Prior Methods Research Articles

A Curriculum-Style Self-Training Approach for Source-Free Semantic Segmentation.

Source-free domain adaptation has developed rapidly in recent years, where the well-trained source model is adapted to the target domain instead of the source data, offering the potential for privacy concerns and intellectual property protection. However, a number of feature alignment techniques in prior domain adaptation methods are not feasible in this challenging problem setting. Thereby, we resort to probing inherent domain-invariant feature learning and propose a curriculum-style self-training approach for source-free domain adaptive semantic segmentation. In particular, we introduce a curriculum-style entropy minimization method to explore the implicit knowledge from the source model, which fits the trained source model to the target data using certain information from easy-to-hard predictions. We then train the segmentation network by the proposed complementary curriculum-style self-training, which utilizes the negative and positive pseudo labels following the curriculum-learning manner. Although negative pseudo-labels with high uncertainty cannot be identified with the correct labels, they can definitely indicate absent classes. Moreover, we employ an information propagation scheme to further reduce the intra-domain discrepancy within the target domain, which could act as a standard post-processing method for the domain adaptation field. Furthermore, we extend the proposed method to a more challenging black-box source model scenario where only the source model's predictions are available. Extensive experiments validate that our method yields state-of-the-art performance on source-free semantic segmentation tasks for both synthetic-to-real and adverse conditions datasets.

Transformer-Based Visual Segmentation: A Survey.

Visual segmentation seeks to partition images, video frames, or point clouds into multiple segments or groups. This technique has numerous real-world applications, such as autonomous driving, image editing, robot sensing, and medical analysis. Over the past decade, deep learning-based methods have made remarkable strides in this area. Recently, transformers, a type of neural network based on self-attention originally designed for natural language processing, have considerably surpassed previous convolutional or recurrent approaches in various vision processing tasks. Specifically, vision transformers offer robust, unified, and even simpler solutions for various segmentation tasks. This survey provides a thorough overview of transformer-based visual segmentation, summarizing recent advancements. We first review the background, encompassing problem definitions, datasets, and prior convolutional methods. Next, we summarize a meta-architecture that unifies all recent transformer-based approaches. Based on this meta-architecture, we examine various method designs, including modifications to the meta-architecture and associated applications. We also present several specific subfields, including 3D point cloud segmentation, foundation model tuning, domain-aware segmentation, efficient segmentation, and medical segmentation. Additionally, we compile and re-evaluate the reviewed methods on several well-established datasets. Finally, we identify open challenges in this field and propose directions for future research.

Recovering NDVI over lake surfaces: Initial insights from CYGNSS data enhanced by ERA-5 inputs

The escalating water pollution in many lakes has led to more frequent occurrences of algal bloom disasters in recent decades. The severity of these disasters can be assessed through remote sensing techniques, specifically using the Normalized Difference Vegetation Index (NDVI) for measurement. However, NDVI observations using optical sensors are often affected by cloud and fog in areas with numerous water bodies, such as Taihu Lake. Sensors operating in the microwave band can effectively mitigate this issue, particularly the emerging Global Navigation Satellite System Reflectometry (GNSS-R), which offers high temporal resolution and cost-effectiveness. In this paper, we propose a new method to recover lake-surface NDVI on cloudy days, utilizing GNSS-R observables and auxiliary meteorological data in conjunction with a machine learning regression algorithm called Bagging Tree. We also examine the effective range of GNSS-R data within this application scenario. Meanwhile, the Weighted Linear Regression-Laplacian Prior Regulation Method (WLR-LPRM) image gap-filling algorithm is used as a benchmark to evaluate recovery accuracy. The regression coefficient of NDVI retrieved using the proposed method is 0.95, with a root mean square error (RMSE) of 0.021 and a mean absolute error (MAE) of 0.010. Compared to the previous work on GNSS-R algal bloom detection with overall accuracy of 0.82, this work shows significant improvement in both accuracy and utility. The recovery of lake surface NDVI provides detailed insights into algal blooms, including quantifiable metrics such as the amount and spatial distribution, which are crucial for effective monitoring and management. Additionally, the recovered image textures exhibit high clarity and closely resemble the reference NDVI images. Experimental evaluation using simulated and actual cloud blocks indicates the model’s robustness to recover NDVI under varying cloud cover conditions. In summary, this study demonstrates the capability of GNSS-R aided by supplementary data for recovering missing NDVI values on lake surfaces when optical observations are absent for the first time.

