Background Model Research Articles

Lymphocyte radiosensitivity: An extension to the linear-quadratic model?

Background and purposeThe linear-quadratic (LQ) model has been pivotal for evaluating the effects of radiation on cells, but it is primarily characterized by linear responses, which has exhibited limitations when applied to lymphocyte data. The present research aims to address these limitations and to explore an alternative model extended from the conventional LQ model. Materials and methodsLiterature providing lymphocyte counts from assays investigating apoptosis and survival after in vitro irradiation was selected. To address the nonlinearity in lymphocyte responses to radiation, we developed a saturation model characterized by a negative exponential relationship between radiation dose and cellular response. We compared the performance of this saturation model against that of conventional models, including the LQ model and its variants (linear model LM and linear-quadratic-cubic model LQC), as well as the repair-misrepair (RMR) model. The models were evaluated based on prediction-residual plots, residual standard errors, and the Akaike information criterion (AIC). We applied the saturation model to two additional datasets: (1) a dataset from the existing literature that assessed stimulated and unstimulated human lymphocytes exposed to gamma irradiation in vitro and (2) a novel dataset involving T lymphocytes from rodent spleens after exposure to various radiation types (X-rays and protons). ResultsThe literature (n = 15 out of 2342) showed that lymphocyte apoptosis varies with dose, time and experimental conditions. The saturation model had a lower AIC of 718 compared to the LM, LQ, LQC and RMR models (AIC of 728, 720, 720 and 734, respectively). The saturation model had a lower residual error and more consistent error distribution. Integrating time as a covariate, the saturation model also had a better AIC for demonstrating time-dependent variations in lymphocyte responses after irradiation. For datasets involving unstimulated lymphocytes before irradiation, the saturation model provided a more accurate fit than did the LM, LQ, and RMR models. In these cases, the fit of the saturation model was comparable to that of the LQC model but offered an advantage when extrapolating to higher doses, where the LQC model might underestimate survival. For stimulated lymphocytes, which are radioresistant, all the models approximated the LM. Both the LQ and saturation models indicated greater radiosensitivity to protons in vitro. ConclusionThe new “saturation model” performed better than the LQ model in quantifying lymphocyte apoptosis and survival, estimating time dependency and assessing the role of radiation modalities or lymphocyte stimulation. Further experiments are warranted to experimentally explore the validity of the saturation model as a promising alternative in the clinical setting.

Measurement of background radiations and spectra of X-ray sources using low-cost stratospheric balloon missions

In small balloon missions, some in situ radiation is added to the targeted strong persistent astronomical sources (such as, the Sun, Crab, Cygnus X-1, etc.) or transient sources, such as solar flares, gamma-ray bursts, etc., during their detection, i.e., the background radiation. For measurement and modeling of background radiation, we use the data of the low-cost, small, stratospheric, high-altitude balloon-borne missions conducted by Indian Centre for Space Physics. In this paper, we show several models to estimate the background counts where the strong X-ray sources are detected at the altitude range of ∼35–42 km even in the absence of onboard pointing system and severe atmospheric absorption. For all data analysis purposes, the photons detected in the detector are in the energy range of 10–80 keV. Moreover, we found that the observed data is within ±1σ of the calculated background model. Finally, using these background models, we estimate the spectra and luminosity of the Crab nebula and the black hole candidate Cygnus X-1, which agree with the previously published results.

Automated detection of pavement distress based on enhanced YOLOv8 and synthetic data with textured background modeling

Regular inspections of pavement distress are essential for accident prevention, and deep learning based algorithms have been developed to ensure high accuracy of inspections. However, the lack of available data is a critical challenge for existing algorithms. To address this issue, an elaborate image synthesis strategy is proposed. By combining textured background modeling for real pavement and Unreal Engine-based distress block stitching technique, high-resolution virtual images that are indistinguishable from real images are generated, including five types of pavement distresses. In addition, an enhanced YOLOv8 network utilizing synthetic data is designed in this paper. The enhanced YOLOv8 network is embedded with the Squeeze and Excitation attention module, and the Swin Transformer module, which are designed to distinguish different types of pavement distress accurately and suppress the interference of complex backgrounds (noise, shadow, blur). The results show that the performance of the algorithms trained based on appropriately ratios of synthetic data (2:1 ratio of virtual to real) improves by more than 10% compared with no virtual data. The enhanced YOLOv8 network achieves a Mean Average Precision (MAP) of 94.8% for transverse cracks, longitudinal cracks, cross-cracks, alligator cracks, and potholes, which is better than seven existing object detection models. The proposed image synthesis method can contribute to improving the accuracy and reliability of pavement inspection and alleviate the reliance on large number of distress samples collection.

