Search NASA⌕ Search

SEARCH · Search NASA

Results for “static testing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,045 records · Page 58

Electron stripping processes of H{sup −} ion beam in the 80 kV high voltage extraction column and low energy beam transport line at LANSCE

Basic vacuum calculations were performed for various operating conditions of the Los Alamos National Neutron Science H{sup −} Cockcroft-Walton (CW) injector and the Ion Source Test Stand (ISTS). The vacuum pressure was estimated for both the CW and ISTS at five different points: (1) inside the H{sup −} ion source, (2) in front of the Pierce electrode, (3) at the extraction electrode, (4) at the column electrode, and (5) at the ground electrode. A static vacuum analysis of residual gases and the working hydrogen gas was completed for the normal ion source working regime. Gas density and partial pressure were estimated for the injected hydrogen gas. The attenuation of H{sup −} beam current and generation of electron current in the high voltage acceleration columns and low energy beam transport lines were calculated. The interaction of H{sup −} ions on molecular hydrogen (H{sub 2}) is discussed as a dominant collision process in describing electron stripping rates. These results are used to estimate the observed increase in the ratio of electrons to H{sup −} ion beam in the ISTS beam transport line.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Wide-field microwave magnetic field imaging with nitrogen-vacancy centers in diamond

Non-invasive imaging of microwave (MW) magnetic fields with microscale lateral resolution is pivotal for various applications, such as MW technologies and integrated circuit failure analysis. Diamond nitrogen-vacancy (NV) center magnetometry has emerged as an ideal tool, offering micrometer-scale resolution, millimeter-scale field of view, high sensitivity, and non-invasive imaging compatible with diverse samples. However, up until now, it has been predominantly used for imaging of static or low-frequency magnetic fields or, concerning MW field imaging, to directly characterize the same microwave device used to drive the NV spin transitions. In this work, we leverage an NV center ensemble in diamond for wide-field imaging of MW magnetic fields generated by a test device employing a differential measurement protocol. The microscope is equipped with a MW loop to induce Rabi oscillations between NV spin states, and the MW field from the device-under-test is measured through local deviations in the Rabi frequency. This differential protocol yields magnetic field maps of a 2.57 GHz MW field with a sensitivity of ∼9 μT Hz −1/2 for a total measurement duration of T=357 s, covering a 340 × 340 μm 2 field of view with a micrometer-scale spatial resolution and a device-under-test input power dynamic range of 30 dB. This work demonstrates a novel NV magnetometry protocol, based on differential Rabi frequency measurement, that extends NV wide-field imaging capabilities to imaging of weak MW magnetic fields that would be difficult to measure directly through standard NV Rabi magnetometry.

crystallographic defects↗

Observations of Atmospheric Corrosion Testing on Heated Austenitic Stainless Steel Surfaces: Exploring the Role of Inert Dust Particulates and Seawater

Relevant atmospheric corrosion laboratory testing environments were developed to explore the influence of inert dust and seawater on the corrosion susceptibility of stainless steel in spent nuclear fuel dry storage conditions. Measurements from dust collected on in-service dry storage canisters were applied to develop exposure conditions. Three atmospheric exposure conditions, two static and one cyclic, were examined with three different surface coverages: co-deposited large dust and seawater, co-deposited small dust and seawater, and solely seawater. Stainless steel coupons representative of spent nuclear fuel dry storage canister material were subjected to the various corrosion environments, with the results from exposures up to 1 year presented here. Post exposure, corrosion damage was analyzed using optical microscopy, scanning electron microscopy, and energy dispersive X-ray spectroscopy. Initial observations are presented herein, and potential implications with respect to the influence of inert dust particles on corrosion susceptibility are summarized. In general, the co-deposition of dust and salt resulted in larger pits and exhibited mixed modes of corrosion that were not observed in the no-dust conditions (i.e., crevicing, filiform, and pits within pits). The presence of the inert dust may influence brine spreading and/or act as crevice formers, leading to enhanced corrosion. This study highlights the significance of incorporating dust particulate(s) beyond the deliquescent chemistries to fully evaluate atmospheric corrosion severity.

