Results for “MONITORING” — Page 46

NASA NTRS2023

Contractors Road Heavy Equipment Area (SWMU 055) 2022 Annual Groundwater Monitoring Report

This document presents a summary of activities completed at the Contractors Road Heavy Equipment (CRHE) Area, located at Kennedy Space Center (KSC), Florida, from June through December 2022. The activities conducted at the CRHE Area include: - Annual groundwater sampling activities associated with sitewide plume monitoring and Underground Injection Control (UIC) monitoring in the former bioremediation Interim Measures (IM) Hot Spot 1 (HS1) area; - Direct push technology (DPT) groundwater investigation; - Semiannual vapor intrusion (VI) evaluation activities. This facility is designated Solid Waste Management Unit (SWMU) Number 055 (SWMU 055) under KSC’s Resource Conservation and Recovery Act (RCRA) Corrective Action program. HydroGeoLogic, Inc. (HGL) prepared this report for the National Aeronautics and Space NASA) under contract number 80KSC019F0096/80KSC019D0012, Technical Directive-03. An Advance Data Package (ADP) presentation of the elements of this report received Team (KSCRT) agreement at the April 5, 2023, KSCRT meeting. The chlorinated volatile organic compound plume appears to be stable vertically and horizontally. Additional DPT groundwater sampling will be completed to support a remedial alternatives evaluation for potential means to expedite groundwater cleanup. There are per- and polyfluorinated alkyl substances in site groundwater, and assessment is planned in the future. Sub-slab soil gas monitoring results do not indicate any concerns with indoor vapor intrusion. Based on the 2022 annual groundwater sampling results, it is recommended that annual sampling continue to alternate with the wet and dry season, with the next sampling event planned for December 2023.

groundwater↗

NASA NTRSDate not supplied

Developmental Hardware Testing Results and Forward Plans for the Spacecraft Water Impurity Monitor (SWIM) Organic Water Module (OWM)

We present testing results for developmental hardware of the Spacecraft Water Impurity Monitor (SWIM) Organic Water Module (OWM). SWIM-OWM will monitor spacecraft potable water and system water for trace organic contaminants. The system will detect and identify the specific organic chemical that makes up a given total organic carbon reading. We have built a first development unit (1DU) for SWIM-OWM, which directly injects aqueous water samples and detects chemicals with both a thermal conductivity detector and mass spectrometer sensor. The gas chromatography mass spectrometer (GCMS) system that comprises SWIM-OWM draws on the success of ISS-proven mass spectrometer hardware, and the demonstration of GCMS detection of trace organic contaminants in ISS cabin air. SWIM-OWM benefits from the excellent sensitivity and specificity afforded by GCMS. We have demonstrated detection of a set of chemicals relevant to both crew health and performance as well as system monitoring; these target chemicals range from light, volatile organics such as acetone and ethanol, to heavier, very non-volatile compounds such as dimethyl sulfone and o-phthalaldehyde. Direct aqueous injection was chosen for the general applicability of the technique to clean water sampling and to preclude sample pre-processing, which facilitates an on-line implementation of the SWIM-OWM when deployed in a spacecraft or habitation module. A specific advantage of direct aqueous injection when coupled with appropriate methods is that both the light, volatile organics and heavier non-volatiles can be detected from a single injection, in a single chromatogram. Results from 1DU testing will be discussed, and forward plans will be outlined for continued maturation of SWIM-OWM with the goal of implementing a technology demonstration for the purposes of maturing the engineering design and operations in an environment relevant to NASA’s future goals of exploring and setting up habitation on the Moon and Mars.

