Improve Measurement Precision Research Articles

Pre-natal and early life lead exposure and childhood inhibitory control: an item response theory approach to improve measurement precision of inhibitory control

BackgroundNeurodevelopmental performance tasks are often separately analyzed, even when they tap into a similar construct. This may yield mixed findings for associations of an exposure-neurobehavioral outcome. We develop an item response theory (IRT) approach to integrate multiple task variables together to improve measurement precision of the underlying construct. We apply this approach to create an integrative measure of childhood inhibitory control, and study impacts of pre/post-natal lead exposure.MethodsUsing data from a prospective cohort based in Mexico (N = 533), we created an inhibitory control scale that integrates accuracy and reaction time information from four inhibitory control tasks (Go/NoGo Letter, Go/NoGo Neutral, Go/NoGo Happy, Delis-Kaplan Executive Function System (D-KEFS) Color-Word Interference Test, Condition 3). Using a generalized partial credit item response theory model, we estimated an inhibitory control index for each participant. We then assessed adjusted associations between umbilical cord blood and 4-year lead and childhood inhibitory control. We developed a resampling approach to incorporate error estimates from the inhibitory control variable to confirm the consistency of the lead-inhibitory control associations. We modeled time-varying associations of lead with each inhibitory control measure separately.ResultsParticipants had a median age of 9 years; 51.4% were males. Umbilical cord blood [-0.06 (95% CI: -0.11, -0.01)] and 4-year lead [-0.07 (95% CI: -0.12, -0.02)] were associated with inhibitory control index at 8–10 years. A resampling approach confirmed that 4-year lead was consistently associated with childhood inhibitory control index. Umbilical cord blood and 4-year lead were each associated with 3 out of 8 measures in separate models.ConclusionThis is the first application of IRT in environmental epidemiology to create a latent variable for inhibitory control that integrates accuracy and reaction time information from multiple, related tasks. This framework can be applied to other correlated neurobehavioral assessments or other phenotype data.

Guest Editorial: Quantum industry: Applications in quantum communication (Quantum.Tech Europe 2022)

AbstractQuantum technology harnesses the principles of quantum mechanics to accomplish tasks in a different way as compared to the classical technologies. This includes quantum computing, which uses qubits for parallel information processing, greatly enhancing computation speed and entanglement and empowering problem‐solving abilities. Quantum communication provides secure data transmission through quantum cryptography, while quantum sensing offers improved measurement precision, benefiting areas such as cryptography, material science, and pharmaceuticals. Additionally, the implementation and commercialisation of quantum technology involve transitioning theoretical quantum concepts into practical applications and marketable products. To achieve widespread adoption of quantum industry, significant research efforts are crucial among academia, industry, and government.

A multi-channel photoplethysmography array with contact-force regulation for tonoarteriographic imaging

Aim: Photoplethysmography (PPG) has gained widespread popularity as a non-invasive method for potential cuff-less blood pressure (BP) measurement in smart devices. However, the accuracy of PPG-based devices is often hindered by motion artifacts, site variability, and inconsistent contact force (CF). This study aims to investigate the influence of CF variations on PPG signals. Methods: To address these challenges, we present a novel approach involving a multi-channel PPG array integrated with CF regulation in the form of a wearable wristband. This platform enables the visualization of regional PPG/BP distribution while simultaneously monitoring CF. Moreover, our research explored the relationship between PPG waveform characteristics and CF during wrist extension. Results: The results of this study reveal that the PPG amplitude (PPGA) and the b/a ratios, computed from the second derivative peaks of the PPG AC pulse wave, exhibit inconsistency in reaction to CF variations. Notably, a shape correlation coefficient of 0.65069 between normalized PPG and flipped CF sheds light on how changes in posture affect PPG measurements. Conclusions: The proposed platform shows promise in mitigating the effects of CF and spatial positioning on PPG, thereby improving measurement precision and offering a novel approach to image tonoarteriographic (TAG) activities for continuous hypertension management.