Higher Socioeconomic Status is Associated with Greater Rates of Surgical Resource Utilization Prior to Spine Fusion Surgery

Previous research has demonstrated an association between socioeconomic status (SES) and patient health, specifically noting that patients of lower SES have poor health outcomes. Understanding how social factors, including socioeconomic status (SES), relate to disparities in health outcomes is critical to closing gaps in equitable care to patients. While several studies have examined the effect of SES on postoperative spine outcomes, there is limited spine literature evaluating SES in the context of barriers to spine care. The primary objective of this study was to determine if socioeconomic status is associated with resource utilization prior to spine surgery consultation. As part of a sub-analysis, this paper also explores the effect of other social factors on pre-visit resource utilization. Retrospective cohort study. Adult patients who underwent elective cervical or lumbar spinal fusion between 2020 and 2021. Pre-visit resource utilization including 1) epidural steroid injection, 2) opioid use, 3) physical therapy, 4) prior spine surgeon, and 5) prior spine surgery METHODS: Each patient was assigned a "distressed score" using the Distressed Communities Index (DCI) and a socioeconomic status (SES) score using the Social Vulnerability Index (SVI) based on their zip code. Patient charts were manually reviewed to collect data regarding pre-visit resource utilization. The cohort was analyzed based on DCI quintile and SVI quartile. Additional analyses were conducted based on marital status and race. Our study included 996 patients in the final analysis. Based on DCI, patients from prosperous communities were more likely to have previously visited a spine surgeon (13.2% (prosperous) vs. 7.58% vs. 6.92% vs. 9.09% vs. 3.70% (distressed), p=0.015) and to have had prior spine surgery (11.1% (prosperous) vs. 9.57% vs. 9.09% vs. 2.52% vs. 6.36% (distressed), p=0.015). Similarly, when evaluated based on SES SVI, patients who lived in a low-risk community were more likely to have previously visited a spine surgeon (13.0% low-risk vs. 7.26% low-medium risk vs. 16.9% medium-high risk vs. 10.6% high risk, p=0.049) and to have had prior spine surgery (13.0% low-risk vs. 7.26% vs. 16.9% vs. 10.6% high risk, p=0.030). When evaluated based on marital status, there was no difference in any resource utilization. Non-black and non-white patients were more likely to have tried physical therapy compared to their black and white counterparts (76.9% (other) vs. 60.9% (Black) vs. 54.3% (White), p=0.026). This study examined the relationship between socioeconomic status and resource utilization and found a positive correlation between higher social standing and access to spine surgery and spine surgeons. These findings demonstrate a propensity for earlier evaluation of spine-related conditions among patients from prosperous communities compared to patients from less prosperous communities.

Outlier detection of clustered functional data with image and signal processing applications by archetype analysis.

In this study, we introduce an innovative methodology for anomaly detection of curves, applicable to both multivariate and multi-argument functions. This approach distinguishes itself from prior methods by its capability to identify outliers within clustered functional data sets. We achieve this by extending the recent AA + kNN technique, originally designed for multivariate analysis, to functional data contexts. Our method demonstrates superior performance through a comprehensive comparative analysis against twelve state-of-the-art techniques, encompassing simulated scenarios with either a single functional cluster or multiple clusters. Additionally, we substantiate the effectiveness of our approach through its application in three distinct computer vision tasks and a signal processing problem. To facilitate transparency and replication of our results, we provide access to both the code and the datasets used in this research.