Water Monitoring↗

NASA NTRSDate not supplied

Development of a Photoacoustic Formaldehyde Monitor

Key indoor air quality pollutant formaldehyde (H 2 CO) is tracked on International Space Station (ISS) using passive badges returned to the ground periodically for analysis. The process is time-consuming both in preparation and for analysis upon return 6-12 months later. Badges also require precious crew time for deploy, retrieval and stow. As NASA’s focus in space exploration shifts to the Moon and Mars, archival sample return becomes increasingly impractical, so the aim of this project is to develop a highly reliable real-time analyzer for H 2 CO at low concentrations with data downlinked. Potential sources of H 2 CO include materials off gassing, use of formalin as a tissue fixative in biological payloads and overheating of acetal polymers. The Spacecraft Maximum Allowable Concentration (SMAC) for H 2 CO in air is 100 ppb for exposures of 7 days or longer. ISS concentrations of H 2 CO are running only 10 - 30 ppb in the recent several years but have spiked as high as 60 ppb. Gateway real time monitoring requirements for H 2 CO call for a range of 8 - 138 ppb. For this project, a concentration range of 5 - 500 ppb H 2 CO is targeted. The core tunable diode laser spectroscopy (TDLS) technology used for this project was developed by Vista Photonics through the NASA and US Navy Small Business Innovation Research (SBIR) programs. Monitors based on TDLS have been demonstrated on ISS, trialed on a nuclear submarine and are being certified as portable Anomaly Gas Analyzers (AGA) for both ISS and Orion. Initially, direct absorption TDLS was used exclusively in these devices. The H 2 CO target concentration, however, is low ppb, and a longer wavelength is required, so a photoacoustic spectroscopy (PAS) technique was adapted, where the laser excitation is detected by a sensitive microphone vs. a conventional photodetector. This paper will discuss the results of NASA-JSC laboratory testing of a prototype PAS based formaldehyde in air monitor and briefly discuss next steps.

Photoacoustic Formaldehyde Monitor↗

NASA NTRSDate not supplied

Developmental Hardware Testing Results and Forward Plans for Spacecraft Water Impurity Monitor (SWIM) Organic Water Module (OWM)

We present testing results for developmental hardware of the Spacecraft Water Impurity Monitor (SWIM) Organic Water Module (OWM). SWIM-OWM will monitor spacecraft potable water and system water for trace organic contaminants. The system will detect and identify the specific organic chemical that makes up a given total organic carbon reading. We have built a first development unit (1DU) for SWIM-OWM, which directly injects aqueous water samples and detects chemicals with both a thermal conductivity detector and mass spectrometer sensor. The gas chromatography mass spectrometer (GCMS) system that comprises SWIM-OWM draws on the success of ISS-proven mass spectrometer hardware, and the demonstration of GCMS detection of trace organic contaminants in ISS cabin air. SWIM-OWM benefits from the excellent sensitivity and specificity afforded by GCMS. We have demonstrated detection of a set of chemicals relevant to both crew health and performance as well as system monitoring; these target chemicals range from light, volatile organics such as acetone and ethanol, to heavier, very non-volatile compounds such as dimethyl sulfone and o-phthalaldehyde. Direct aqueous injection was chosen for the general applicability of the technique to clean water sampling and to preclude sample pre-processing, which facilitates an on-line implementation of the SWIM-OWM when deployed in a spacecraft or habitation module. A specific advantage of direct aqueous injection when coupled with appropriate methods is that both the light, volatile organics and heavier non-volatiles can be detected from a single injection, in a single chromatogram. Results from 1DU testing will be discussed, and forward plans will be outlined for continued maturation of SWIM-OWM with the goal of implementing a technology demonstration for the purposes of maturing the engineering design and operations in an environment relevant to NASA’s future goals of exploring and setting up habitation on the Moon and Mars.

Water Monitoring↗

DOE OSTI2020

Leveraging Environmental Monitoring Results from Department of Energy Sites for Research on the Food-Energy-Water Nexus

The Annual Site Environmental Report (ASER) from a U.S. Department of Energy (DOE) Site summarizes environmental data; environmental management performance; compliance with applicable DOE, federal, state, and local regulations; and restoration and surveillance monitoring programs. The Savannah River Site (SRS) produces ASERs that include results of monitoring mercury levels in fish samples obtained in the vicinity of SRS. Mercury is a toxic trace metal and the methylated organic form is a neurotoxin that bioaccumulates in the food web, endangering humans and other organisms. In United States, power plants are the largest source of mercury emissions. Research Objective: To determine the patterns of mercury levels in fish samples reported in the SRS environmental monitoring reports. We developed box plot visualizations to make sense of the data as well as identify fish species, dates and location with outlier mercury content. The integration and visual representation of multi-year datasets can support decision making processes on environmental compliance and environmental sustainability conducted at U.S. DOE Sites. We are constructing datasets from textual, image and multivariable data sources in the multi-year ASERs of the U.S. DOE Savannah River Site (SRS https://www.srs.gov/general/pubs/ERsum/index.html). The Fish Surveillance for mercury presents a case study for the intersection of food, energy and water systems. The dataset of 864 records (2008 and 2009) of mercury levels includes water location of fish, fish species, collection date and mercury content of fish sample. The patterns of mercury levels in the fish species as well as locations provide knowledge for further investigations. A research area for future studies is the composition of microbial communities in fish guts of varying levels of mercury.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

