Point Error Research Articles

Model selection for long-term load forecasting under uncertainty

Purpose The purpose of this study is to propose a systematic model selection procedure for long-term load forecasting (LTLF) for ex-ante and ex-post cases considering uncertainty in exogenous predictors. Design/methodology/approach The different variants of regression models, namely, Polynomial Regression (PR), Generalised Additive Model (GAM), Quantile Polynomial Regression (QPR) and Quantile Spline Regression (QSR), incorporating uncertainty in exogenous predictors like population, Real Gross State Product (RGSP) and Real Per Capita Income (RPCI), temperature and indicators of breakpoints and calendar effects, are considered for LTLF. Initially, the Backward Feature Elimination procedure is used to identify the optimal set of predictors for LTLF. Then, the consistency in model accuracies is evaluated using point and probabilistic forecast error metrics for ex-ante and ex-post cases. Findings From this study, it is found PR model outperformed in ex-ante condition, while QPR model outperformed in ex-post condition. Further, QPR model performed consistently across validation and testing periods. Overall, QPR model excelled in capturing uncertainty in exogenous predictors, thereby reducing over-forecast error and risk of overinvestment. Research limitations/implications These findings can help utilities to align model selection strategies with their risk tolerance. Originality/value To propose the systematic model selection procedure in this study, the consistent performance of PR, GAM, QPR and QSR models are evaluated using point forecast accuracy metrics Mean Absolute Percentage Error, Root Mean Squared Error and probabilistic forecast accuracy metric Pinball Score for ex-ante and ex-post cases considering uncertainty in the considered exogenous predictors such as RGSP, RPCI, population and temperature.

Deformation Analysis of Existing Buildings Affected by Shield Tunnels Based on Intelligent Inversion and Measured Data

In the construction of urban underground shield tunnels, uneven deformation can easily occur when the shield passes through soft soil and other poor strata. Such deformation has a significant impact on surface settlement and may cause potential safety hazards to the surrounding existing buildings, directly affecting the safety of urban operation. When simulating and predicting surface settlements, the small-strain soil hardening model can more accurately characterize the mechanical parameters of soil. Nevertheless, its parameters are numerous and complicated to determine accurately, so parameter inversion is needed to determine the accurate parameters of the soft soil layer in order to more accurately predict the surface settlement. This study uses the EFAST method to analyse the sensitivity of the HSS model parameters of soft soil strata. It is determined that the parameters that have the most significant impact on the surface settlement are the reference tangent modulus, rebound modulus, and effective cohesion. Then, XGBoost’s fast calculation speed and high precision of SSA inversion are used to inverse and optimize the parameters with high sensitivity. Finally, according to the parameters of the soft soil layer obtained from inversion and measured data, the settlement deformation and safety behaviour of existing buildings are analysed. Combined with the actual shield tunnel project in a city along a river, the inversion calculation shows that the overall average error of the transverse monitoring section is 1.04 mm, and the average maximum error of each monitoring point in the overall shield process is 2.87 mm. The prediction effect is significantly improved compared with the original parameters. The accuracy of the inversion of soil layer parameters is verified from the perspective of time and space. The average settlement of the river embankment foundation is 2.5 mm. Compared with the original parameter data, the prediction results have been greatly improved, and the settlement deformation results are more consistent with the measured data.

An in vitro comparison of mixed-reality navigation to traditional freehand and patient-specific instrumentation techniques for glenoid guide pin insertion during shoulder arthroplasty