Deep Learning-Based Student Engagement Classification in Online Learning

Online education has gained significant popularity during the COVID-19 pandemic. The availability of massive open online courses has further strengthened the online learning environment. One of the considerable challenges in online learning environments is to measure the learner’s engagement to meet the educational objectives. This paper addresses the challenge and proposes a convolutional neural network-based architecture that extracts discriminative features from the face (Affective) and the upper body (Behavioral) of the learner and classifies the engagement. The performance of the proposed architecture is evaluated on the publicly available Dataset of Affective States In E-Environments (DAiSEE) and the learning-centered affective state dataset curated from open-source datasets. The experimental results demonstrate that the proposed methodology with affective and behavioral features improves the engagement measurement results. The proposed method outperforms the state-of-the-art in terms of Unweighted Average Recall (UAR), Unweighted Average Precision (UAP), and Unweighted Average F1 (UAF1) with 79.14%, 49.62%, and 55.29% values, respectively. It is threefold more computationally efficient than the previous state-of-the-art method. The proposed method also improves the accuracy of the curated dataset by 2.07% of the prior method.

A Computational Mechanism for Encoding, Recalling and Forgetting of Real-Life Episodic Events

Episodic memory is one of the crucial dimensions of a human memory system; by imitating them, an agent can achieve human intelligence. A memory of a particular occurrence is referred to as an episodic memory. Due to the fact that every individual has a unique perspective and experience of a certain event, the episodic memories that they have of that event are also unique to each individual. It contains your recollections of individual occurrences, as well as personal information, generic occurrences and pictures of precise times in time you have experienced. To achieve human cognitive abilities, this paper presents a computational mechanism that encodes and forgets the real-life episodic events. The proposed model is based on role entity bindings that encode the real-life conceptual events. Furthermore, the model incorporates a computational gradual forgetting mechanism to achieve higher performance while controlling memory consumption over time. According to the findings of our assessments of the agents, forgetfulness lessens the impact of out-of-date information and states that are not frequently visited on the policies that are generated by the episodic control system. To validate the proposed model, an experimental study is conducted, where the evaluation is done based on event recalling’s hit ratio. The model produces robust performance and got an enhanced hit ratio in event recalling concerning prior methods.

Divergent neurodegenerative patterns: comparison of fluorodeoxyglucose-PET and MRI-based Alzheimer’s disease subtypes

Abstract [18F] fluorodeoxyglucose (FDG)-PET and MRI are key imaging markers for neurodegeneration in Alzheimer's disease. It has been well established that parieto-temporal hypometabolism on FDG-PET is closely associated with medial temporal atrophy on MRI in Alzheimer's disease. Substantial biological heterogeneity, expressed as distinct subtypes of hypometabolism or atrophy patterns, has been previously described in Alzheimer's disease using data-driven and hypothesis-driven methods. However, the link between these two imaging modalities has not yet been explored in the context of Alzheimer's disease subtypes. To investigate this link, the current study utilised FDG-PET and MRI scans from 180 amyloid-beta positive Alzheimer's disease dementia patients, 339 amyloid-beta positive mild cognitive impairment and 176 amyloid-beta negative cognitively normal controls from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Random forest hierarchical clustering, a data-driven model for identifying subtypes, was implemented in the two modalities: one with standard uptake value ratios and the other with grey matter volumes. Five hypometabolism- and atrophy-based subtypes were identified, exhibiting both cortical-predominant and limbic-predominant patterns although with differing percentages and clinical presentations. Three cortical-predominant hypometabolism subtypes found were: Cortical Predominant (32%), Cortical Predominant+ (11%), Cortical Predominant posterior (8%); and two limbic-predominant hypometabolism subtypes: Limbic Predominant (36%) and Limbic Predominant frontal (13%). In addition, little atrophy (minimal) and widespread (diffuse) neurodegeneration subtypes were observed from the MRI data. The five atrophy subtypes were found: Cortical Predominant (19%), Limbic Predominant (27%), Diffuse (29%), Diffuse+ (6%) and Minimal (19%). Inter-modality comparisons showed that all FDG-PET subtypes displayed medial temporal atrophy, whereas the distinct MRI subtypes showed topographically similar hypometabolic patterns. Further, allocations of FDG-PET and MRI subtypes were not consistent when compared at an individual-level. Additional analysis comparing the data-driven clustering model with prior hypothesis-driven methods showed only partial agreement between these subtyping methods. FDG-PET subtypes had greater differences between limbic-predominant and cortical-predominant patterns and MRI subtypes had greater differences in severity of atrophy. In conclusion, this study highlighted that Alzheimer's disease subtypes identified using both FDG-PET and MRI capture distinct pathways showing cortical- versus limbic-predominance of neurodegeneration. However, the subtypes do not share a bidirectional relationship between modalities and are thus not interchangeable.