In-line, High-Throughput Quality Monitoring for Fuel Cell and Electrolyzer Components Based on Transmission and Reflection Imaging

During the manufacturing of fuel cell and electrolyzer membranes and membrane electrode assemblies (MEAs), real-time, in-line, high-throughput optical-based quality monitoring methods are essential for detecting defects and monitoring thickness variations, thus improving the performance and increasing the durability of fuel cell and electrolyzer in the hydrogen industry. For the MEAs with very opaque coatings, optical transmission-based imaging has been developed and applied in the Roll-to-Roll system using a flashlight and a high-sensitivity CCD camera. We observed high signal-to-noise ratio images while the Roll-to-Roll system ran at 5 ft/min. The entire sample image could quickly be recovered from the discrete frames using customized Python codes for automatic frame cropping and stitching. We detected significant non-uniformities in our experimental MEAs specimen. For fuel cell and low-temperature electrolysis (LTE) transparent membranes, we used optical reflectance hyperspectral imaging with interference fringe-based thickness mapping. We set up a hyperspectral camera to measure various rolls of commercial membranes. The measurement results are analyzed to find the thickness distribution of each roll and to check for defects. Transmission and reflection imaging-based quality monitoring techniques demonstrated in this project can be widely used in the mass production environment to improve the production yield and performance of hydrogen devices.

DIRECT ENERGY CONVERSION,ENGINEERING↗

DOE OSTI2024

Ecological Monitoring and Compliance Program 2023 Report

Annual report of activities conducted under the EMAC program including biological surveys, desert tortoise compliance, ecosystem monitoring, sensitive and protected/regulated plant and animal monitoring, and habitat restoration implementation and monitoring.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2024

Acoustic Monitoring of Pyroprocessing for Safeguards

As pyroprocessing continues to be an attractive option for the reprocessing of spent nuclear fuel worldwide, safeguards technologies are needed to address the monitoring capabilities that can help state level authorities, or the International Atomic Energy Agency (IAEA) maintain continuity of knowledge of the plant operations. Idaho National Laboratory (INL) is studying the possibility of using acoustic monitoring as a means to monitor a pyroprocessing facility for safeguards purposes. This paper discusses the experimental design and some preliminary results of tests conducted at the Fuel Conditioning Facility, a pyrochemical capable facility, at INL.

98 - NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL↗

DOE OSTI2025

Risk-informed Graded Approach for Reliability and Performance Assessment of Machine Learning and Artificial Intelligence for Advanced Condition Monitoring Techniques

With the shift away from time-based maintenance and toward condition-based maintenance, and to reduce overall maintenance costs, there has been an upsurge in the usage and development of advanced condition monitoring (ACM) techniques for real-time monitoring of nuclear power plant (NPP) components. ACM is particularly useful in the development of digital twins, which are designed to predict the failure or degradation of plant components. Successful implementation of ACM requires an assessment to inform the development of a risk-informed approach to evaluate the use of ACM to meet Nuclear Regulatory Committee (NRC) regulations for in-service testing (IST) programs. This includes the monitoring and diagnostics of reactor components and systems in current, new, and advanced reactors. A key component in ACM is the usage of machine learning (ML) and artificial intelligence (AI) algorithms that can employ real-time data from instrumentation and sensors to detect and predict reactor component degradations. Such predictive capabilities enable early detection of component degradation so as to help plant personnel plan and execute necessary maintenance. For successful implementation of ML/AI in ACM such that regulatory requirements are met, a risk-informed graded approach is needed to assess the reliability and performance of ML/AI for ACM. The American Society for Mechanical Engineers (ASME) developed their Operations and Maintenance (O&M) Code to provide guidance on safe, reliable O&M of NPPs. The IST section of the O&M Code specifically establishes requirements for IST and examination to gauge operational readiness of components in water-cooled NPPs. This paper presents a state-of-the-art review of how reliability and risk assessment can be integrated with ACM to assess component performance by non-nuclear industries. This is followed by different methodologies and approaches for conducting performance and reliability assessments so as to meet IST requirements for NPP components.

