Inter-Rater Reliability of EyeSpy Mobile for Pediatric Visual Acuity Assessments by Parent Volunteers

Purpose

To assess the inter-rater test reliability of the EyeSpy Mobile visual acuity smartphone algorithm when administered to children by eye professionals and parent volunteers.

Patients and Methods

Visual acuity test-retest results were analyzed for 106 children assigned to one of three different screenings: (1) An eye technician and pediatric ophthalmologist using their typical visual acuity testing method on a M&S computer; (2) An eye technician and pediatric ophthalmologist using EyeSpy Mobile; (3) An eye technician and parent volunteer using EyeSpy Mobile.

Results

All three phases demonstrated a strong agreement between the two testers, with mean test-retest equivalency results within 0.05 logMAR (2.5 letters, 90% CI). Whether testing using their typical technique on an M&S computer or using EyeSpy Mobile, eye professionals obtained statistically closer mean test-retest results than parent volunteers by 1 letter, with equivalency results within 0.03 logMAR (1.5 letters, 90% CI). Conversely, the number of retests within 2 vision lines was statistically greater when EyeSpy mobile was used by parents as compared to eye professional's customary technique on the M&S computer.

Conclusion

EyeSpy Mobile provides clinically useful visual acuity test-retest results even when used by first-time parent volunteers. Adaptive visual acuity algorithms have the potential to improve reliability, lessen training requirements, and expand the number of vision screening volunteers in community settings.

A three-dimensional adaptive rational interpolation algorithm for removing TMS-EEG pulse artifacts

Objective.Transcranial magnetic stimulation in combination with electroencephalography (TMS-EEG) has been widely used to study the reactivity and connectivity of brain regions. In order to efficiently and fast solve the pulse artifacts problem caused by TMS electromagnetic pulses, a three-dimensional adaptive rational quadratic Hermite interpolation algorithm is proposed.Approach.Firstly, a three-dimensional signal matrix is obtained by a signal recombination algorithm, where the removed window is automatically obtained by a derivative threshold. Secondly, the adaptive rational quartic Hermite interpolation algorithm is used to interpolate the removed window. Finally, the performance of the algorithm is verified using simulated and public database data.Main results.The simulation results show that the proposed algorithm improves the SNR by 23.88%-47.60%, reduces the RMSE by 46.52%-81.11%, reduces the average MAE by 47.83%-58.33%, and reduces the time consumption of the proposed algorithm by 45.90% compared with the piecewise cubic Hermite interpolation algorithm.Significance.Therefore, TMS-EEG pulse artifacts can be removed effectively and quickly with the proposed algorithm.

Tracking online low-rank approximations of higher-order incomplete streaming tensors

In this paper, we propose two new provable algorithms for tracking online low-rank approximations of high-order streaming tensors with missing data. The first algorithm, dubbed adaptive Tucker decomposition (ATD), minimizes a weighted recursive least-squares cost function to obtain the tensor factors and the core tensor in an efficient way, thanks to an alternating minimization framework and a randomized sketching technique. Under the canonical polyadic (CP) model, the second algorithm, called ACP, is developed as a variant of ATD when the core tensor is imposed to be identity. Both algorithms are low-complexity tensor trackers that have fast convergence and low memory storage requirements. A unified convergence analysis is presented for ATD and ACP to justify their performance. Experiments indicate that the two proposed algorithms are capable of streaming tensor decomposition with competitive performance with respect to estimation accuracy and runtime on both synthetic and real data.