Effi

We delve into the realm of offline meta-reinforcement learning (OMRL), a practical paradigm in the field of reinforcement learning that leverages offline data to adapt to new tasks. While prior approaches have not explored the utilization of context-based dynamical models to tackle OMRL problems, our research endeavors to fill this gap. Our investigation uncovers shortcomings in existing context-based methods, primarily related to distribution shifts during offline learning and challenges in establishing stable task representations. To address these issues, we formulate the problem as Hidden-Parameter MDPs and propose a framework for effective model adaptation using meta-models plus latent variables, which is inferred by the transformer-based system recognition module trained in an unsupervised fashion. Through extensive experimentation encompassing diverse simulated robotics and control tasks, we validate the efficacy of our approach and demonstrate its superior generalization ability compared to existing schemes, and explore multiple strategies for obtaining policies with personalized models. Our method achieves a model with reduced prediction error, outperforming previous methods in policy performance, and facilitating efficient adaptation when compared to prior dynamic model generalization methods and OMRL algorithms.

ParSEL: Parameterized Shape Editing with Language

The ability to edit 3D assets with natural language presents a compelling paradigm to aid in the democratization of 3D content creation. However, while natural language is often effective at communicating general intent, it is poorly suited for specifying exact manipulation. To address this gap, we introduce ParSEL, a system that enables controllable editing of high-quality 3D assets with natural language. Given a segmented 3D mesh and an editing request, ParSEL produces a parameterized editing program. Adjusting these parameters allows users to explore shape variations with exact control over the magnitude of the edits. To infer editing programs which align with an input edit request, we leverage the abilities of large-language models (LLMs). However, we find that although LLMs excel at identifying the initial edit operations, they often fail to infer complete editing programs, resulting in outputs that violate shape semantics. To overcome this issue, we introduce Analytical Edit Propagation (AEP), an algorithm which extends a seed edit with additional operations until a complete editing program has been formed. Unlike prior methods, AEP searches for analytical editing operations compatible with a range of possible user edits through the integration of computer algebra systems for geometric analysis. Experimentally, we demonstrate ParSEL's effectiveness in enabling controllable editing of 3D objects through natural language requests over alternative system designs.

Stripe Embedding: Efficient Maps with Exact Numeric Computation

We consider the fundamental problem of injectively mapping a surface mesh with disk topology onto a boundary constrained convex domain. We start from the basic observation that mapping a strip of triangles onto a rectangular shape always yields a valid embedding, if the vertices that bound the strip are sorted coherently along the sides of the rectangle. Based on this intuition, we propose a straightforward algorithm, called Stripe Embedding, that operates by decomposing the input mesh into a set of triangle strips and then embeds each strip into the target domain by means of linear interpolation between two previously embedded vertices. Thanks to its simplicity, Stripe Embedding is extremely efficient and permits to switch to an exact implementation without almost increasing its running times. Stripe Embedding is up to three orders of magnitude faster than the Tutte embedding for same numerical model and, even when implemented with costly rational numbers, it is faster than any floating point implementation of prior methods at any scale.

Sequential Adaptive Design Method for Incorporating External Data.

External data (e.g., real-world data (RWD) and historical data) have become more readily available. This has led to rapidly increasing interest in exploring and evaluating ways of utilizing external data to facilitate traditional clinical trials (TCT), especially for rare diseases with high unmet medical needs where a TCT would be impractical and/or unethical. In this article, we focus on hybrid studies that incorporate external data into randomized clinical trials to augment the control arm and explore a complex innovative design. A sequential adaptive design conducts multiple interim assessments to improve the accuracy of estimates of agreement between external data and current data. At each interim assessment, we apply the inverse probability weighted power prior (IPW-PP) method to adaptively borrow information from external data to account for confounding and heterogeneity. The randomization ratio is dynamically adjusted during the interim assessment based on accumulatively augmented information to reduce the sample size of the current trial. Additionally, the proposed design can be extended to allowinterim analyses for early efficacy/futility stopping, that is, early assessment of trial success or failure based on accumulated data, potentially reducing ineffective treatment exposure and unnecessary time and resources. The performance of the proposed method and design is evaluated via extensive simulation studies. The sequential adaptive design and IPW-PP approach having desirable properties are implemented.