BackgroundAccurate insertion of the glenoid guide pin in shoulder arthroplasty (RSA) is important for obtaining optimized glenoid component position and orientation. The objective of this study was to evaluate and compare the accuracy of three glenoid guide pin insertion techniques: 1) traditional software planning using freehand guide pin insertion (freehand), 2) guide pin insertion utilizing patient-specific instrumentation (PSI), and 3) using a mixed reality navigation (MR-NAV) system. MethodsTwenty (20) computer tomography (CT) scans were obtained from patients exhibiting glenoid erosion patterns according to the Walch and Favard classifications. Cases were planned using validated three-dimensional (3D) preoperative planning software. The CT data was then used to 3D print triplicate plastic models of each glenoid to evaluate the three guide pin insertion techniques. The first technique employed traditional software planning with freehand guide pin insertion. The second method used preoperatively planned PSI guides, while the third utilized a MR-NAV system, which provided real-time holographic guidance during guide pin insertion. Once all guide pins had been inserted into the models, an independent optical tracking system and custom digitization device was used to quantify the position and orientation of each guide pin relative to the glenoid. The outcomes for this study included the absolute mean error in guide pin inclination, version, and entry point relative to the preoperative plan. The absolute Total Global Error was also assessed, which was defined as the sum of the absolute guide pin orientation and position error relative to the preoperative plan. ResultsNo statistically significant differences between MR-NAV and PSI were found for the inclination error (2±1° versus 2±1°; P=0.056), version error (1±1° versus 1±1°; P=1.000), and Total Global Error (5±1 [mm+deg] versus 5±1 [mm+deg], P=1.000), respectively. The freehand technique produced significantly greater error than MR-NAV and PSI for inclination (5±3°, P≤0.017), version (4±3°, P≤0.032) and Total Global Error (8±3 [mm+deg], P<0.001). No statistically significant differences in the entry point error were observed between all guide pin insertion methods (P≥0.058). DiscussionThese results demonstrate that the precision and accuracy of MR-NAV is comparable to PSI and superior to a freehand technique for glenoid guide pin insertion in-vitro. Further study is needed to compare the accuracy of these techniques intra-operatively, in addition to assessing a potential learning curve between surgeons of varying experience with the MR-NAV system.

Accuracy of Boltless Frame-Based Stereo-Electroencephalography Electrode Implantation.

Boltless implantation of stereo-electroencephalography electrode is a useful alternative especially when anchor bolt is not available such as in country with limited resources or is less appropriate such as placement in patients with thin skull or at the occiput area, despite some drawbacks including potential dislodgement. While the accuracy of implantation using anchor bolt is well-studied, data on boltless implantation remain scarce. This study aimed to reveal the accuracy, permissible error for actual placement of electrodes within the grey matter, and delayed electrode dislodgement in boltless implantation. A total of 120 electrodes were implanted in 15 patients using a Leksell Stereotactic G Frame with each electrode fixed on the scalp using sutures. Target point error was defined as the Euclidean distance between the planned target and the electrode tip on immediate postimplantation computed tomography. Similarly, delayed dislodgement was defined as the Euclidean distance between the electrode tips on immediate postimplantation computed tomography and delayed MRI. The factors affecting accuracy were evaluated using multiple linear regression. The permissible error was defined as the largest target point error that allows the maximum number of planned gray matter electrode contacts to be actually placed within the gray matter as intended. The median (IQR) target point error was 2.6 (1.7-3.5) mm, and the permissible error was 3.2 mm. The delayed dislodgement, with a median (IQR) of 2.2 (1.4-3.3) mm, was dependent on temporal muscle penetration (P = 5.0 × 10-4), scalp thickness (P < 5.1 × 10-3), and insertion angle (P = 3.4 × 10-3). Boltless implantation of stereo-electroencephalography electrode offers an accuracy comparable to those using anchor bolt. During the planning of boltless implantation, target points should be placed within 3.2 mm from the gray-white matter junction and a possible delayed dislodgement of 2.2 mm should be considered.

A method for accuracy of placement analysis on radiolucent polyether-ether-keton facial implants: A case series

Facial asymmetry is defined as a bilateral difference between facial components. Correction, often desired by the patient, can be performed with the aim of bone born patient-specific solid implants designed using 3D CAD software. This treatment is embedded in the daily practice of today's healthcare. However, an analysis of the implant's accuracy of placement has not been reported. This case series describes the accuracy analysis of bone born aesthetic facial implants manufactured out of polyether-ether-ketone (PEEK). The accuracy analysis was based on postoperative (cone beam) computed tomography ((CB)CT) data and preoperative 3D planning. The analysis showed a median entry point error of 0.7 mm (min: 0.1, max: 3.3, interquartile range: 0.78). The median maximal orientation error was 5.5° (min: 0.1, max: 36.8, interquartile range: 7.13). Both parameters showed an excellent intraobserver and interobserver agreement with an ICC above 0.84. The described cases show that the analysis method is an objective approach for determining the accuracy of PSI placement and indicates that these implants can be placed accurately on the osseous face.