The variability in nociceptive flexion reflex threshold measurement is mostly caused by probabilistic effects of the estimation algorithms: a simulation study

Objectives The nociceptive flexion reflex (NFR) and its threshold are frequently used to investigate spinal nociception in humans. Since this threshold (NFRT) is a probabilistic measure, specific algorithms are used for NFRT estimation based on the stochastic occurrence of reflexes at different stimulus intensities. We used a validated simulation model of the NFR to investigate the amount of NFRT measurement variability induced by different estimation algorithms in a steady setting of reduced external influences. Methods We simulated the behavior of different estimation algorithms in subjects with an artificially steady baseline NFRT variability (standard deviation: 0 mA) or low baseline NFRT variability (standard deviation: 0.156 mA), equaling a quiet experimental setting. The obtained data were analyzed for NFRT measurement variability caused by the algorithms compared to the baseline variability reflecting other physiological influences. Results The standard deviation of the NFRT estimated by the different algorithms ranged between 0.381 and 3.464 mA with 96.8% to 99.6% of the measurement variability attributed to the algorithm used. Out of the investigated algorithms the dynamic staircase algorithm was most precise. Conclusion The NFRT measurement variability observed during quiet and steady experimental sessions is mostly caused by the properties of the estimation algorithms, due to the probabilistic nature of the reflex occurrence. Our results give reference for choosing the optimal estimation algorithm to improve measurement precision.

Characterizing correlations in partial credit speech recognition scoring with beta-binomial distributions.

Partial credit scoring for speech recognition tasks can improve measurement precision. However, assessing the magnitude of this improvement with partial credit scoring is challenging because meaningful speech contains contextual cues, which create correlations between the probabilities of correctly identifying each token in a stimulus. Here, beta-binomial distributions were used to estimate recognition accuracy and intraclass correlation for phonemes in words and words in sentences in listeners with cochlear implants (N = 20). Estimates demonstrated substantial intraclass correlation in recognition accuracy within stimuli. These correlations were invariant across individuals. Intraclass correlations should be addressed in power analysis of partial credit scoring.

Administering selected subscales of patient-reported outcome questionnaires to reduce patient burden and increase relevance: a position statement on a modular approach

Patient-reported outcome (PRO) questionnaires considered in this paper contain multiple subscales, although not all subscales are equally relevant for administration in all target patient populations. A group of measurement experts, developers, license holders, and other scientific-, regulatory-, payer-, and patient-focused stakeholders participated in a panel to discuss the benefits and challenges of a modular approach, defined here as administering a subset of subscales out of a multi-scaled PRO measure. This paper supports the position that it is acceptable, and sometimes preferable, to take a modular approach when administering PRO questionnaires, provided that certain conditions have been met and a rigorous selection process performed. Based on the experiences and perspectives of all stakeholders, using a modular approach can reduce patient burden and increase the relevancy of the items administered, and thereby improve measurement precision and eliminate wasted data without sacrificing the scientific validity and utility of the instrument. The panelists agreed that implementing a modular approach is not expected to have a meaningful impact on item responses, subscale scores, variability, reliability, validity, and effect size estimates; however, collecting additional evidence for the impact of context may be desirable. It is also important to recognize that adequate rationale and evidence (e.g., of fit-for-purpose status and relevance to patients) and a robust consensus process that includes patient perspectives are required to inform selection of subscales, as in any other measurement circumstance, is expected. We believe that the considerations discussed within (content validity, administration context, and psychometric factors) are relevant across multiple therapeutic areas.

Preliminary comparisons between a point-of-care ketometer and reference method using Steller sea lion pup whole blood and plasma.

