by preoperative AL elongation due to high IOP.

Accuracy of optimized Sirius ray-tracing method in intraocular lens power calculation.

To evaluate the accuracy and predictability of ray tracing-assisted intraocular lens (IOL) calculation function in Sirius internal software and further improve the accuracy by optimizing the calculation of predicted lens position (PLP). This retrospective study recruited 52 eyes of 49 patients. All of the cases with cataract had undergone phacoemulsification combined with IOL implantation. SRK-T, Haigis formula, and Sirius ray-tracing method were all used for each eye's IOL calculation. The mean absolute value of prediction error (prediction error=predicted refraction-postoperative refraction) was defined as mean absolute prediction error (MAPE) and was determined for each method. Calculation of PLP was optimized by effective lens position (ELP). Optimized PLP was entered to Sirius internal software again to verify whether the method was improved. Compared with SRK-T and Haigis formulas, less accuracy was shown in Sirius ray-tracing method (P=0.001). The ELP of the IOL moved forward compared to PLP (P<0.001). The MAPE of the ELP-inputted Sirius ray-tracing method was reduced. ELP and PLP were well correlated. Taking ELP as y and PLP given by Sirius soft as x, a linear regression formula y=0.1637x+3.1741 was concluded (R2 =0.1066, P=0.018). It was shown that the optimized Sirius ray-tracing method (optimized PLP entered), compared with SRK-T and Haigis formulas, worked with the same accuracy (P=0.038). The original Sirius ray tracing method is not satisfactory enough. However, in normal eyes, the optimized Sirius ray-tracing method in IOL calculation was as accurate as SRK-T and Haigis formulas.

Reversible data hiding in encrypted images based on pixel prediction and multi-MSB planes rearrangement

Great concern of reversible data hiding in encrypted images (RDHEI) has arisen due to the development of cloud computing and privacy protection. RDHEI is an effective technology that can embed additional data after image encryption, extract additional data error-free and reconstruct original images losslessly. In this paper, a high-capacity and fully reversible RDHEI method is proposed, which is based on pixel prediction and multi-MSB (most significant bit) planes rearrangement. First, the median edge detector (MED) predictor is used to calculate the predicted value. Next, unlike previous methods, in the proposed method, signs of prediction errors (PEs) are represented by one bit plane and absolute values of PEs are represented by other bit planes. Then, we divide bit planes into uniform blocks and non-uniform blocks, and rearrange these blocks. Finally, according to different pixel prediction schemes, different sizes of additional data are embedded adaptively. The experimental results prove that our method has a higher embedding capacity compared with state-of-the-art RDHEI methods.

Surprise Acts as a Reducer of Outcome Value in Human Reinforcement Learning.

Surprise occurs because of differences between a decision outcome and its predicted outcome (prediction error), regardless of whether the error is positive or negative. It has recently been postulated that surprise affects the reward value of the action outcome; studies have indicated that increasing surprise as an absolute value of prediction error decreases the value of the outcome. However, how surprise affects the value of the outcome and subsequent decision making is unclear. We suggest that, on the assumption that surprise decreases the outcome value, agents will increase their risk-averse choices when an outcome is often surprising. Here, we propose the surprise-sensitive utility model, a reinforcement learning model that states that surprise decreases the outcome value, to explain how surprise affects subsequent decision making. To investigate the properties of the proposed model, we compare the model with previous reinforcement learning models on two probabilistic learning tasks by simulations. As a result, the proposed model explains the risk-averse choices like the previous models, and the risk-averse choices increase as the surprise-based modulation parameter of outcome value increases. We also performed statistical model selection by using two experimental datasets with different tasks. The proposed model fits these datasets better than the other models with the same number of free parameters, indicating that the model can better capture the trial-by-trial dynamics of choice behavior.