Sea salt↗

Astrophysical black holes in screened modified gravity

Chameleon, environmentally dependent dilaton, and symmetron gravity are three models of modified gravity in which the effects of the additional scalar degree of freedom are screened in dense environments. They have been extensively studied in laboratory, cosmological, and astrophysical contexts. In this paper, we present a preliminary investigation into whether additional constraints can be provided by studying these scalar fields around black holes. By looking at the properties of a static, spherically symmetric black hole, we find that the presence of a non-uniform matter distribution induces a non-constant scalar profile in chameleon and dilaton, but not necessarily symmetron gravity. An order of magnitude estimate shows that the effects of these profiles on in-falling test particles will be sub-leading compared to gravitational waves and hence observationally challenging to detect.

79 ASTRONOMY AND ASTROPHYSICS↗

Magnetic excitations in stripe-ordered La 1.875 Ba 0.125 CuO 4 studied using resonant inelastic x-ray scattering

The charge and spin correlations in La 1.875 Ba 0.125 CuO 4 (LBCO 1/8) are studied using Cu L 3 edge resonant inelastic x-ray scattering (RIXS). The static charge order (CO) is observed at a wave vector of ( 0.24 , 0 ) and its charge nature confirmed by measuring the dependence of this peak on the incident x-ray polarization. The damped spin excitation or “paramagnon” in LBCO 1/8 is then measured as it disperses through the CO wave vector. Within the experimental uncertainty no changes are observed in the paramagnon at that wave vector and the paramagnon seems to be similar to that measured in other cuprates, which have no static CO. Given that the stripe correlation modulates both the charge and spin degrees of freedom, it is likely that subtle changes do occur in the paramagnon due to CO. Consequently, we propose that future RIXS measurements, realized with higher-energy resolution and sensitivity, should be performed to test for these effects.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Shifting Left for Machine Learning: An Empirical Study of Security Weaknesses in Supervised Learning-based Projects

Context: Supervised learning-based projects (SLPs), i.e., software projects that use supervised learning algorithms, such as decision trees are useful for performing classification-related tasks. Yet, security weaknesses, such as the use of hard-coded passwords in SLPs, can make SLPs susceptible to security attacks. A characterization of security weaknesses in SLPs can help practitioners understand the security weaknesses that are frequent in SLPs and adopt adequate mitigation strategies. Objective: The goal of this paper is to help practitioners se-curely develop supervised learning-based projects by conducting an empirical study of security weaknesses in supervised learning-based projects. Methodology: We conduct an empirical study by quantifying the frequency of security weaknesses in 278 open source SLPs. Results: We identify 22 types of security weaknesses that occur in SLPs. We observe ‘use of potentially dangerous function’ to be the most frequently occurring security weakness in SLPs. Of the identified 3,964 security weaknesses, 23.79 % and 40.49 % respectively, appear for source code files used to train and test models. We also observe evidence of co-location, e.g., instances of command injection co-locates with instances of potentially dangerous function. Conclusion: Based on our findings, we advocate for a shift left approach for SLP development with security-focused code reviews, and application of security static analysis.

Bhuiyan, Farzana Ahamed↗

SU-F-T-255: Accuracy and Precision of Dynamic Tracking Irradiation with VERO-4DRT System

Purpose: The VERO-4DRT system is able to provide dynamic tracking irradiation (DTI) for the target with respiratory motion. This technique requires enough commissioning for clinical implementation. The purpose of this study is to make sure the accuracy and precision of DTI using VERO- 4DRT through commissioning from fundamental evaluation to end-to-end test. Method: We evaluated several contents for DTI commissioning: the accuracy of absorption dose at isocenter in DTI, the field size and penumbra of DTI, the accuracy of 4D modeling in DTI. All evaluations were performed by respiratory motion phantom (Quasar phantom). These contents were compared the results between static irradiation and DTI. The shape of radiation field was set to square from 3 cm × 3 cm to 10 cm × 10 cm. The micro 3D chamber and Gafchromic EBT3 film were used for absorbed dose and relative dose distribution measurement, respectively. The sine and irregular shaped waves were used for demonstrative respiratory motion. The visicoil was implanted into the phantom for guidance of respiratory motion. The respiration patterns of frequency and motion amount were set to 10–15 BPM and 1–2 cm, respectively. Results: As the result of absorbed dose of DTI in comparison with static irradiation, the average dose error at isocenter was 0.5% even though various respiratory patterns were set on. As the result of relative dose distribution, the field size (set it on 50% dose line) was not significantly changed in all respiratory patterns. However, the penumbra was larger in greater respiratory motion (up to 4.1 mm). The 4D modeling coincidence between actual and created waves was within 1%. Conclusion: The DTI using VERO-4DRT can provide sufficient accuracy and precision in absorbed dose and distribution. However, the patientspecific quantitative internal margin corresponding respiratory motion should be taken into consideration with image guidance.