We evaluated the Precision Xtra™ ketometer as part of a larger study categorizing fasting status of free-ranging Steller sea lion (Eumetopias jubatus; SSL) pups which necessitated the identification of plasma β-hydroxybutyrate concentrations ([β-HBA]) around a threshold of <0.3 and ≥0.3mmol/l. Whole blood samples mixed with sodium heparin (NaHep) or ethylenediaminetetraacetic acid liquid anticoagulants were tested <10minutes after collection (n = 14; triplicate technical replicates). Plasma (stored at -80°C, NaHep, Thaw1) measured via our laboratory's Reference Assay (Sigma Aldrich, St. Louis, MO, Kit #MAK041) served as the standard [β-HBA] for ketometer comparisons. Our observed β-HBA range (0.0-1.6mmol/l), consistent with published [β-HBA] of free-ranging Otariid pups, represented the lower 20% of the ketometer's range (0.0-8.0mmol/l). The maximal coefficient of variation (%CV) of ketometer technical replicates was 9.1% (NaHep, whole blood). The majority of ketometer technical replicate sets (84%, including all matrices, anticoagulants and thawings) were identical (CV = 0%). We found linear relationships and agreement of ketometer [β-HBA] between whole blood preserved with different anticoagulants and between whole blood and plasma (Thaw1) measurements. The ketometer produced results with linearity to the Reference Assay for both whole blood and plasma (Thaw1). We identified a non-linear relationship between plasma at Thaw1 and Thaw2 (tested four months apart, NaHep), as only samples with higher SSL [β-HBA] decreased in concentration, and all others remained the same. With respect to categorizing SSL pup fasting in our larger study, the ketometer's %Accuracy, %Sensitivity and %Specificity for samples with Reference Assay β-HBA <0.2 and >0.4mmol/l were 100%. We adopted a modified procedure: plasma samples with mean ketometer concentrations ±0.1mmol/l of 0.3mmol/l β-HBA were re-evaluated using the Reference Assay, improving measurement precision from tenths (ketometer) to thousandths (assay) mmol/l. The Precision Xtra™ ketometer was valuable to our application over the range of [β-HBA] observed in SSL pup plasma and whole blood samples.

Enhancing the sensitivity of atom-interferometric inertial sensors using robust control

Atom-interferometric quantum sensors could revolutionize navigation, civil engineering, and Earth observation. However, operation in real-world environments is challenging due to external interference, platform noise, and constraints on size, weight, and power. Here we experimentally demonstrate that tailored light pulses designed using robust control techniques mitigate significant error sources in an atom-interferometric accelerometer. To mimic the effect of unpredictable lateral platform motion, we apply laser-intensity noise that varies up to 20% from pulse-to-pulse. Our robust control solution maintains performant sensing, while the utility of conventional pulses collapses. By measuring local gravity, we show that our robust pulses preserve interferometer scale factor and improve measurement precision by 10× in the presence of this noise. We further validate these enhancements by measuring applied accelerations over a 200 μg range up to 21× more precisely at the highest applied noise level. Our demonstration provides a pathway to improved atom-interferometric inertial sensing in real-world settings.

Use of technology-based assessments: a systematic review covering over 30 countries

ABSTRACT The widespread adoption of technology-based assessments during the COVID-19 pandemic has exacerbated concerns about the digital divide, given global disparities in digital access, connectivity, and coping strategies. This systematic review was intended to assess how the use of technology-based assessments has affected the education system’s functioning in pre-college education, compared to traditional assessments. It covered 34 countries from 34 full-text English sources in 2018–2022. A total of 12 assumptions were tested, corresponding to six hypotheses about the learning, educating, and management facets associated with the use of technology-based assessments. Our findings revealed mixed evidence about technology-based assessments in reducing cheating, enhancing learning, supporting monitoring, improving instruction, and reducing non-teaching workload. However, strong evidence supported the assumptions that technology-based assessments improve measurement precision, interpretability, engagement, interaction, and teacher-parent communication. Limited but positive evidence supported the assumptions that technology-based assessments may reduce the cost and time of test administration.

Assessing inhibitory control in kindergarten children: Validity of integrating response accuracy and response latency

Assessing inhibitory control in young children poses a challenge because of its rapid and non-linear development. This study examined the validity of integrating response accuracy and latency through a two-factor model, based on the data of 271 children who completed a computerized inhibitory control task. Although integrating response accuracy and latency slightly improved measurement precision, multigroup analyses of younger and older children showed inconsistent associations between response accuracy and latency if response latencies from incorrect responses were not excluded. A time-on-task analysis revealed that the extent of the accuracy gain by taking more time depended on the individual’s skill level. The validity of task performance as an indicator of inhibitory control was highest when response accuracy was the primary determinant of the inhibitory control score and response latency was only considered after the child had surpassed an accuracy threshold to further improve the score. These findings suggest that integrating response accuracy and latency into a single score should only be performed for children who can maintain high accuracy levels despite giving fast responses.