A method for determining the actual coordinate system of spiral bevel gears by the measured positions of eight tooth surfaces

Abstract The coordinate system for the measurement of spiral bevel gear (SBG) by gear measuring center (GMC) is discussed and the misalignment between the actual coordinate system and the theoretical coordinate system of the SBG is analyzed. The misalignment can cause measurement errors and cannot be neglected in SBG precision measurements. To calculate the misalignment and accurately establish the actual coordinate system, a novel re-fitting method based on the measured positions of eight tooth surfaces is proposed. First, eight sets of gear axis direction vectors and original points can be obtained from the measurement of eight tooth surfaces, by the single tooth surface matching method. Second, the actual axis vector and the original point of the gear can be obtained, by re-fitting those eight sets of gear axis vectors and original points. Then, the actual coordinate system can be established by the actual gear axis direction vector and the original point. The least square method is used to calculate the original point, and a basis vector fitting method is proposed to solve the gear axis direction vector. Finally, two groups of preset coordinate system deviation tests are carried out to verify the accuracy and feasibility of the fitting algorithm. The results show that the fitting error of the original point is far less than 1 × 10 − 1 μ m and the fitting error of the angle is far less than 1 × 10 − 7 degree, which are superior to the design accuracy of the GMC. Therefore, the proposed algorithm can be applied to improve the measurement accuracy of current GMC.

High-precision all-in-one dual robotic arm strategy in oral implant surgery

IntroductionDental implantation has emerged as an efficient substitute for missing teeth, which is essential for restoring oral function and aesthetics. Compared to traditional denture repair approaches, dental implants offer better stability and sustainability. The position, angle, and depth of dental implants are crucial factors for their long-term success and necessitate high-precision operation and technical support.MethodWe propose an integrated dual-arm high-precision oral implant surgery navigation positioning system and a corresponding control strategy. Compared with traditional implant robots, the integrated dual-arm design greatly shortens the preparation time before surgery and simplifies the operation process. We propose a novel control flow and module for the proposed structure, including an Occluded Target Tracking Module (OTTM) for occlusion tracking, a Planting Plan Development Module (PPDM) for generating implant plans, and a Path Formulation Module (PFM) for controlling the movement path of the two robot arms.ResultUnder the coordinated control of the aforementioned modules, the robot achieved excellent accuracy in clinical trials. The average angular error and entry point error for five patients who underwent implant surgery using the proposed robot were 2.1° and 0.39 mm, respectively.ConclusionIn essence, our study introduces an integrated dual-arm high-precision navigation system for oral implant surgery, resolving issues like lengthy preoperative preparation and static surgical planning. Clinical results confirm its efficacy, emphasizing its accuracy and precision in guiding oral implant procedures.

Abnormal User Identification in Substation Area Based on Restricted Ridge Regression Model

The rapid development of the smart grid affects every aspect of people’s lives at all times. Among them, the identification accuracy of power metering anomalies is a serious problem that needs to be solved urgently. To further improve the identification accuracy of power metering anomalies, a user identification method for substation area anomalies is designed based on the restricted ridge regression model. First, a clustering algorithm is designed using high-frequency voltage data of users in the substation area to conduct phase separation and terminal branch identification for each user metering point in the substation area. Then establish a substation area energy conservation model equation set, and merge the end branch users based on genetic algorithm clustering results, thereby reducing the unknown quantity dimension. Then, a line resistance-constrained condition is constructed using the topological structure, and the model equation set is solved using quadratic programming based on the constrained ridge regression method to obtain the measurement point error results of users in the substation area. The experimental results show that this recognition method’s clustering effect and recognition accuracy are high, with a recognition rate of 98.21%. It can be seen that this abnormal user identification method has application value.

Beyond Traditional Training: Exploring the Benefits of Virtual Reality Simulator in Lumbar Pedicle Screw Insertion - A Randomized Controlled Trial

This study compares the effectiveness of virtual reality simulators (VRS) and a saw bone model for learning lumbar pedicle screw insertion (LPSI) in neurosurgery. A single-center, cross-sectional, randomized controlled laboratory investigation was conducted involving residents and fellows from a tertiary care referral hospital. Participants were divided into two groups (A and B). Group A performed 3 LPSI tasks: the first on a saw bone model, the second on VRS, and the third on another saw bone model. Group B completed 2 LPSI tasks: the first on a saw bone model and the second on another saw bone model. The accuracy of LPSI was evaluated through noncontrast computed tomography scans for the saw bone models, while the in-built application of VRS was utilized to check for accuracy of screw placement using the simulator. The study included 38 participants (19 in each group). Group A participants showed reduced mean entry point error (0.11mm, P 0.024), increased mean purchase length (4.66cm, P 0.007), and no cortical breaches (P 0.031) when placing the second saw bone model screw. Similar improvements were observed among group A participants in PGY 1-3 while placing the second saw bone model screws. Virtual reality simulators (VRS) prove to be an invaluable tool for teaching complex neurosurgical skills, such as LPSI, to trainees. This technology investment can enhance the learning curve while maintaining patient safety.