An Adaptive Algorithm and Additively Manufactured Punch Used to Form Aluminum Sheet Metal Parts

Self-adaptive mechanisms are gaining momentum in industrial processes. It is understandable that as the complexity increases, the human work must be augmented. Considering this, the authors have developed one such solution for the punch-forming process, using additive manufacturing, i.e., a 3D-printed punch, to draw into shape 6061-T6 aluminum sheets. This paper aims to highlight the topological study used to optimize the punch form shape, the methodology of the 3D printing process, and the material used. For the adaptive algorithm, a complex Python-to-C++ bridge was created. It was necessary as the script has computer vision (used for calculating stroke and speed), punch force, and hydraulic pressure measurement capabilities. The algorithm uses the input data to control its subsequent actions. Two approaches are used in this experimental paper, a pre-programmed direction and an adaptive one, for comparison purposes. The results, namely the drawing radius and flange angle, were statistically analyzed using the ANOVA methodology for significance. The results indicate significant improvements when using the adaptive algorithm.

Achieving Adaptive Visual Multi-Object Tracking with Unscented Kalman Filter

As an essential part of intelligent monitoring, behavior recognition, automatic driving, and others, the challenge of multi-object tracking is still to ensure tracking accuracy and robustness, especially in complex occlusion environments. Aiming at the issues of the occlusion, background noise, and motion state violent change for multi-object in a complex scene, an improved DeepSORT algorithm based on YOLOv5 is proposed for multi-object tracking to enhance the speed and accuracy of tracking. Firstly, a general object motion model is devised, which is similar to the variable acceleration motion model, and a multi-object tracking framework with the general motion model is established. Then, the latest YOLOv5 algorithm, which has satisfactory detection accuracy, is utilized to obtain the object information as the input of multi-object tracking. An unscented Kalman filter (UKF) is proposed to estimate the motion state of multi-object to solve nonlinear errors. In addition, the adaptive factor is introduced to evaluate observation noise and detect abnormal observations so as to adaptively adjust the innovation covariance matrix. Finally, an improved DeepSORT algorithm for multi-object tracking is formed to promote robustness and accuracy. Extensive experiments are carried out on the MOT16 data set, and we compare the proposed algorithm with the DeepSORT algorithm. The results indicate that the speed and precision of the improved DeepSORT are increased by 4.75% and 2.30%, respectively. Especially in the MOT16 of the dynamic camera, the improved DeepSORT shows better performance.

An Adaptive User Tracking Algorithm Using Irregular Data Frames for Passive Fingerprint Positioning

Wi-Fi fingerprinting is the most popular indoor positioning method today, representing received signal strength (RSS) values as vector-type fingerprints. Passive fingerprinting, unlike the active fingerprinting method, has the advantage of being able to track location without user participation by utilizing the signals that are naturally emitted from the user's smartphone. However, since signals are generated depending on the user's network usage patterns, there is a problem in that data are irregularly collected according to the patterns. Therefore, this paper proposes an adaptive algorithm that shows stable tracking performances for fingerprints generated at irregular time intervals. The accuracy and stability of the proposed tracking method were verified by experiments conducted in three scenarios. Through the proposed method, it is expected that the stability of indoor positioning and the quality of location-based services will improve.

The Adaptive Olfactory Measure of Threshold (ArOMa-T): A rapid test of olfactory function

PURPOSE

Many widely-used psychophysical tests of olfaction have limitations that can create barriers to adoption outside research settings. For example, tests that measure the ability to identify odors may confound sensory performance with memory recall, verbal ability, and past experience with the odor. Conversely, threshold-based tests typically avoid these issues, but are labor intensive. Additionally, many commercially available olfactory tests are slow and may require a trained administrator, making them impractical for use in a short wellness visit or other broad clinical assessment.

METHODS

We tested the performance of the Adaptive Olfactory Measure of Threshold (ArOMa-T) -- a novel odor detection threshold test that employs an adaptive Bayesian algorithm paired with a disposable odor-delivery card -- in a non-clinical sample of individuals (n=534) at the 2021 Twins Day Festival in Twinsburg, OH.

RESULTS

Participants successfully completed the test in under 3 min with a false alarm rate of 9.6% and a test-retest reliability of 0.61. Odor detection thresholds differed by sex (~3.2-fold) and between the youngest and oldest age groups (~8.7-fold), consistent with prior work. In an exploratory analysis, we failed to observe evidence of detection threshold differences between participants who reported a history of COVID-19 and matched controls who did not. We also found evidence for broad-sense heritability of odor detection thresholds.