60 APPLIED LIFE SCIENCES↗

Minimum Detectable Quantity Calculation for Radiation Portal Monitors

The minimum detectable quantity of a radiation portal monitor is the smallest amount of radioactive material that can be detected passing through the monitor (with a specified detection probability). The minimum detectable quantity is a function of many factors, including isotope, velocity, distance between pillars, height of portal and vehicle, source distribution and position, background radiation, background suppression, detector volume and positions, detector efficiency, decision metrics and algorithms, the presence of NORM (naturally occurring radioactive material), and alarm thresholds. Experimentally testing a radiation portal monitor to failure for all combinations of these factors is extremely time-consuming and often cost prohibitive. This document outlines a straightforward method for quickly estimating the minimum detectable quantity for moving sources over a large variety of conditions with a minimal number of static measurements. The document also describes a proof of concept software application that the International Atomic Energy Agency has designed based on the described method that can be used by Member States to estimate a monitor’s minimum detectable quantity.

Blessinger, Christopher S.↗

Multicenter Collaborative Quality Assurance Program for the Province of Ontario, Canada: First-Year Results

Purpose: The objective of this work was to develop a collaborative quality assurance (CQA) program to assess the performance of intensity modulated radiation therapy (IMRT) planning and delivery across the province of Ontario, Canada. Methods and Materials: The CQA program was designed to be a comprehensive end-to-end test that can be completed on multiple planning and delivery platforms. The first year of the program included a head-and-neck (H and N) planning exercise and on-site visit to acquire dosimetric measurements to assess planning and delivery performance. A single dosimeter was used at each institution, and the planned to measured dose agreement was evaluated for both the H and N plan and a standard plan (linear-accelerator specific) that was created to enable a direct comparison between centers with similar infrastructure. Results: CQA program feasibility was demonstrated through participation of all 13 radiation therapy centers in the province. Planning and delivery was completed on a variety of infrastructure (treatment planning systems and linear accelerators). The planning exercise was completed using both static gantry and rotational IMRT, and planned-to-delivered dose agreement (pass rates) for 3%/3-mm gamma evaluation were greater than 90% (92.6%-99.6%). Conclusions: All centers had acceptable results, but variation in planned to delivered dose agreement for the same planning and delivery platform was noted. The upper end of the range will provide an achievable target for other centers through continued quality improvement, aided by feedback provided by the program through the use of standard plans and simple test fields.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Dosimetric properties of an amorphous silicon EPID for verification of modulated electron radiotherapy

Purpose: To investigate the dosimetric properties of an electronic portal imaging device (EPID) for electron beam detection and to evaluate its potential for quality assurance (QA) of modulated electron radiotherapy (MERT). Methods: A commercially available EPID was used to detect electron beams shaped by a photon multileaf collimator (MLC) at a source-surface distance of 70 cm. The fundamental dosimetric properties such as reproducibility, dose linearity, field size response, energy response, and saturation were investigated for electron beams. A new method to acquire the flood-field for the EPID calibration was tested. For validation purpose, profiles of open fields and various MLC fields (square and irregular) were measured with a diode in water and compared to the EPID measurements. Finally, in order to use the EPID for QA of MERT delivery, a method was developed to reconstruct EPID two-dimensional (2D) dose distributions in a water-equivalent depth of 1.5 cm. Comparisons were performed with film measurement for static and dynamic monoenergy fields as well as for multienergy fields composed by several segments of different electron energies. Results: The advantageous EPID dosimetric properties already known for photons as reproducibility, linearity with dose, and dose rate were found to be identical for electron detection. The flood-field calibration method was proven to be effective and the EPID was capable to accurately reproduce the dose measured in water at 1.0 cm depth for 6 MeV, 1.3 cm for 9 MeV, and 1.5 cm for 12, 15, and 18 MeV. The deviations between the output factors measured with EPID and in water at these depths were within {+-}1.2% for all the energies with a mean deviation of 0.1%. The average gamma pass rate (criteria: 1.5%, 1.5 mm) for profile comparison between EPID and measurements in water was better than 99% for all the energies considered in this study. When comparing the reconstructed EPID 2D dose distributions at 1.5 cm depth to film measurements, the gamma pass rate (criteria: 2%, 2 mm) was better than 97% for all the tested cases. Conclusions: This study demonstrates the high potential of the EPID for electron dosimetry, and in particular, confirms the possibility to use it as an efficient verification tool for MERT delivery.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Thermite and intermetallic projectiles examined experimentally in air and inert gas environments