97 - MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Risk-informed Graded Approach for Reliability and Performance Assessment for Advanced Condition Monitoring Techniques

With the shift away from time-based maintenance and toward condition-based maintenance, and to reduce overall maintenance costs, there has been an upsurge in the usage and development of advanced condition monitoring (ACM) techniques for real-time monitoring of nuclear power plant (NPP) components. ACM is particularly useful in the development of digital twins, which are designed to predict the failure or degradation of plant components. Successful implementation of ACM requires an assessment to inform the development of a risk-informed approach to evaluate the use of ACM to meet Nuclear Regulatory Committee (NRC) regulations for in-service testing (IST) programs. This includes the monitoring and diagnostics of reactor components and systems in current, new, and advanced reactors. A key component in ACM is the usage of machine learning (ML) and artificial intelligence (AI) algorithms that can employ real-time data from instrumentation and sensors to detect and predict reactor component degradations. Such predictive capabilities enable early detection of component degradation so as to help plant personnel plan and execute necessary maintenance. For successful implementation of ML/AI in ACM such that regulatory requirements are met, a risk-informed graded approach is needed to assess the reliability and performance of ML/AI for ACM. The American Society for Mechanical Engineers (ASME) developed their Operations and Maintenance (O&M) Code to provide guidance on safe, reliable O&M of NPPs. The IST section of the O&M Code specifically establishes requirements for IST and examination to gauge operational readiness of components in water-cooled NPPs. This paper presents a state-of-the-art review of how reliability and risk assessment can be integrated with ACM to assess component performance by non-nuclear industries. This is followed by different methodologies and approaches for conducting performance and reliability assessments so as to meet IST requirements for NPP components.

99 - GENERAL AND MISCELLANEOUS↗

DOE OSTI2025

The CMS Phase-2 Fast Beam Condition Monitor prototype test with beam

The Fast Beam Condition Monitor (FBCM) is a standalone luminometer for the High Luminosity LHC (HL-LHC) program of the CMS Experiment at CERN. The detector is under development and features a new, radiation-hard, front-end application-specific integrated circuit (ASIC) designed for beam monitoring applications. The achieved timing resolution of a few nanoseconds enables the measurement of both the luminosity and the beam-induced background. The ASIC, called FBCM23, features six channels with adjustable shaping times, enabling in-field fine-tuning. Each ASIC channel outputs a single binary asynchronous signal encoding time-of-arrival and time-over-threshold information. The FBCM is based on silicon-pad sensors, with two sensor designs presently being considered. This paper presents the results of tests of the FBCM detector prototype using both types of silicon sensors with hadron, muon, and electron beams. Irradiated FBCM23 ASICs and silicon-pad sensors were also tested to simulate the expected conditions near the end of the detector's lifetime in the HL-LHC radiation environment. Based on test results, direct bonding between the sensor and ASIC was chosen, and an optimal bias voltage and ASIC threshold for FBCM operation were proposed. The current design of the front-end test board was validated following the beam test and is now being used for the first front-end module, which is expected to be produced in summer 2025. These results represent a major step forward in validating the FBCM concept, first version of the firmware and establishing a reliable design path for the final detector.

Beam-line instrumentation (beam position and profi↗

NASA NTRS2013

Advanced Signal Processing for High Temperatures Health Monitoring of Condensed Water Height in Steam Pipes

An advanced signal processing methodology is being developed to monitor the height of condensed water thru the wall of a steel pipe while operating at temperatures as high as 250deg. Using existing techniques, previous study indicated that, when the water height is low or there is disturbance in the environment, the predicted water height may not be accurate. In recent years, the use of the autocorrelation and envelope techniques in the signal processing has been demonstrated to be a very useful tool for practical applications. In this paper, various signal processing techniques including the auto correlation, Hilbert transform, and the Shannon Energy Envelope methods were studied and implemented to determine the water height in the steam pipe. The results have shown that the developed method provides a good capability for monitoring the height in the regular conditions. An alternative solution for shallow water or no water conditions based on a developed hybrid method based on Hilbert transform (HT) with a high pass filter and using the optimized windowing technique is suggested. Further development of the reported methods would provide a powerful tool for the identification of the disturbances of water height inside the pipe.