CONCLUSION

Together, these data indicate the ArOMa-T can determine odor detection thresholds. The ArOMa-T may be particularly valuable in clinical or field settings where rapid and portable assessment of olfactory function is needed.

The Adaptive Olfactory Measure of Threshold (ArOMa-T): a rapid test of olfactory function

Many widely used psychophysical olfactory tests have limitations that can create barriers to adoption. For example, tests that measure the ability to identify odors may confound sensory performance with memory recall, verbal ability, and prior experience with the odor. Conversely, classic threshold-based tests avoid these issues, but are labor intensive. Additionally, many commercially available tests are slow and may require a trained administrator, making them impractical for use in situations where time is at a premium or self-administration is required. We tested the performance of the Adaptive Olfactory Measure of Threshold (ArOMa-T)-a novel odor detection threshold test that employs an adaptive Bayesian algorithm paired with a disposable odorant delivery card-in a non-clinical sample of individuals (n = 534) at the 2021 Twins Day Festival in Twinsburg, OH. Participants successfully completed the test in under 3 min with a false alarm rate of 7.5% and a test-retest reliability of 0.61. Odor detection thresholds differed by sex (~3.2-fold lower for females) and age (~8.7-fold lower for the youngest versus the oldest age group), consistent with prior studies. In an exploratory analysis, we failed to observe evidence of detection threshold differences between participants who reported a history of COVID-19 and matched controls who did not. We also found evidence for broad-sense heritability of odor detection thresholds. Together, this study suggests the ArOMa-T can determine odor detection thresholds. Additional validation studies are needed to confirm the value of ArOMa-T in clinical or field settings where rapid and portable assessment of olfactory function is needed.

Illumination Adaptation in a Multi-Wavelength Opto-Electronic Patch Sensor

In capturing high-quality photoplethysmographic signals, it is crucial to ensure that appropriate illumination intensities are used. The purpose of the study was to deliver controlled illumination intensities for a multi-wavelength opto-electronic patch sensor that has four separate arrays each consisting of four light-emitting diodes (LEDs), the wavelength of the light generated by each array being different. The study achieved the following: (1) a linear constant current source LED driver incorporating series negative feedback using an integrated operational amplifier circuit; (2) the fitting of a linear regression equation to provide rapid determination of the LEDs driver voltage; and (3) an algorithm for the automatic adjustment of the output voltage to ensure suitable LED illumination. The data from a single centrally-located photo detector, which is capable of capturing all four channels of back-light in a time-multiplexed manner, were used to monitor heart rate and blood oxygen saturation. This paper provides circuitry for driving the LEDs and describes an adaptive algorithm implemented on a microcontroller unit that monitors the quality of the photo detector signals received in order to control each of the individual currents being supplied to the LED arrays. The study demonstrated that the operation of the new circuitry in its ability to adapt LED illumination to the strength of the signal received and the performance of the adaptive system was compared with that of a non-adaptive approach.

Path Planning of Pattern Transfer Based on Dual-Operator and a Dual-Population Ant Colony Algorithm for Digital Mask Projection Lithography

In the process of digital micromirror device (DMD) digital mask projection lithography, the lithography efficiency will be enhanced greatly by path planning of pattern transfer. This paper proposes a new dual operator and dual population ant colony (DODPACO) algorithm. Firstly, load operators and feedback operators are used to update the local and global pheromones in the white ant colony, and the feedback operator is used in the yellow ant colony. The concept of information entropy is used to regulate the number of yellow and white ant colonies adaptively. Secondly, take eight groups of large-scale data in TSPLIB as examples to compare with two classical ACO and six improved ACO algorithms; the results show that the DODPACO algorithm is superior in solving large-scale events in terms of solution quality and convergence speed. Thirdly, take PCB production as an example to verify the time saved after path planning; the DODPACO algorithm is used for path planning, which saves 34.3% of time compared with no path planning, and is about 1% shorter than the suboptimal algorithm. The DODPACO algorithm is applicable to the path planning of pattern transfer in DMD digital mask projection lithography and other digital mask lithography.