Intermetallic (aluminum and zirconium) and thermite (aluminum and molybdenum trioxide) projectiles were launched using a high velocity impact ignition testing system. The experiments were designed to simulate reactivity in high (argon) and low (air) altitude environments. The projectiles were launched into a chamber that included a steel target plate for projectile penetration before impacting a rear witness plate. The chamber was semi-sealed and instrumented for quasi-static pressure data. The results provide an understanding of energy release from the projectile materials and of the environmental influence on performance. The transient pressure traces provide insight into reaction kinetics. A bifurcation in transient pressure rise was an indication of a shift in reaction kinetics from the inherent reactive material to metal oxidation with the environment. The bifurcation was delayed by about 0.15 ms for the intermetallic relative to the thermite, evidence that the thermite reaction proceeded faster upon impact than the intermetallic. The two-step process (impact ignition of the reactive material followed by metal oxidation) was shown to produce higher energy conversion efficiencies than projectiles composed of pure fuel (i.e., aluminum) reported previously. Both reactive materials showed energy conversion efficiencies greater than 30% (for air) and 50% (for argon), and an explanation of underestimated efficiency and energy losses is provided. These results have implications for advancing formulations for ballistic applications. Structural reactive materials can be used to modify the effective reactivity of metal-containing formulations in varied atmospheric environments.

36 MATERIALS SCIENCE↗

Improving the nonlinear control performance of the supply fan at air handling units using a gain scheduling control strategy

Due to its nonlinear nature, the supply fan at air handling units with the controller tuned at the design condition tends to be aggressive and oscillate under partial load conditions. Here, the objective of this paper is to develop and validate a gain scheduling control strategy to improve its nonlinear control performance. First, a control-oriented model, which does not require numerous physical parameters and extensive test data, is developed to study the nonlinearity of the fan system. Based on the theoretical model and experimental verifications, the issue of an aggressive response with a conventional fixed-gain controller is caused by the fact that the system gain is proportional to the ratio of the duct static pressure to the fan speed. To address the issue, a scheduling function of the measurable duct static pressure and fan speed is proposed to be included in the conventional fixed-gain controller to compensate for the fan system gain variation. The gain scheduling control strategy is found to approximately maintain the identical control performance under all operation conditions. Most importantly, the gain scheduling control strategy can be readily implemented without intensive computation and additional measurements, showing a promising potential in industrial applications.

42 ENGINEERING↗

Initial conditions for cosmological N-body simulations of the scalar sector of theories of Newtonian, Relativistic and Modified Gravity

We present a description for setting initial particle displacements and field values for simulations of arbitrary metric theories of gravity, for perfect and imperfect fluids with arbitrary characteristics. We extend the Zel'dovich Approximation to nontrivial theories of gravity, and show how scale dependence implies curved particle paths, even in the entirely linear regime of perturbations. For a viable choice of Effective Field Theory of Modified Gravity, initial conditions set at high redshifts are affected at the level of up to 5% at Mpc scales, which exemplifies the importance of going beyond Λ-Cold Dark Matter initial conditions for modifications of gravity outside of the quasi-static approximation. In addition, we show initial conditions for a simulation where a scalar modification of gravity is modelled in a Lagrangian particle-like description. Our description paves the way for simulations and mock galaxy catalogs under theories of gravity beyond the standard model, crucial for progress towards precision tests of gravity and cosmology.