Anomaly detection method for urban river water quality based on background model prediction using three-dimensional fluorescence

Abstract Three-dimensional fluorescence spectroscopy has great potential for detecting water quality anomalies in urban rivers and protecting against organic pollution. However, current detection methods inadequately address critical application scenarios, such as fluctuations in river water background, low concentration of pollutants, and the fluorescence peaks overlap between pollutants and background. In this paper, a fluorescence spectrum feature extraction method which is effective for the above scenarios was proposed. The proposed method involves a sequential process. First, the original spectrum undergoes preprocessing using a novel method. Next, an alternating residual tri-linearization technique is applied to establish a predictive model for river water spectrum changes. Subsequently, the background model is utilized for spectrum decomposition and reconstruction of the test sample. This reconstructed spectrum is then used to derive the residual spectrum by comparison with the original. Finally, frequency domain features are extracted from the residual spectrum to enable classification, while the background model undergoes real-time updates. In the three meaningful scenarios mentioned above, the accuracy of the proposed method for anomaly detection reached 99%, 82%, and 98%, respectively. Our accuracy is higher than several typical benchmark methods.

Calibrating gravitational-wave search algorithms with conformal prediction

In astronomy, we frequently face the decision problem: does this data contain a signal? Typically, a statistical approach is used, which requires a threshold. The choice of threshold presents a common challenge in settings where signals and noise must be delineated, but their distributions overlap. Gravitational-wave astronomy, which has gone from the first discovery to catalogs of hundreds of events in less than a decade, presents a fascinating case study. For signals from colliding compact objects, the field has evolved from a frequentist to a Bayesian methodology. However, the issue of choosing a threshold and validating noise contamination in a catalog persists. Confusion and debate often arise due to the misapplication of statistical concepts, the complicated nature of the detection statistics, and the inclusion of astrophysical background models. We introduce conformal prediction (CP), a framework developed in machine learning to provide distribution-free uncertainty quantification to point predictors. We show that CP can be viewed as an extension of the traditional statistical frameworks whereby thresholds are calibrated such that the uncertainty intervals are statistically rigorous and the error rate can be validated. Moreover, we discuss how CP offers a framework to optimally build a metapipeline combining the outputs from multiple independent searches. We introduce CP with a toy cosmic-ray detector, which captures the salient features of most astrophysical search problems and allows us to demonstrate the features of CP in a simple context. We then apply the approach to a recent gravitational-wave mock data challenge using multiple search algorithms for compact binary coalescence signals in interferometric gravitational-wave data. Finally, we conclude with a discussion on the future potential of the method for gravitational-wave astronomy. Published by the American Physical Society 2024

Research on a Human Moving Object Detection Method Based on Gaussian Model and Deep Learning

In order to understand human motion object detection methods, the author proposes a research on human motion object detection method based on Gaussian model. Firstly, traditional Gaussian models are unable to detect complex scenes or slow moving targets. Therefore, an improved Gaussian model based moving object detection algorithm is proposed. Secondly, multiple Gaussian models are used to represent the features of each pixel in the moving target image, and based on the matching of each pixel in the image with the Gaussian model, it is considered as a background point. Conversely, it is based on the principle of the foreground, and the Gaussian model is updated. Finally, by updating the foreground model and calculating short-term stability indicators, the detection effect of moving targets is improved. By determining the Gaussian distribution and pixel relationship, new parameters are set to construct the background model and eliminate the impact caused by sudden changes in lighting. The experimental analysis results show that this method can effectively detect and track moving targets, with good noise resistance, high clarity, and an accuracy rate of up to 99%. Compared with traditional Gaussian model methods, the improved method can more effectively detect moving targets and has better robustness.

The analysis of the skeletal muscle metabolism is crucial for designing optimal exercise paradigms in type 2 diabetes mellitus

BackgroundType 2 diabetes mellitus (T2DM) is a chronic metabolic disease that commonly results from a high-calorie diet and sedentary lifestyle, leading to insulin resistance and glucose homeostasis perturbation. Physical activity is recommended as one first-line treatment in T2DM, but it leads to contrasted results. We hypothesized that, instead of applying standard exercise protocols, the prescription of personalized exercise programs specifically designed to reverse the potential metabolic alterations in skeletal muscle could result in better results.MethodsTo test this hypothesis, we drew the metabolic signature of the fast-twitch quadriceps muscle, based on a combined unbiased NMR spectroscopy and RT-qPCR study, in several T2DM mouse models of different genetic background (129S1/SvImJ, C57Bl/6J), sex and aetiology (high-fat diet (HFD) or HFD/Streptozotocin (STZ) induction or transgenic MKR (FVB-Tg Ckm-IGF1R*K1003R)1Dlr/J) mice. Three selected mouse models with unique muscular metabolic signatures were submitted to three different swimming-based programs, designed to address each metabolic specificity.ResultsWe found that depending on the genetic background, the sex, and the mode of T2DM induction, specific muscular adaptations occurred, including depressed glycolysis associated with elevated PDK4 expression, shift to β-oxidation, or deregulation of amino-acid homeostasis. Interestingly, dedicated swimming-based exercises designed to restore specific metabolic alterations in muscle were found optimal in improving systemic T2DM hallmarks, including a significant reduction in insulin resistance, the improvement of glucose homeostasis, and a delay in sensorimotor function alterations.ConclusionThe muscle metabolism constitutes an important clue for the design of precision exercises with potential clinical implications for T2DM patients.