Urban Traffic Simulation with Shared Mobility Services: An Approach Using Spatiotemporal Network Kernel Density Estimation and MATSim

The popularity of smartphones and 4G/5G network have enabled various novel transportation modes based on shared mobility, such as app-based ride-hailing services (e.g., Uber and Lyft) and shared micromobility services (e.g., Veo and Gotcha). However, little is known about to what degree their operations impact urban traffic, which is important for transportation planning and policy making. These companies seldom share their ride data due to business and user privacy reasons, and they are young and still exploring and growing their market to new locations. Recently, transportation engineering researchers began to collect data in large cities trying to understand the transportation impacts of shared mobility services, but (1) there does not exist a general analytics framework applicable to any city, and (2) the studies were based on historical data and cannot project the future easily to catch up with the rapid development of shared mobility services. In this paper, we introduce a general framework for multi-modal urban traffic analytics. The goal is to build a digital twin of the transportation in a city, i.e., an accurate agent-based transportation simulation model, based on a medium-sized dataset of the interested transportation modes collected by the research group combined with other open data sources such as US Census Bureau. With this digital twin, transportation engineering researchers can flexibly analyze the impact of shared mobility services under different scenarios, such as “if the number of Uber drivers doubles” or “if the number of deployed e-scooters doubles”. The digital twin may also enable new opportunities, such as being an environment for learning policies with reinforcement learning. Our framework consists of three steps: (1) fitting the spatiotemporal distribution of the shared mobility rides on the underlying road network, (2) generating the travel day-plans of the entire urban population based on the learned spatiotemporal ride distribution and user-specified parameters, and (3) configuring an agent-based simulation software such as MATSim to execute the generated realistic day-plans, which provides detailed transportation data for analytics. At the core of our approach is a new spatiotemporal network kernel density estimation (KDE) method that fixes the flaw of prior methods where the contributions of different data samples are not equal. We also propose a crowdsourcing method to collect app-based ride data that is easy to carry out in any city. Using Birmingham, AL as an example, we demonstrate how our framework can be applied to help transportation engineering researchers analyze the impacts of Uber ride-hailing and Veo e-scooter services.

Image forgery localization integrating multi-scale and boundary features

Abstract Image forgery localization identifies tampered regions within an image by extracting distinctive forgery features. Current methods mainly use convolutional neural networks (CNNs) to extract features. However, CNNs’ limited receptive field emphasizes local features, impeding the global modeling of crucial lower-level features like edges and textures, leading to decreased precision. Moreover, prior methods use pyramid networks for multi-scale feature extraction but show deficiencies in multi-scale and interlayer modeling, leading to inadequate multi-scale information representation and limiting flexibility to tampered regions of varying sizes. To address these issues, this paper proposes a Transformer-based model integrating multi-scale and boundary features. The model employs a Pyramid Vision Transformer as the encoder, using self-attention over convolution to enhance global context modeling. Building on this, the model incorporates a multi-scale feature enhancement module that enriches forgery features by paralleling various convolutional layers. Features at various encoder stages are integrated through a cross-stage interaction module, enabling multi-level feature correlation for a strong feature representation. Furthermore, the model includes a forgery boundary information-guided branch, which focuses precisely on tampered region structures without introducing irrelevant noise. Experiments demonstrate that our model surpasses previous methods in localization accuracy, with F1 and AUC improving by 8.5% and 2.2% in pre-training, respectively.

Rapid and General Amination of Aryl Boronic Acids and Esters Using O-(Diphenylphosphinyl)hydroxylamine (DPPH).