79 ASTRONOMY AND ASTROPHYSICS↗

Drift orbit islands of energetic particles due to 3D fields in ITER

In this study, we report a comparative study of the orbit islands formed by the guiding center drift motion of test energetic particles (EPs), including both high-energy deuterium ions and fusion-born α-particles, with respect to the magnetic islands formed by tracing the field lines, for an ITER plasma representing the 15 MA baseline scenario. The particle drift orbit is modified by the presence of static resonant magnetic perturbations (RMPs) of different toroidal mode numbers n (n = 1 and 3 in this study), forming orbit islands in the Poincaré plane by passing EPs. The key findings are as follows. (i) With an n = 1 RMP field, the size of the orbit islands combined with the total RMP field, including the plasma response, is about three times smaller than that obtained by assuming the corresponding vacuum field. (ii) The orbit island size is not sensitive to the EP energy, and is comparable to those of the magnetic islands. (iii) Passing EPs with outward radial orbital drift form orbit islands that shift inward with respect to the corresponding magnetic islands, and vice versa. This radial shift, ΔΨ, measured in the normalized equilibrium poloidal flux, is quantified as ΔΨ ≃ 0.03X for the ITER baseline plasma and assuming n = 1 RMPs, with $X\equiv {(M/{M}_{\mathrm{p}})}^{1/2}{(E[\mathrm{M}\mathrm{e}\mathrm{V}])}^{1/2}{Z}^{-1}$, where M is the particle mass, Mp is the proton mass, E is the particle energy, and Z is the particle charge number. (iv) Trapped EPs do not form drift orbit islands, even in the presence of the 3D fields produced by the 90 kAt ELM control coil current in ITER, and thus possess good confinement properties. Nevertheless, the deformation of the trapped particle drift orbits in the Poincaré plane shows that the canonical toroidal angular momentum of EPs is no longer conserved in the presence of the RMP fields.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Toward Automated Detection of Portability Bugs in Kokkos Parallel Programs

Performance-portable programming frameworks provide abstractions for parallel execution to allow easily porting an application to multiple backend programming models, such as CUDA, HIP, and OpenMP. However, programs may still have portability bugs that manifest only on specific backends. Traditional testing is ineffective in discovering these bugs, as it would require concrete execution on all supported hardware configurations for a potentially infinite set of inputs. To mitigate this issue, we focused on a specific programming framework, Kokkos, and identified several categories of common portability bugs. We then developed Klokkos, a static analysis approach based on symbolic execution that can run on commodity hardware, before execution on supercomputers. As a proof-of-concept, we ran Klokkos on examples encoding the identified bugs. Our results show that Klokkos is effective, efficient, and precise: it detected all the considered bugs, quickly, and without any false positives. Although preliminary, our results motivate further research and development in this direction.

Kale, Vivek↗

TH-AB-202-04: Auto-Adaptive Margin Generation for MLC-Tracked Radiotherapy

Purpose: To develop an auto-adaptive margin generator for MLC tracking. The generator is able to estimate errors arising in image guided radiotherapy, particularly on an MR-Linac, which depend on the latencies of machine and image processing, as well as on patient motion characteristics. From the estimated error distribution, a segment margin is generated, able to compensate errors up to a user-defined confidence. Method: In every tracking control cycle (TCC, 40ms), the desired aperture D(t) is compared to the actual aperture A(t), a delayed and imperfect representation of D(t). Thus an error e(t)=A(T)-D(T) is measured every TCC. Applying kernel-density-estimation (KDE), the cumulative distribution (CDF) of e(t) is estimated. With CDF-confidence limits, upper and lower error limits are extracted for motion axes along and perpendicular leaf-travel direction and applied as margins. To test the dosimetric impact, two representative motion traces were extracted from fast liver-MRI (10Hz). The traces were applied onto a 4D-motion platform and continuously tracked by an Elekta Agility 160 MLC using an artificially imposed tracking delay. Gafchromic film was used to detect dose exposition for static, tracked, and error-compensated tracking cases. The margin generator was parameterized to cover 90% of all tracking errors. Dosimetric impact was rated by calculating the ratio between underexposed points (>5% underdosage) to the total number of points inside FWHM of static exposure. Results: Without imposing adaptive margins, tracking experiments showed a ratio of underexposed points of 17.5% and 14.3% for two motion cases with imaging delays of 200ms and 300ms, respectively. Activating the margin generated yielded total suppression (<1%) of underdosed points. Conclusion: We showed that auto-adaptive error compensation using machine error statistics is possible for MLC tracking. The error compensation margins are calculated on-line, without the need of assuming motion or machine models. Further strategies to reduce consequential overdosages are currently under investigation. This work was funded by the SoRTS consortium, which includes the industry partners Elekta, Philips and Technolution.