Banana Bunch Weight Estimation and Stalk Central Point Localization in Banana Orchards Based on RGB-D Images

Precise detection and localization are prerequisites for intelligent harvesting, while fruit size and weight estimation are key to intelligent orchard management. In commercial banana orchards, it is necessary to manage the growth and weight of banana bunches so that they can be harvested in time and prepared for transportation according to their different maturity levels. In this study, in order to reduce management costs and labor dependence, and obtain non-destructive weight estimation, we propose a method for localizing and estimating banana bunches using RGB-D images. First, the color image is detected through the YOLO-Banana neural network to obtain two-dimensional information about the banana bunches and stalks. Then, the three-dimensional coordinates of the central point of the banana stalk are calculated according to the depth information, and the banana bunch size is obtained based on the depth information of the central point. Finally, the effective pixel ratio of the banana bunch is presented, and the banana bunch weight estimation model is statistically analyzed. Thus, the weight estimation of the banana bunch is obtained through the bunch size and the effective pixel ratio. The R2 value between the estimated weight and the actual measured value is 0.8947, the RMSE is 1.4102 kg, and the average localization error of the central point of the banana stalk is 22.875 mm. The results show that the proposed method can provide bunch size and weight estimation for the intelligent management of banana orchards, along with localization information for banana-harvesting robots.

An Adaptive Cubature Kalman Filter Based on Resampling-Free Sigma-Point Update Framework and Improved Empirical Mode Decomposition for INS/CNS Navigation

For the degradation of the filtering performance of the INS/CNS navigation system under measurement noise uncertainty, an adaptive cubature Kalman filter (CKF) is proposed based on improved empirical mode decomposition (EMD) and a resampling-free sigma-point update framework (RSUF). The proposed algorithm innovatively integrates improved EMD and RSUF into CKF to estimate measurement noise covariance in real-time. Specifically, the improved EMD is used to reconstruct measurement noise, and the exponential decay weighting method is introduced to emphasize the use of new measurement noise while gradually discarding older data to estimate the measurement noise covariance. The estimated measurement noise covariance is then imported into RSUF to directly construct the posterior cubature points without a resampling step. Since the measurement noise covariance is updated in real-time and the prediction cubature points error is directly transformed to the posterior cubature points error, the proposed algorithm is less sensitive to the measurement noise uncertainty. The proposed algorithm is verified by simulations conducted on the INS/CNS-integrated navigation system. The experimental results indicate that the proposed algorithm achieves better performance for attitude angle.

Development of Augmented Reality Vision for Osteosynthesis Using a 3D Camera.

We developed a 3D camera system to track motion in a surgical field. This system has the potential to introduce augmented reality (AR) systems non-invasively, eliminating the need for the invasive AR markers conventionally required. The present study was performed to verify the real-time tracking accuracy of this system, assess the feasibility of integrating this system into the surgical workflow, and establish its potential to enhance the accuracy and efficiency of orthopedic procedures. To evaluate the accuracy of AR technology using a 3D camera, a forearm bone model was created. The forearm model was depicted using a 3D camera, and its accuracy was verified in terms of the positional relationship with a 3D bone model created from previously imaged CT data. Images of the surgical field (capturing the actual forearm) were taken and saved in nine poses by rotating the forearm from pronation to supination. The alignment of the reference points was computed at the three points of CT versus the three points of the 3D camera, yielding a 3D rotation matrix representing the positional relationship. In the original system, a stereo vision-based 3D camera, with a depth image resolution of 1280×720 pixels, 30 frames per second, and a lens field of view of 64 specifications, with a baseline of 3 cm, capable of optimally acquiring real-time 3D data at a distance of 40-60 cm from the subject was used. In the modified system, the following modifications were made to improve tracking performance: (1) color filter processing was changed from HSV to RGB, (2) positional detection accuracy was modified with supporting marker sizes of 8 mm in diameter, and (3) the detection of marker positions was stabilized by calculating the marker position for each frame. Tracking accuracy was examined with the original system and modified system for the following parameters: differences in the rotation matrix, maximum and minimum inter-reference point errors between CT-based and camera-based 3D data, and the average error for the three reference points. In the original system, the average difference in rotation matrices was 5.51±2.68 mm. Average minimum and maximum errors were 1.10±0.61 and 15.53±12.51 mm, respectively. The average error of reference points was 6.26±4.49 mm. In the modified system, the average difference in rotation matrices was 4.22±1.73 mm. Average minimum and maximum errors were 0.79±0.49 and 1.94±0.87 mm, respectively. The average error of reference points was 1.41±0.58 mm. In the original system, once tracking failed, it was difficult to recover tracking accuracy. This resulted in a large maximum error in supination positions. These issues were resolved by the modified system. Significant improvements were achieved in maximum errors and average errors using the modified system (P<0.05). AR technology using a 3D camera was developed. This system allows direct comparisons of 3D data from preoperative CT scans with 3D data acquired from the surgical field using a 3D camera. This method has the advantage of introducing AR into the surgical field without invasive markers.