Modification of Optical System for Improving Measurement Precision and Sensitivity of a Laboratory-type X-ray Absorption Spectrometer and Measurement of Battery Material Metal Elements

Modification of Optical System for Improving Measurement Precision and Sensitivity of a Laboratory-type X-ray Absorption Spectrometer and Measurement of Battery Material Metal Elements

Quantifying structural racism in cohort studies to advance prospective evidence

Calls-to-action in health research have described a need to improve research on race, ethnicity, and structural racism. Well-established cohort studies typically lack access to novel structural and social determinants of health (SSDOH) or precise race and ethnicity categorization, contributing to a loss of rigor to conduct informative analyses and a gap in prospective evidence on the role of structural racism in health outcomes. We propose and implement methods that prospective cohort studies can use to begin to rectify this, using the Women's Health Initiative (WHI) cohort as a case study. To do so, we evaluated the quality, precision, and representativeness of race, ethnicity, and SSDOH data compared with the target US population and operationalized methods to quantify structural determinants in cohort studies. Harmonizing racial and ethnic categorization to the current standards set by the Office of Management and Budget improved measurement precision, aligned with published recommendations, disaggregated groups, decreased missing data, and decreased participants reporting “some other race”. Disaggregation revealed sub-group disparities in SSDOH, including a greater proportion of Black-Latina (35.2%) and AIAN-Latina (33.3%) WHI participants with income below the US median compared with White-Latina (42.5%) participants. We found similarities in the racial and ethnic patterning of SSDOH disparities between WHI and US women but less disparity overall in WHI. Despite higher individual-level advantage in WHI, racial disparities in neighborhood resources were similar to the US, reflecting structural racism. Median neighborhood income was comparable between Black WHI ($39,000) and US ($34,700) women. WHI SSDOH-associated outcomes may be generalizable on the basis of comparing across race and ethnicity but may quantitatively (but not qualitatively) underestimate US effect sizes. This paper takes steps towards data justice by implementing methods to make visible hidden health disparity groups and operationalizing structural-level determinants in prospective cohort studies, a first step to establishing causality in health disparities research.

Repeatability of brain phase-based magnetic resonance electric properties tomography methods and effect of compressed SENSE and RF shimming

Magnetic resonance electrical properties tomography (MREPT) is an emerging imaging modality to noninvasively measure tissue conductivity and permittivity. Implementation of MREPT in the clinic requires repeatable measurements at a short scan time and an appropriate protocol. The aim of this study was to investigate the repeatability of conductivity measurements using phase-based MREPT and the effects of compressed SENSE (CS), and RF shimming on the precision of conductivity measurements. Conductivity measurements using turbo spin echo (TSE) and three-dimensional balanced fast field echo (bFFE) with CS factors were repeatable. Conductivity measurement using bFFE phase showed smaller mean and variance that those measured by TSE. The conductivity measurements using bFFE showed minimal deviation with CS factors up to 8, with deviation increasing at CS factors > 8. Subcortical structures produced less consistent measurements than cortical parcellations at higher CS factors. RF shimming using full slice coverage 2D dual refocusing echo acquisition mode (DREAM) and full coverage 3D dual TR approaches further improved measurement precision. BFFE is a more optimal sequence than TSE for phase-based MREPT in brain. Depending on the area of the brain being measured, the scan can be safely accelerated with compressed SENSE without sacrifice of precision, offering the potential to employ MREPT in clinical research and applications. RF shimming with better field mapping further improves precision of the conductivity measures.

Accurate SiC MOSFET Chip Extraction Based on Parasitic Parameter Impact Compensation

With the growing use of SiC <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">mosfet</small> in power electronic equipment, it is meaningful to extract precise typical waveforms and characteristics for device manufacture, converter design, and operating evaluation. Many solutions have been proposed to improve measurement precision by equipment and essential techniques. However, the measured object should also be accurately identified as the existing measurement result consisting of the expected gate-source voltage ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">v</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GS</sub> ) and voltage drops on parasitic parameters between measuring points. This article, first, reveals the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">v</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GS</sub> measurement deviation of TO-247-4 and TO-247-3 devices during the switching process and crosstalk, based on the equivalent circuit model and simulation waveforms. According to the clarification of the measured object, an improved <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">v</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GS</sub> extraction method is proposed to eliminate this deviation. Voltage drops on parasitic parameters are compensated, with the help of high-precision equipment and the proven parasitic parameter extraction technique. Experiments are constructed to verify the analysis of measurement deviation and the availability of the proposed method. Comparison between various operating conditions emphasizes its necessity, especially in high-capacity and low-driving-resistor applications, contributing to characterization, operating condition monitoring, protection design, and measurement understanding of wide band-gap devices.