Distributed Learning Fractal Algorithm for Optimizing a Centralized Control Topology of Wireless Sensor Network Based on the Hilbert Curve L-System

Wireless sensor networks (WSNs) consist of a large number of small devices or nodes, called micro controller units (MCUs) and located in homes and/or offices, to be operated through the internet from anywhere, making these devices smarter and more efficient. Quality of service routing is one of the critical challenges in WSNs, especially in surveillance systems. To improve the efficiency of the network, in this article we proposes a distributed learning fractal algorithm (DFLA) to design the control topology of a wireless sensor network (WSN), whose nodes are the MCUs distributed in a physical space and which are connected to share parameters of the sensors such as concentrations of C O 2 , humidity, temperature within the space or adjustment of the intensity of light inside and outside the home or office. For this, we start defining the production rules of the L-systems to generate the Hilbert fractal, since these rules facilitate the generation of this fractal, which is a fill-space curve. Then, we model the optimization of a centralized control topology of WSNs and proposed a DFLA to find the best two nodes where a device can find the highly reliable link between these nodes. Thus, we propose a software defined network (SDN) with strong mobility since it can be reconfigured depending on the amount of nodes, also we employ a target coverage because distributed learning fractal algorithm (DLFA) only consider reliable links among devices. Finally, through laboratory tests and computer simulations, we demonstrate the effectiveness of our approach by means of a fractal routing in WSNs, by using a large amount of WSNs devices (from 16 to 64 sensors) for real time monitoring of different parameters, in order to make efficient WSNs and its application in a forthcoming Smart City.

Adaptive algorithm utilizing acceptance rate for eliminating noisy epochs in block-design functional near-infrared spectroscopy data: application to study in attention deficit/hyperactivity disorder children

Functional near-infrared spectroscopy (fNIRS) signals are prone to problems caused by motion artifacts and physiological noises. These noises unfortunately reduce the fNIRS sensitivity in detecting the evoked brain activation while increasing the risk of statistical error. In fNIRS measurements, the repetitive resting-stimulus cycle (so-called block-design analysis) is commonly adapted to increase the sample number. However, these blocks are often affected by noises. Therefore, we developed an adaptive algorithm to identify, reject, and select the noise-free and/or least noisy blocks in accordance with the preset acceptance rate. The main features of this algorithm are personalized evaluation for individual data and controlled rejection to maintain the sample number. Three typical noise criteria (sudden amplitude change, shifted baseline, and minimum intertrial correlation) were adopted. Depending on the quality of the dataset used, the algorithm may require some or all noise criteria with distinct parameters. Aiming for real applications in a pediatric study, we applied this algorithm to fNIRS datasets obtained from attention deficit/hyperactivity disorder (ADHD) children as had been studied previously. These datasets were divided for training and validation purposes. A validation process was done to examine the feasibility of the algorithm regardless of the types of datasets, including those obtained under sample population (ADHD or typical developing children), intervention (nonmedication and drug/placebo administration), and measurement (task paradigm) conditions. The algorithm was optimized so as to enhance reproducibility of previous inferences. The optimum algorithm design involved all criteria ordered sequentially (0.047 mM mm of amplitude change, 0.029    mM    mm / s of baseline slope, and 0.6 × interquartile range of outlier threshold for each criterion, respectively) and presented complete reproducibility in both training and validation datasets. Compared to the visual-based rejection as done in the previous studies, the algorithm achieved 71.8% rejection accuracy. This suggests that the algorithm has robustness and potential to substitute for visual artifact-detection.