Wrongful birth and wrongful life lawsuits in obstetrics and gynecology

Developments in preconception and prenatal technologies have led to undeniable advances in how healthcare providers screen and treat patients. Despite these advances, at any point errors can occur leading to misdiagnosis or a missed diagnosis. In some instances, the missed information can lead to the birth of a child with health issues where short of the error, the decision to avoid conception or terminate the pregnancy might have been made. When these lapses unfold, there exists the potential for a wrongful birth or wrongful life lawsuit to ensue. While these two actions are based on the same set of events, they are distinct legal claims with varying degrees of judicial permissibility. Global legal acceptability of wrongful birth and life lawsuits tends to resemble patterns in the United States. Analyzing prior wrongful birth and wrongful life claims can reveal common trends in events leading to these types of lawsuits, as well as an understanding of their potential outcomes. A familiarity with wrongful birth and wrongful life lawsuits demonstrates how these cases are a unique from other forms of prenatal or birth injury tort lawsuits and can provide insights to common shortcomings in clinical practice. Applying these lessons to clinical practice highlights key approaches towards limiting the risk of certain errors leading to wrongful birth and wrongful life lawsuits, with the goal of healthcare providers offering high quality healthcare.

Phantom and in vivo accuracy of frameless optical navigation in stereotactic laser interstitial thermal therapy.

Targeting accuracy presents a key factor in achieving maximal safe ablation in laser interstitial thermal therapy (LITT). The VarioGuide system has proven precise for brain biopsies, but data showing its accuracy in combination with LITT are limited. The aim of this study was to determine the phantom and in vivo accuracy of LITT probe placement using the VarioGuide system and to evaluate the effect of targeting error on maximum possible ablation volume. Stereotactic LITT probe placement was performed using the VarioGuide system in 3 phantom skulls. The same system was used in 10 patients treated with LITT, for which data were retrospectively analyzed. Target point error (TPE), target depth deviation (TDD), target lateral deviation (TLD), and angular deviation (AD) were derived from intraprocedural MRI scans of both the phantom and in vivo trajectories. In vivo, the effect of targeting error on the maximum reachable ablation was calculated as the difference between the planned maximal achievable tumor ablation (PTA) and the actual maximal achievable tumor ablation (ATA). In total, 24 phantom and 16 in vivo trajectories were analyzed. In the phantom setting, the median TPE was 3.3 mm and median AD was 1.9°. Targeting accuracy significantly decreased for longer trajectories and those less perpendicular to the skull. In patients, the authors observed a comparable median TPE of 4.0 mm but significantly higher AD of 3.2°. In vivo, targeting inaccuracy resulted in a median decrease in maximum achievable ablation volume of 6% as compared to the planned trajectory. The authors' study indicates that utilizing the VarioGuide system in combination with LITT yields an average targeting error as large as 4 mm, which was smaller for shorter and straighter trajectories. In patients, targeting inaccuracy resulted in a median 6% decrease of the planned tumor ablation volume. These are important factors that should be considered in optimal case planning and patient selection in LITT.

The 3-Dimensional Intelligent Structured Light Technique: A New Registration Method in Stereotactic Neurosurgery.