health monitoring↗

DOE OSTI2020

Organic Microbial Electrochemical Transistor Monitoring Extracellular Electron Transfer

Abstract Extracellular electron transfer (EET) denotes the process of microbial respiration with electron transfer to extracellular acceptors and has been exploited in a range of microbial electrochemical systems (MESs). To further understand EET and to optimize the performance of MESs, a better understanding of the dynamics at the microscale is needed. However, the real‐time monitoring of EET at high spatiotemporal resolution would require sophisticated signal amplification. To amplify local EET signals, a miniaturized bioelectronic device, the so‐called organic microbial electrochemical transistor (OMECT), is developed, which includes Shewanella oneidensis MR‐1 integrated onto organic electrochemical transistors comprising poly(3,4‐ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) combined with poly(vinyl alcohol) (PVA). Bacteria are attached to the gate of the transistor by a chronoamperometric method and the successful attachment is confirmed by fluorescence microscopy. Monitoring EET with the OMECT configuration is achieved due to the inherent amplification of the transistor, revealing fast time‐responses to lactate. The limits of detection when using microfabricated gates as charge collectors are also investigated. The work is a first step toward understanding and monitoring EET in highly confined spaces via microfabricated organic electronic devices, and it can be of importance to study exoelectrogens in microenvironments, such as those of the human microbiome.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2023

Effects of Livestock Exclusion on Stream Habitat and Aquatic Biota: A Review and Recommendations for Implementation and Monitoring

Abstract To inform riparian restoration, research, and monitoring and to provide management recommendations, we reviewed published studies evaluating the physical and biological effectiveness of livestock exclusion and grazing reduction on various metrics in riparian and aquatic areas. We identified 95 North American studies that reported the effects of livestock grazing reduction on physical habitat (channel morphology, mesohabitats, substrate, and bank stability), biological assemblages (riparian vegetation, macroinvertebrates, fish, and birds), and water quality metrics (temperature, nitrates, phosphorus, and turbidity). Most studies reported that methods to reduce or exclude livestock decreased channel width, width-to-depth ratio, bank erosion, soil bulk density, bare ground, water temperature, nitrogen, and phosphorus and increased riparian vegetation (cover, height, productivity, biomass, and abundance), riparian bird abundance, and young-of-the-year fishes. Results for channel depth, instream substrate, mesohabitats, water depth, juvenile and adult fishes, and macroinvertebrates showed no consistent response to exclusion. Project success was influenced by the time since exclusion; whether there was complete exclusion or continued grazing; and local climate, geology, and soils. Apart from bank erosion and stability, most of the physical and biological metrics took more than a decade to respond to livestock exclusion. However, coupling exclusion with planting and other restoration measures decreased the recovery time. Complete exclusion of livestock produced more consistent improvements in riparian condition and other metrics than rest–rotation or other grazing management strategies. Understanding how physical and biological metrics respond to livestock exclusion will require (1) focused, long-term studies using before–after or before–after, control–impact designs; and (2) monitoring of metrics that most consistently respond to exclusion. Ultimately, the design of exclusions should be driven by local climate, geology, biophysical conditions, and management history. Our results highlight the need for watershed-scale approaches to excluding livestock from broad areas and the need for implementation monitoring to ensure that fencing and other exclusion measures continue to exclude livestock and produce the desired responses.

Krall, Michelle↗

DOE OSTI2021

Developing a strategy for the national coordinated soil moisture monitoring network

Soil moisture is a critical land surface variable, affecting a wide variety of climatological, agricultural, and hydrological processes. Determining the current soil moisture status is possible via a variety of methods, including in situ monitoring, remote sensing, and numerical modeling. Although all of these approaches are rapidly evolving, there is no cohesive strategy or framework to integrate these diverse information sources to develop and disseminate coordinated national soil moisture products that will improve our ability to understand climate variability. The National Coordinated Soil Moisture Monitoring Network initiative has developed a national strategy for network coordination with NOAA's National Integrated Drought Information System. The strategy is currently in review within NOAA, and work is underway to implement the initial milestones of the strategy. This update reviews the goals and steps being taken to establish this national-scale coordination for soil moisture monitoring in the United States.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

The Dynamic Networks Experiments: Virtual Experiments to Quantify Gains in Nuclear Explosion Monitoring