New Vegetation Monitoring Field Tools Improve Plot Demarcation Consistency

The Heartland Inventory and Monitoring Network uses a scaled plot design for vegetation monitoring. Redesign of aging field plot frames resulted in easier to use equipment. The new designs were aimed at improving measurement precision to reduce sources of nonsampling relocation error. Although sampling sites are permanently monumented at the corners, the center point of subplots is measured from these fixed corners. Increasing confidence in long-term monitoring data through consistent plot deployment will increase the value of that data. The 1 m2 circular plot uses flexible PVC tubing with a PVC pipe inserted and fastened for deployment or coiled for transportation. The redesigned 10 m2 plot delineation device for woodlands utilized a swivel mounted on a threaded rod. Upgrades to aging field gear improved field operations by increasing precision during plot deployment.

Nanocapillary sampling coupled to liquid chromatography mass spectrometry delivers single cell drug measurement and lipid fingerprints.

This work describes the development of a new approach to measure drug levels and lipid fingerprints in single living mammalian cells. Nanocapillary sampling is an approach that enables the selection and isolation of single living cells under microscope observation. Here, live single cell nanocapillary sampling is coupled to liquid chromatography for the first time. This allows molecular species to be separated prior to ionisation and improves measurement precision of drug analytes. The efficiency of transferring analytes from the sampling capillary into a vial was optimised in this work. The analysis was carried out using standard flow liquid chromatography coupled to widely available mass spectrometry instrumentation, highlighting opportunities for widespread adoption. The method was applied to 30 living cells, revealing cell-to-cell heterogeneity in the uptake of different drug molecules. Using this system, we detected 14-158 lipid features per single cell, revealing the association between bedaquiline uptake and lipid fingerprints.

REEXAMINING RACE AND ETHNICITY FROM A STRUCTURAL RACISM AND SOCIAL DETERMINANTS OF HEALTH LENS IN COHORT STUDIES

Abstract To advance research on the role of race, ethnicity, and structural racism in health disparities, recent calls-to-action have described methods to correct race and ethnicity categorization and to measure structural and social determinants of health (SSDOH). Following the rapid advance in the understanding of race and SSDOH, long running cohort studies typically lack access to this novel data, which has limited SSDOH evidence to cross-sectional and ecological associations. We implemented methods to rectify this data gap in prospective cohort studies, using The Women’s Health Initiative (WHI) cohort of women age 50–79 years at baseline as a case study. We evaluated the quality, precision, and representativeness of race, ethnicity, and SSDOH data compared with the target US population and operationalized methods to quantify structural determinants in cohort studies. The use of a revised race and ethnicity categorization aligned with theory-based recommendations and improved measurement precision, including decreasing missing data and participants reporting “other” race. The revised categorization was harmonized to the US Census permitting description of the generalizability of WHI race and ethnicity groups in relation to the target population of US women. We found that WHI women reflected broader trends to the target population in relation to structural racism between race and ethnicity groups, while mean rates of SSDOH were systematically more advantageous in WHI. These results underscore the need to implement theoretically justified race and ethnicity measurement to accurately investigate the intersectional role of SSDOH underlying racial and ethnic health disparities among older age women.