O-(Diphenylphosphinyl)hydroxylamine (DPPH) is a general reagent for the conversion of (hetero)aryl boronic acids and esters to primary anilines. The transformation proceeds rapidly at rt and exhibits a broad substrate scope and exceptional functional-group tolerance. In terms of rate, the reaction is relatively insensitive to the electronic properties of the substrate, in contrast to similar reactions using electrophilic amination reagents such as hydroxylamine-O-sulfonic acid. Consequently, this protocol is particularly useful for accessing electron-deficient (hetero)aryl anilines, which had been challenging to prepare using prior methods.

Abstract B063: DiMMER: A robust computational pipeline for differential methylation marker evaluation in R of cell-free DNA fragments

Abstract Background: Decoding the origins of cell-free DNA (cfDNA) released from dying cells in a liquid biopsy sample offers the potential to provide insight into the dynamic, organism-wide changes reflective of health and disease, making cfDNA an ideal target for serial, minimally invasive monitoring of disease-related changes. To this end, cell-type specific DNA methylation patterns offer a promising target to facilitate tissue of origin analysis, yet limited methods exist to identify differentially methylated regions that distinguish cell-types. Methods: We develop a robust differentially methylated marker region identification pipeline, DiMMER, that leverages cfDNA fragment level information of neighboring, co-regulated CpG sites on methylome-wide sequencing reads. Compared to prior methods that utilize simple heuristics in one-vs-all average methylation rate comparisons, we implement an individual pairwise statistical test procedure across all cell-types or groups under consideration. We assess the cell-specific nature of identified differentially methylated marker regions by generating in-silico mixtures from known cell-type of origin at each region, and calculating the area under the receiver operating characteristic curve (AUROC). We utilize our differentially methylated marker finding pipeline to identify melanocyte specific methylation marker regions and assess their functional role through annotations. We show these cell-specific regions are conserved in melanoma cell-lines and are distinct from the aberrant changes in a cancer context. Results: We identified the most cell-type specific differentially methylated marker regions across 24 distinct cell-type groups using our pipeline. Compared to a simpler one-vs-all heuristic, we demonstrate improved cell-type specificity evidenced by a higher AUROC on average. Using our pipeline we identified melanocyte-specific marker regions, which were commonly found in intronic and promoter regions of genes related to melanocyte development and differentiation. When comparing methylation patterns between melanocytes and melanoma cell-lines, we demonstrate melanocyte specific methylation marks are conserved through the transformation to malignant melanoma. Conclusion: As methylome-wide sequencing methods continue to rapidly develop and more cell-specific methylation data is generated, there is unmet need for more statistically robust differentially methylated marker finding tools. Here we present one such tool, DiMMER, and demonstrate the potential utility of identifying such regions to be used to assess organ-specific load as a measure of residual disease, as opposed to traditional circulating tumor DNA quanitification. Citation Format: Arthur P McDeed, Sidharth S Jain, Megan E McNamara, Amber Alley, Anton Wellstein, Jaeil Ahn. DiMMER: A robust computational pipeline for differential methylation marker evaluation in R of cell-free DNA fragments [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr B063.

ScDOT: optimal transport for mapping senescent cells in spatial transcriptomics

The low resolution of spatial transcriptomics data necessitates additional information for optimal use. We developed scDOT, which combines spatial transcriptomics and single cell RNA sequencing to improve the ability to reconstruct single cell resolved spatial maps and identify senescent cells. scDOT integrates optimal transport and expression deconvolution to learn non-linear couplings between cells and spots and to infer cell placements. Application of scDOT to lung spatial transcriptomics data improves on prior methods and allows the identification of the spatial organization of senescent cells, their neighboring cells and novel genes involved in cell-cell interactions that may be driving senescence.