Local tangent space alignment and relevance vector machine as nonlinear methods for estimating sensory quality of tea using NIR spectroscopy

The quality of green tea is conventionally tested by human sensory panel, who assign quality scores to different teas. In order to overcome the deficiency of sensory evaluation methods, near infrared (NIR) reflectance spectroscopy combined with nonlinear methods was used to evaluate the sensory quality of green tea in this study. Nonlinear dimension reduction methods such as Kernel Principal Component Analysis (KPCA), Sparse Principal Component Analysis (SPCA), and Local Tangent Space Alignment (LTSA) were used to reduce the dimension of the spectral data, and compared with linear dimension reduction methods such as Principal Component Analysis (PCA), Multidimensional Scaling (MDS). Relevance vector machine (RVM) as a nonlinear quantitative analysis method was performed comparatively to develop calibration models between the sensory scores of the sensory panels and objective determination (NIR spectra measurements), and compared with back-propagation artificial network and linear PLS model. Experiment results showed that the non-linear reduction algorithm LTSA could obtain better effect on spatial reduction dimensional for NIR spectral data of tea. And when the number K of the neighborhoods in LTSA was 15, the RMSECV of RVM model was the minimum. The performance of LTSA-RVM model was superior to the others when 8 intrinsic variables were used. In the prediction set, Rp, RMSEP and RPD were 0.963, 1.461 and 3.788, respectively. The absolute value of prediction error of all prediction samples in LTSA-RVM are below 3.0, which indicated that the sensory scores predicted by NIR were consistent with the scores by tea assessors. The results of recovery study were 98.42%–102.81% at 5 real tea samples and the relative error lower than 3%. The research demonstrates that NIR technology combined with nonlinear analysis methods can effectively estimate sensory quality of tea and also present a potential way for fast and effective determination of the sensory quality of green tea.

Short-Term Rain Attenuation Predictor for Terrestrial Links in Tropical Area

This letter describes a short-term rain attenuation predictor model for terrestrial radio links located in tropical area. The predictor is based on the fade-slope statistics that are modeled by a generalization of Van de Kamp (VDK) distribution for rain attenuation levels between 1 and 30 dB. The predictor can improve adaptive fade countermeasures as it predicts the rain attenuation in the next 10 s based on the current attenuation value. The model is applied to data collected in four links located in tropical region, and its results show that the absolute values of prediction errors are under 2.9 dB for 99.9% of the predictions.

Parameterization of the nonlinear Greham predictor for digital image compression

Parameterization of the nonlinear Greham predictor is performed for a digital image compression method based on the differential pulse code modulation. The predictor automatically selects different methods for calculating each pixel in the image based on the availability and intensity of that pixel contour. A fast learning procedure that optimizes the prediction parameters is performed before the actual compression. In the course of optimization, the minimum sum of absolute values of prediction errors is provided. For this purpose, a recursive procedure is used, whose computational complexity is independent of the image size. The estimation of the computational complexity of the proposed predictor is conducted. To study the predictors computational experiments are carried out on real images. A gain the proposed predictor offers in terms of the root mean square error when compared with the prototypes is demonstrated. In addition, a gain that the compression method based on the differential pulse code modulation with the proposed predictor has over the JPEG compression method in terms of the maximum error is demonstrated.

H∞Guaranteed Cost Control for Networked Control Systems under Scheduling Policy Based on Predicted Error

Scheduling policy based on model prediction error is presented to reduce energy consumption and network conflicts at the actuator node, where the characters of networked control systems are considered, such as limited network bandwidth, limited node energy, and high collision probability. The object model is introduced to predict the state of system at the sensor node. And scheduling threshold is set at the controller node. Control signal is transmitted only if the absolute value of prediction error is larger than the threshold value. Furthermore, the model of networked control systems under scheduling policy based on predicted error is established by taking uncertain parameters and long time delay into consideration. The design method ofH∞guaranteed cost controller is presented by using the theory of Lyapunov and linear matrix inequality (LMI). Finally, simulations are included to demonstrate the theoretical results.

On autoregressive moving-average models as a tool of virtual stock-exchange: experimental investigation

The objective of this work is to investigate experimentally the well-known autoregressive models as simplest algorithms simulating prediction processes of the stockholders using the historical stock rates only. The “virtual” stock exchange which applies these algorithms can help in testing various assumptions of investor behavior. To represent users that prefer linear utility functions, the autoregressive moving-average model (ARMA-ABS(p, q)), minimizing the absolute values of prediction errors is regarded, in addition to the traditional ARMA(p, q) model which minimize the least square errors. The results of two hundred actual financial time series and a hundred of virtual ones are discussed in short.

Refractive accuracy after intraocular lens implantation in pediatric cataract.