Surface-based facial scanning registration emerged as an essential registration method in the robot-assisted neuronavigation surgery, providing a marker-free way to align a patient's facial surface with the imaging data. The 3-dimensional (3D) structured light was developed as an advanced registration method based on surface-based facial scanning registration. We aspire to introduce the 3D structured light as a new registration method in the procedure of the robot-assisted neurosurgery and assess the accuracy, efficiency, and safety of this method by analyzing the relative operative results. We analyzed the results of 47 patients who underwent Ommaya reservoir implantation (n = 17) and stereotactic biopsy (n = 30) assisted by 3D structured light at our hospital from January 2022 to May 2023. The accuracy and additional operative results were analyzed. For the Ommaya reservoir implantation, the target point error was 3.2 ± 2.2 mm and the entry point error was 3.3 ± 2.4 mm, while the operation duration was 35.8 ± 8.3 minutes. For the stereotactic biopsy, the target point error was 2.3 ± 1.3 mm and the entry point error was 2.7 ± 1.2 mm, while the operation duration was 24.5 ± 6.3 minutes. The 3D structured light technique reduces the patients' discomfort and offers the advantage of a simpler procedure, which can improve the clinical efficiency with the sufficient accuracy and safety to meet the clinical requirements of the puncture and navigation.

Validation and Safety Profile of a Novel, Noninvasive Fiducial Attachment for Stereotactic Robotic-Guided Stereoelectroencephalography: A Case Series.

We developed, tested, and validated a novel, noninvasive, Leksell G frame-based fiducial attachment, for use in stereotactic registration for stereoelectroencephalography (sEEG). Use of the device increased the number of fixed reference points available for registration, while obviating the need for additional scalp incisions. We report here on our experience and safety profile of using the device. We collected registration data using the fiducial device across 25 adult and pediatric patients with epilepsy consecutively undergoing robotic-guided sEEG for invasive epilepsy monitoring, treated between May 2022 and July 2023. ROSA One Brain was used for trajectory planning and electrode implantation. Postoperative clinical and radiographic data were computed and quantified, including mean registration error for all patients. Entry point, target point (TP), and angular errors were measured. Descriptive statistics and correlation coefficients for error were calculated. Twenty-five patients underwent robotic-guided sEEG implantation (11 patients, bilateral; 10 patients, left unilateral; 4 patients, right). The mean number of electrodes per patient was 18 ± 3. The average mean registration error was 0.77 ± 0.11 mm. All patients were implanted with Ad-Tech depth electrodes. No clinically relevant complications were reported. Analysis of trajectory error was performed on 446 electrodes. The median entry point error was 1.03 mm (IQR 0.69-1.54). The median TP error was 2.26 mm (IQR 1.63-2.93). The mean angular error was 0.03 radians (IQR 0.02-0.05). There was no significant correlation between root mean square error and lead error. Root mean square error did not appreciably change over time, nor were there any significant changes in average angular, entry point, or TP error metrics. A novel, noninvasive, Leksell G frame-based fiducial attachment was developed, tested, and validated, facilitating O-arm-based stereotactic registration for sEEG. This simple innovation maintained an excellent accuracy and safety profile for sEEG procedures in epilepsy patients, with the added advantages of providing additional reference points for stereotactic registration, without requiring additional scalp incisions.

Geo-guided deep learning for spatial downscaling of solute transport in heterogeneous porous media

Resolving solute transport in heterogeneous porous media is a complex task, because of the sparse experimental data and the high computational cost of numerical simulations. This work proposes a unique two-stage deep learning architecture comprising a dual-branch autoencoder and a geo-guided super-resolution generative adversarial network (Gg-SRGAN) to address this dual challenge. The dual-branch autoencoder addresses the issue of sparsity by constructing a continuous, but coarse representation of concentration and pressure profiles from a sparse, discontinuous profile with up to 85% missing data points. The Gg-SRGAN is then employed to generate a finer representation of field variables from the outputs generated by the dual-branch autoencoder (i.e., downscaling). We train and test our framework using six solute transport cases with varying levels of heterogeneity and compare the results with standalone methods, namely the vanilla autoencoder and vanilla SRGAN, in addition to ground truth profiles generated by the finite element method (FEM). The comparisons are performed based on several statistical metrics, such as absolute point error (APE), mean squared error (MSE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and learned perceptual image patch similarity (LPIPS). The first four cases are used for training, evaluation, and testing. The last two cases are utilized for blind testing to determine the generalizability of the framework. Our results show that the dual-branch autoencoder outperforms the vanilla autoencoder, and the Gg-SRGAN outperforms the SRGAN during both the training and evaluation phases. Moreover, the proposed framework can successfully construct the fine representation of concentration profiles, compared to FEM, using the coarse representation of the pressure, concentration, and domain permeability fields. When tested using the two blind test cases, the proposed dual-branch autoencoder and Gg-SRGAN exhibit superior performance compared to their counterparts in terms of all evaluation metrics.