Adaptive Enhancement of X-Band Marine Radar Imagery to Detect Oil Spill Segments

Oil spills generate a large cost in environmental and economic terms. Their identification plays an important role in oil-spill response. We propose an oil spill detection method with improved adaptive enhancement on X-band marine radar systems. The radar images used in this paper were acquired on 21 July 2010, from the teaching-training ship “YUKUN” of the Dalian Maritime University. According to the shape characteristic of co-channel interference, two convolutional filters are used to detect the location of the interference, followed by a mean filter to erase the interference. Small objects, such as bright speckles, are taken as a mask in the radar image and improved by the Fields-of-Experts model. The region marked by strong reflected signals from the sea’s surface is selected to identify oil spills. The selected region is subject to improved adaptive enhancement designed based on features of radar images. With the proposed adaptive enhancement technique, calculated oil spill detection is comparable to visual interpretation in accuracy.

Large-scale analysis of high-speed atomic force microscopy data sets using adaptive image processing

Modern high-speed atomic force microscopes generate significant quantities of data in a short amount of time. Each image in the sequence has to be processed quickly and accurately in order to obtain a true representation of the sample and its changes over time. This paper presents an automated, adaptive algorithm for the required processing of AFM images. The algorithm adaptively corrects for both common one-dimensional distortions as well as the most common two-dimensional distortions. This method uses an iterative thresholded processing algorithm for rapid and accurate separation of background and surface topography. This separation prevents artificial bias from topographic features and ensures the best possible coherence between the different images in a sequence. This method is equally applicable to all channels of AFM data, and can process images in seconds.

[Not Available]

Only very recently, Sayas [The validity of Johnson-Nédélec's BEM-FEM coupling on polygonal interfaces. SIAM J Numer Anal 2009;47:3451-63] proved that the Johnson-Nédélec one-equation approach from [On the coupling of boundary integral and finite element methods. Math Comput 1980;35:1063-79] provides a stable coupling of finite element method (FEM) and boundary element method (BEM). In our work, we now adapt the analytical results for different a posteriori error estimates developed for the symmetric FEM-BEM coupling to the Johnson-Nédélec coupling. More precisely, we analyze the weighted-residual error estimator, the two-level error estimator, and different versions of (h-h/2)-based error estimators. In numerical experiments, we use these estimators to steer h-adaptive algorithms, and compare the effectivity of the different approaches.

Spatial independent component analysis of functional MRI time-series: to what extent do results depend on the algorithm used?

Independent component analysis (ICA) has been successfully employed to decompose functional MRI (fMRI) time-series into sets of activation maps and associated time-courses. Several ICA algorithms have been proposed in the neural network literature. Applied to fMRI, these algorithms might lead to different spatial or temporal readouts of brain activation. We compared the two ICA algorithms that have been used so far for spatial ICA (sICA) of fMRI time-series: the Infomax (Bell and Sejnowski [1995]: Neural Comput 7:1004-1034) and the Fixed-Point (Hyvärinen [1999]: Adv Neural Inf Proc Syst 10:273-279) algorithms. We evaluated the Infomax- and Fixed Point-based sICA decompositions of simulated motor, and real motor and visual activation fMRI time-series using an ensemble of measures. Log-likelihood (McKeown et al. [1998]: Hum Brain Mapp 6:160-188) was used as a measure of how significantly the estimated independent sources fit the statistical structure of the data; receiver operating characteristics (ROC) and linear correlation analyses were used to evaluate the algorithms' accuracy of estimating the spatial layout and the temporal dynamics of simulated and real activations; cluster sizing calculations and an estimation of a residual gaussian noise term within the components were used to examine the anatomic structure of ICA components and for the assessment of noise reduction capabilities. Whereas both algorithms produced highly accurate results, the Fixed-Point outperformed the Infomax in terms of spatial and temporal accuracy as long as inferential statistics were employed as benchmarks. Conversely, the Infomax sICA was superior in terms of global estimation of the ICA model and noise reduction capabilities. Because of its adaptive nature, the Infomax approach appears to be better suited to investigate activation phenomena that are not predictable or adequately modelled by inferential techniques.