Differences in the measurement of cognition for the assessment of dementia across geographic contexts: Recommendations for cross‐national research

AbstractBackgroundAlthough 71% of individuals with dementia are projected to live and low‐ and middle‐income countries by 2050, most dementia research to‐date has been conducted in high‐income countries. Cross‐national studies are needed to understand how the causes and consequences of dementia may differ by country. However, factors such as urbanicity, language of administration or race/ethnicity, which vary across countries, affect the measurement of cognition. A better understanding of which cognitive items perform well across countries or within specific countries will allow researchers to optimize dementia measurement in future cross‐national and international studies.MethodWe used data from the Harmonized Cognitive Assessment Protocol surveys in the US, Mexico, India, England, and South Africa (combined N = 11,364). To define cognitive impairment consistently across countries, we compared participants’ cognitive performance with robust neuropsychological norms within each country. For each HCAP country, we assessed item performance by estimating associations between each cognitive item and cognitive impairment using logistic regression models, controlling for age and gender. We avoided circularity in the analysis by using an iterative quasi‐leave‐one‐out approach. We compared patterns of associations across countries using median odds ratios, and visually, using heatmaps.ResultThe associations between cognitive items and cognitive impairment were stronger in the US (Median Odds Ratio [OR] = 0.17) and England (Median OR = 0.19), in comparison to South Africa (Median OR = 0.23), India (Median OR = 0.29), and Mexico (Median OR = 0.28). Memory items, notably delayed recall tasks, had the most consistent associations of the largest magnitudes across countries. In comparison, there was variability in performance between settings for many language items (e.g. naming a hammer). Items requiring numeracy, including the Trail‐making test, did not perform well in countries with low educational attainment.ConclusionThe performance of cognitive items for the classification of cognitive impairment was not consistent across countries. Although items performing well in a single country could be used to improve measurement precision in specific settings, transporting items between countries without prior validation warrants caution. However, we did identify items that performed well across a range of countries (i.e. delayed recall, animal naming); these items may be of particular utility in future cross‐national research.

A Standardized Protocol for Measuring Bioelectrical Impedance in Green Turtles (Chelonia mydas).

AbstractBioelectrical impedance analysis (BIA) is gaining popularity in wildlife studies as a portable technology for immediate and nondestructive predictions of body composition components, such as fat-free and fat masses. Successful application of BIA for field-based research requires the identification and control of potential sources of error, as well as the creation of and adherence to a standardized protocol for measurement. The aim of our study was to determine sources of error and to provide a standardization protocol to improve measurement precision of BIA on juvenile green turtles (Chelonia mydas; ). We assessed the effects of altered environmental temperature (20°C-30°C), postprandial state (2-72 h), and time out of the water (2 h) on five impedance parameters (resistance at infinite frequency [Rinf], resistance at zero frequency [R0], resistance at 50 kHz [R50], phase angle at 50 kHz [PhA50], and intracellular resistance [Ri]) using a bioimpedance spectroscopy device. Technical reproducibility of measurements and interanimal variability were also assessed. We found an inverse exponential relationship between change in environmental temperature and impedance parameters Rinf, R0, and R50. Postprandial state significantly increased Rinf and Ri 72 h after feeding. BIA measurements were reproducible within individual juvenile green turtles at temperatures from 20°C to 30°C. Significant variation in impedance values was found between animals at all temperatures, sampling times, and postprandial states, but the relative differences (%) were small in magnitude. Our study suggests that measurement precision is improved by measuring animals at consistent environmental temperatures close to their preferred thermal range. We propose a standardized protocol of measurement conditions to facilitate laboratory and field use of BIA for body composition assessment studies in turtles.

Efficient and robust entanglement generation with deep reinforcement learning for quantum metrology

Quantum metrology exploits quantum resources and strategies to improve measurement precision of unknown parameters. One crucial issue is how to prepare a quantum entangled state suitable for high-precision measurement beyond the standard quantum limit. Here, we propose a scheme to optimize the state preparation pulse sequence to accelerate the one-axis twisting dynamics for entanglement generation with the aid of deep reinforcement learning (DRL). We consider the pulse train as a sequence of π/2 pulses along one axis or two orthogonal axes, and the operation is determined by maximizing the quantum Fisher information using DRL. Within a limited evolution time, the ultimate precision bounds of the prepared entangled states follow the Heisenberg-limited scalings. These states can also be used as the input states for Ramsey interferometry and the final measurement precisions still follow the Heisenberg-limited scalings. While the pulse train along only one axis is more simple and efficient, the scheme using pulse sequence along two orthogonal axes show better robustness against atom number difference between simulation and experiment. Our protocol with DRL is efficient and easy to be implemented in state-of-the-art experiments.