We describe an ongoing series of virtual experiments conducted collaboratively by four United States National Laboratories: Sandia National Laboratories, Los Alamos National Laboratory, Lawrence Livermore National Laboratory, and Pacific Northwest National Laboratory. These Dynamic Network Experiments (DNEs) provide an experimental framework to evaluate the potential impact of new research tools on nuclear explosion monitoring. The second DNE (DNE2), completed in 2024, exploited waveform data (seismic, infrasound, and electromagnetic) that was recorded by multi-modal sensors within and near the Nevada National Security Site and synthetic radionuclide signatures over multiple time periods. During the execution of DNE2, we processed and analyzed data through a multi-stage event processing pipeline that ingested raw data, performed quality control, detected signals, built events from these signals, located these events, and characterized the events’ source types and sizes. For each stage and over the entire event processing pipeline, we evaluated performance changes by comparing the performance of new data processing methods, models, and algorithms against a baseline. We also performed an additional execution phase to assess event processing pipeline function, speed, and efficiency against that of an expert analyst, including computational and manual efforts. Finally, we assessed the impact and effort of modern computing infrastructure on the monitoring pipeline. This paper describes key elements of the DNEs, from formulation through execution, as demonstrated in DNE2. The DNEs introduce several novel concepts to quantitatively measure the potential impact of new methods on explosion monitoring, including the collaborative design of multi-modal datasets, performance and logistical metrics, and integrated analyses.

42 ENGINEERING↗

DOE OSTI2020

Groundwater characterization and monitoring at a complex industrial waste site using electrical resistivity imaging

A contaminated industrial waste site in Washington State (USA) containing buried, metallic-waste storage tanks, pipes, and wells, was evaluated to determine the feasibility of monitoring groundwater remediation activities associated with an underlying perched aquifer system using electrical resistivity tomography. The perched aquifer, located ~65 m below ground surface and ~10 m above the regional water table, contains high concentrations of nitrate, uranium, and other contaminants of concern from past tank leaks and intentional releases of wastes to surface disposal sites. The extent of the perched water aquifer is not well known, and the effectiveness of groundwater extraction for contaminant removal is uncertain, so supplemental characterization and monitoring technologies are being evaluated. Numerical simulations of subsurface flow and contaminant transport were performed with a highly resolved model of the hydrogeologic system and waste site infrastructure, and these simulations were used as the physical basis for electrical resistivity tomography modeling. The modeling explicitly accounted for metallic infrastructure at the site. The effectiveness of using surface electrodes versus surface and horizontal subsurface electrodes, for imaging groundwater extraction from the perched water aquifer, was investigated. Although directional drilling is a mature technology, its use for electrode emplacement in the deep subsurface under a complex industrial waste site via horizontal wells has not yet been demonstrated. Results from this study indicate that using horizontal subsurface electrode arrays could significantly improve the ability of electrical resistivity tomography to image deep subsurface features and monitor remediation activities under complex industrial waste sites.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

Out-of-Distribution Detection and Radiological Data Monitoring Using Statistical Process Control

Abstract Machine learning (ML) models often fail with data that deviates from their training distribution. This is a significant concern for ML-enabled devices as data drift may lead to unexpected performance. This work introduces a new framework for out of distribution (OOD) detection and data drift monitoring that combines ML and geometric methods with statistical process control (SPC). We investigated different design choices, including methods for extracting feature representations and drift quantification for OOD detection in individual images and as an approach for input data monitoring. We evaluated the framework for both identifying OOD images and demonstrating the ability to detect shifts in data streams over time. We demonstrated a proof-of-concept via the following tasks: 1) differentiating axial vs. non-axial CT images, 2) differentiating CXR vs. other radiographic imaging modalities, and 3) differentiating adult CXR vs. pediatric CXR. For the identification of individual OOD images, our framework achieved high sensitivity in detecting OOD inputs: 0.980 in CT, 0.984 in CXR, and 0.854 in pediatric CXR. Our framework is also adept at monitoring data streams and identifying the time a drift occurred. In our simulations tracking drift over time, it effectively detected a shift from CXR to non-CXR instantly, a transition from axial to non-axial CT within few days, and a drift from adult to pediatric CXRs within a day—all while maintaining a low false positive rate. Through additional experiments, we demonstrate the framework is modality-agnostic and independent from the underlying model structure, making it highly customizable for specific applications and broadly applicable across different imaging modalities and deployed ML models.

Zamzmi, Ghada↗