To analyze the factors that influence the prediction error (PE) after intraocular lens (IOL) implantation in pediatric cataract. The medical records of cataract patients of no more than 14 years old who had primary IOL implantation were reviewed from 2006 to 2010. The PE, absolute value of PE (APE), and predictability between in different axial length, mean corneal curvature, corneal astigmatism, and age at the surgery were analyzed. Seventy-five children (119 eyes) were included, with a mean age of (5.09±2.54) years. At the follow-up of (1.19±0.69) months, the mean postoperative PE was (-0.22±1.12) D, and APE was (0.87±0.73)D. The PE in eyes with an axial length >20mm but ≤22mm were significantly under-corrected than that in eyes with longer axis, and the APE in eyes with an axial length ≤20mm was more obvious compared with the others. The correlations between PE and axial length, as well as corneal astigmatism, and between APE and axial length were significant. The predictability was significantly poorer in the eyes with an axial length ≤20mm than the others. The axial length is closely related with the PE after IOL implantation in pediatric cataract patients, especially when it is ≤20mm, PE is more significant. The formula that is more suitable to very short axial length should be explored.

Calculation of intraocular lens power after corneal refractive surgery using ray tracing method

Background The calculation of intraocular lens(IOL) power is important for cataract patient following corneal refractive surgery.Theoretically,ray tracing for intraocular lens calculation is an ideal method because pseudophakic optics can be imitated by this method.ObjectiveThis study is to introduce ray tracing method for IOL power calculation after corneal refractive surgery and evaluate the accuracy of ray tracing method.MethodsPseudophakic eye models to perform ray tracing calculations were constructed using optical software Zemax(Zemax-EE Optical Design Program,Focus Software,Inc.USA) by importing the corneal and IOL radii,refractive indexes of cornea,aqueous,IOL and vitreous.Ten eyes of Chinese patients and 15 eyes of German patients undergone corneal refractive surgery were enrolled in this study.Corneal radius was measured by Orbscan Ⅱ z or C-Scan.Axial length was measured by IOL Master or A-scan.The refraction after cataract surgery was predicted by ray tracing method.The difference between actual postoperative refraction and predicted refraction was defined as prediction error,and absolute value of prediction error was defined as absolute prediction error.The mean value,standard deviation and distribution of both prediction error and absolute prediction error were calculated using SPSS11.0 software.ResultsA pseudophakic eye model for IOL power calculation was constructed in this study.Mean prediction error was 0.28±0.73 diopter(D)(range-1.09 D to 1.91 D);mean absolute prediction error was 0.63±0.45 D(range 0.01 D to 1.91 D).For absolute prediction error,9 eyes(36%) were confined within 0.5 D;12 eyes(48%) were confined between 0.5 D and 1.0 D;3 eyes(12%) were confined between 1.0 D and 1.5 D;1 eye(4%) was confined between 1.5 D and 2.0 D.Totally 21 eyes(84%) were confined within 1.0 D.ConclusionRay tracing method is a reliable method for the calculation of intraocular lens power after corneal refractive surgery.

Refractive outcomes with secondary intraocular lens implantation in children

To determine the accuracy of postoperative predicted refractive outcomes in surgically aphakic pediatric patients undergoing secondary intraocular lens (IOL) implantation. Comparisons were also made with other variables historically considered important in cataract surgery. Retrospective review of 50 eyes from 35 consecutive pediatric patients (< or =18 years of age) undergoing secondary IOL implantation within the ciliary sulcus or posterior capsular bag. All cases were performed by 1 of 2 surgeons and all refractions were performed manually using the retinoscope by an experienced pediatric ophthalmologist. The mean patient age at the time of the secondary implantation was 6.5 years of age (range, 0.58-15.04). The mean patient age at the time of the primary cataract extraction was 0.78 years (range, 0.08-5.77). For all patients, mean absolute value of prediction error was 1.64 D (SD 1.58 D). There were no significant associations between mean absolute value of prediction error and any of the variables measured including axial length, corneal mean curvature, bag or sulcus implantation, formula used, or age at primary and secondary surgery (p > 0.05). The mean absolute value of prediction error observed in this study is consistent with previous pediatric primary and secondary IOL data in that it shows a considerable difference from that expected in adult populations. Our findings support the argument that methods currently employed to calculate IOL power may fail to accurately account for all the variations in the eyes of pediatric patients.