Analysis of learning curves in predictive modeling using exponential curve fitting with an asymptotic approach.

The existence of large volumes of data has considerably alleviated concerns regarding the availability of sufficient data instances for machine learning experiments. Nevertheless, in certain contexts, addressing limited data availability may demand distinct strategies and efforts. Analyzing COVID-19 predictions at pandemic beginning emerged a question: how much data is needed to make reliable predictions? When does the volume of data provide a better understanding of the disease's evolution and, in turn, offer reliable forecasts? Given these questions, the objective of this study is to analyze learning curves obtained from predicting the incidence of COVID-19 in Brazilian States using ARIMA models with limited available data. To fulfill the objective, a retrospective exploration of COVID-19 incidence across the Brazilian States was performed. After the data acquisition and modeling, the model errors were assessed by employing a learning curve analysis. The asymptotic exponential curve fitting enabled the evaluation of the errors in different points, reflecting the increased available data over time. For a comprehensive understanding of the results at distinct stages of the time evolution, the average derivative of the curves and the equilibrium points were calculated, aimed to identify the convergence of the ARIMA models to a stable pattern. We observed differences in average derivatives and equilibrium values among the multiple samples. While both metrics ultimately confirmed the convergence to stability, the equilibrium points were more sensitive to changes in the models' accuracy and provided a better indication of the learning progress. The proposed method for constructing learning curves enabled consistent monitoring of prediction results, providing evidence-based understandings required for informed decision-making.

Machine vision-based non-destructive dissolution prediction of meloxicam-containing tablets

Machine vision systems have emerged for quality assessment of solid dosage forms in the pharmaceutical industry. These can offer a versatile tool for continuous manufacturing while supporting the framework of process analytical technology, quality-by-design, and real-time release testing. The aim of this work is to develop a digital UV/VIS imaging-based system for predicting the in vitro dissolution of meloxicam-containing tablets. The alteration of the dissolution profiles of the samples required different levels of the critical process parameters, including compression force, particle size and content of the API. These process parameters were predicted non-destructively by multivariate analysis of UV/VIS images taken from the tablets. The dissolution profile prediction was also executed using solely the image data and applying artificial neural networks. The prediction error (RMSE) of the dissolution profile points was less than 5%. The alteration of the API content directly affected the maximum concentrations observed at the end of the dissolution tests. This parameter was predicted with a relative error of less than 10% by PLS models that are based on the color components of UV and VIS images. In conclusion, this paper presents a modern, non-destructive PAT solution for real-time testing of the dissolution of tablets.

Temperature-Automated Calibration Methods for a Large-Area Blackbody Radiation Source.

High-precision temperature control of large-area blackbodies has a pivotal role in temperature calibration and thermal imaging correction. Meanwhile, it is necessary to correct the temperature difference between the radiating (surface of use) and back surfaces (where the temperature sensor is installed) of the blackbody during the testing phase. Moreover, large-area blackbodies are usually composed of multiple temperature control channels, and manual correction in this scenario is error-prone and inefficient. At present, there is no method that can achieve temperature-automated calibration for a large-area blackbody radiation source. Therefore, this article is dedicated to achieving temperature-automated calibration for a large-area blackbody radiation source. First, utilizing two calibrated infrared thermometers, the optimal temperature measurement location was determined using a focusing algorithm. Then, a three-axis movement system was used to obtain the true temperature at the same measurement location on a large-area blackbody surface from different channels. This temperature was subtracted from the blackbody's back surface. The temperature difference was calculated employing a weighted algorithm to derive the parameters for calibration. Finally, regarding experimental verification, the consistency error of the temperature measurement point was reduced by 85.4%, the temperature uniformity of the surface source was improved by 40.4%, and the average temperature measurement deviation decreased by 43.8%. In addition, this system demonstrated the characteristics of strong environmental adaptability that was able to perform temperature calibration under the working conditions of a blackbody surface temperature from 100 K to 573 K, which decreased the calibration time by 9.82 times.