Enhancing power transformer health assessment through dimensional reduction and ensemble approaches in Dissolved Gas Analysis

AbstractTransformer health analysis using Dissolved Gas Analysis is crucial for diagnosing power transformer faults. This paper proposes an innovative approach to diagnose power transformer faults by integrating machine learning algorithms with Ensemble techniques. The method involves fusing reduced dimensional input features through Principal Component Analysis with Ensemble techniques such as Bagging, Decorate, and Boosting. Various machine learning algorithms, including Decision Tree (DT), K‐Nearest Neighbours, Radial Basis Function Network, and Support Vector Machine, are employed in conjunction with Ensemble techniques. The long short‐term memory algorithm was used to create synthetic data to solve the issue of data imbalance. A dataset of 683 samples is used in the study for training, testing, validation, and comparison with current techniques. The results highlight the effectiveness of Ensemble techniques, particularly Boosting, which demonstrates superior performance across all classification algorithms. The Boosting with DT algorithm achieves an impressive accuracy of 98.32%, surpassing alternative methods. In validation, the proposed Boosting Ensemble technique outperforms various approaches, showcasing its diagnostic accuracy and superiority over alternative methods. The research emphasises the model's effectiveness in smoothing input vectors, enhancing harmony with ensemble techniques, and overcoming limitations in prior methods.

A framework to model charge sharing and pulse pileup for virtual imaging trials of photon-counting CT

Objective.This study describes the development, validation, and integration of a detector response model that accounts for the combined effects of x-ray crosstalk, charge sharing, and pulse pileup in photon-counting detectors.Approach.The x-ray photon transport was simulated using Geant4, followed by analytical charge sharing simulation in MATLAB. The analytical simulation models charge clouds with Gaussian-distributed charge densities, which are projected on a 3×3 pixel neighborhood of interaction location to compute detected counts. For pulse pileup, a prior analytical method for redistribution of energy-binned counts was implemented for delta pulses. The x-ray photon transport and charge sharing components were validated using experimental data acquired on the CdTe-based Pixirad-1/Pixie-III detector using monoenergetic beams at 26, 33, 37, and 50 keV. The pulse pileup implementation was verified with a comparable Monte Carlo simulation. The model output without pulse pileup was used to generate spatio-energetic response matrices for efficient simulation of scanner-specific photon-counting CT (PCCT) images with DukeSim, with pulse pileup modeled as a post-processing step on simulated projections. For analysis, images for the Gammex multi-energy phantom and the XCAT chest phantom were simulated at 120 kV, both with and without pulse pileup for a range of doses (27-1344 mAs). The XCAT images were evaluated qualitatively at 120 mAs, while images for the Gammex phantom were evaluated quantitatively for all doses using measurements of attenuation coefficients and Calcium concentrations.Main results.Reasonable agreement was observed between simulated and experimental spectra with Mean Absolute Percentage Error Values (MAPE) between 10%and 31%across all incident energies and detector modes. The increased pulse pileup from increased dose affected attenuation coefficients and calcium concentrations, with an effect on calcium quantification as high as MAPE of 28%.Significance.The presented approach demonstrates the viability of the model for enabling VITs to assess and optimize the clinical performance of PCCT.

Comparison of Multiple Methods for Supraglacial Melt-Lake Volume Estimation in Western Greenland During the 2021 Summer Melt Season

Supraglacial melt-lakes form and evolve along the western edge of the Greenland Ice Sheet and have proven to play a significant role in ice sheet surface hydrology and mass balance. Prior methods to quantify melt-lake volume have relied upon Landsat-8 optical imagery, available at 30 m spatial resolution but with temporal resolution limited by satellite overpass times and cloud cover. We propose two novel methods to quantify the volume of meltwater stored in these lakes, including a high-resolution surface DEM (ArcticDEM) and an ablation model using daily averaged automated weather station data. We compare our methods to the depth-reflectance method for five supraglacial melt-lakes during the 2021 summer melt season. We find agreement between the depth-reflectance and DEM lake infilling methods, within +/−15% for most cases, but our ablation model underproduces by 0.5–2 orders of magnitude the volumetric melt needed to match our other methods, and with a significant lag in meltwater onset for routing into the lake basin. Further information regarding energy balance parameters, including insolation and liquid precipitation amounts, is needed for adequate ablation modelling. Despite the differences in melt-lake volume estimates, our approach in combining remote sensing and meteorological methods provides a framework for analysis of seasonal melt-lake evolution at significantly higher spatial and temporal scales, to understand the drivers of meltwater production and its influence on the spatial distribution and extent of meltwater volume stored on the ice sheet surface.