60 APPLIED LIFE SCIENCES↗

OECD MCCI Project: Category 4 Integral Test to Validate Severe Accident Codes and Core-Concrete Interaction Test Six (CCI-6) (Rev. 2, Final Report)

Ex-vessel debris coolability is an important light water reactor (LWR) technical challenge. For existing plants, resolution of this challenge will confirm the technical basis for severe accident management guidelines (SAMGs). For new reactors, understanding this challenge will help confirm the effectiveness of the design and implementation of new accident mitigation features and severe accident management design alternatives (SAMDAs). The first OECD-MCCI program conducted reactor material experiments focused on achieving the following technical objectives: i) provide confirmatory evidence and data for various cooling mechanisms through separate effect tests for severe accident model development, and ii) provide long-term 2-D coreconcrete interaction data for code assessment and improvement. In addition to the experimental work, an analysis task was defined to develop and validate coolability models to form the basis for extrapolating the experiment findings to plant conditions. In response to the PRG mandate, a summary test plan for this sixth Core-Concrete Interaction (CCI-6) experiment was prepared, reviewed by the PRG, and subsequently approved for test conduct. The experiment approach was to incorporate an array of water nozzles into the core-concrete interaction test apparatus. The test section was enlarged to incorporate a 70 cm x 70 cm basemat. The initial melt depth was increased to 28 cm, which corresponds to a thermite charge mass of 900 kg. The melt depth is noted to be slightly less than the 30 cm level requested by the PRG. The thermite charge was redesigned to include a reduced amount of concrete decomposition products (i.e., 6 wt %) relative to the previously performed CCI tests. The concrete sidewalls allow for 24 cm of lateral erosion, and the sidewall design was modified to include refractory MgO with a UO 2 pellet and U 3 O 8 powder liner to stop the ablation if the 24 cm ablation depth was reached. The basemat and sidewalls also included an array of 45 concrete nozzles that were arranged on a 14 cm lateral pitch and were located at a uniform depth of 27.5 cm below the initial concrete basemat surface elevation. The nozzles were originally sealed at the top, and were fed with water from a common water supply tank that provided a static water head of nominally 10 kPa measured at the upper surface of the nozzles. The planned operating sequence was to ignite the thermite and then provide 210 kW Direct Electrical Heating (DEH) to simulate decay heat. The concrete would then heat up and begin erosion. At this point, the cavity would be flooded from the top. If the melt did not cool, then the core-concrete interaction would continue until the 27.5 cm axial ablation depth was reached. At this point, the nozzles would open by erosion and thus initiate cooling by bottom water injection. The test would continue to operate until the melt was quenched, or the maximum axial ablation depth of 32.5 cm was reached. The experiment would then be terminated. The CCI-6 experiment was carried out on 18 March 2010. The balance of this report documents the facility, actual test operating procedure, and results from this experiment.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Prediction of chronic kidney disease progression using recurrent neural network and electronic health records

Chronic kidney disease (CKD) is a progressive loss in kidney function. Early detection of patients who will progress to late-stage CKD is of paramount importance for patient care. To address this, we develop a pipeline to process longitudinal electronic heath records (EHRs) and construct recurrent neural network (RNN) models to predict CKD progression from stages II/III to stages IV/V. The RNN model generates predictions based on time-series records of patients, including repeated lab tests and other clinical variables. Our investigation reveals that using a single variable, the recorded estimated glomerular filtration rate (eGFR) over time, the RNN model achieves an average area under the receiver operating characteristic curve (AUROC) of 0.957 for predicting future CKD progression. When additional clinical variables, such as demographics, vital information, lab test results, and health behaviors, are incorporated, the average AUROC increases to 0.967. In both scenarios, the standard deviation of the AUROC across cross-validation trials is less than 0.01, indicating a stable and high prediction accuracy. Our analysis results demonstrate the proposed RNN model outperforms existing standard approaches, including static and dynamic Cox proportional hazards models, random forest, and LightGBM. The utilization of the RNN model and the time-series data of previous eGFR measurements underscores its potential as a straightforward and effective tool for assessing the clinical risk of CKD patients concerning their disease progression.

60 APPLIED LIFE SCIENCES↗