A joint ESTRO and AAPM guideline for development, clinical validation and reporting of artificial intelligence models in radiation therapy

Background and purposeArtificial Intelligence (AI) models in radiation therapy are being developed with increasing pace. Despite this, the radiation therapy community has not widely adopted these models in clinical practice. A cohesive guideline on how to develop, report and clinically validate AI algorithms might help bridge this gap. Methods and materialsA Delphi process with all co-authors was followed to determine which topics should be addressed in this comprehensive guideline. Separate sections of the guideline, including Statements, were written by subgroups of the authors and discussed with the whole group at several meetings. Statements were formulated and scored as highly recommended or recommended. ResultsThe following topics were found most relevant: Decision making, image analysis, volume segmentation, treatment planning, patient specific quality assurance of treatment delivery, adaptive treatment, outcome prediction, training, validation and testing of AI model parameters, model availability for others to verify, model quality assurance/updates and upgrades, ethics. Key references were given together with an outlook on current hurdles and possibilities to overcome these. 19 Statements were formulated. ConclusionA cohesive guideline has been written which addresses main topics regarding AI in radiation therapy. It will help to guide development, as well as transparent and consistent reporting and validation of new AI tools and facilitate adoption.

Uncertainty estimation using a 3D probabilistic U-Net for segmentation with small radiotherapy clinical trial datasets

Background and objectivesBio-medical image segmentation models typically attempt to predict one segmentation that resembles a ground-truth structure as closely as possible. However, as medical images are not perfect representations of anatomy, obtaining this ground truth is not possible. A surrogate commonly used is to have multiple expert observers define the same structure for a dataset. When multiple observers define the same structure on the same image there can be significant differences depending on the structure, image quality/modality and the region being defined. It is often desirable to estimate this type of aleatoric uncertainty in a segmentation model to help understand the region in which the true structure is likely to be positioned. Furthermore, obtaining these datasets is resource intensive so training such models using limited data may be required. With a small dataset size, differing patient anatomy is likely not well represented causing epistemic uncertainty which should also be estimated so it can be determined for which cases the model is effective or not. MethodsWe use a 3D probabilistic U-Net to train a model from which several segmentations can be sampled to estimate the range of uncertainty seen between multiple observers. To ensure that regions where observers disagree most are emphasised in model training, we expand the Generalised Evidence Lower Bound (ELBO) with a Constrained Optimisation (GECO) loss function with an additional contour loss term to give attention to this region. Ensemble and Monte-Carlo dropout (MCDO) uncertainty quantification methods are used during inference to estimate model confidence on an unseen case. We apply our methodology to two radiotherapy clinical trial datasets, a gastric cancer trial (TOPGEAR, TROG 08.08) and a post-prostatectomy prostate cancer trial (RAVES, TROG 08.03). Each dataset contains only 10 cases each for model development to segment the clinical target volume (CTV) which was defined by multiple observers on each case. An additional 50 cases are available as a hold-out dataset for each trial which had only one observer define the CTV structure on each case. Up to 50 samples were generated using the probabilistic model for each case in the hold-out dataset. To assess performance, each manually defined structure was matched to the closest matching sampled segmentation based on commonly used metrics. ResultsThe TOPGEAR CTV model achieved a Dice Similarity Coefficient (DSC) and Surface DSC (sDSC) of 0.7 and 0.43 respectively with the RAVES model achieving 0.75 and 0.71 respectively. Segmentation quality across cases in the hold-out datasets was variable however both the ensemble and MCDO uncertainty estimation approaches were able to accurately estimate model confidence with a p-value < 0.001 for both TOPGEAR and RAVES when comparing the DSC using the Pearson correlation coefficient. ConclusionsWe demonstrated that training auto-segmentation models which can estimate aleatoric and epistemic uncertainty using limited datasets is possible. Having the model estimate prediction confidence is important to understand for which unseen cases a model is likely to be useful.

Hyperspectral Anomaly Detection Using Reconstruction Fusion of Quaternion Frequency Domain Analysis.

Most existing techniques consider hyperspectral anomaly detection (HAD) as background modeling and anomaly search problems in the spatial domain. In this article, we model the background in the frequency domain and treat anomaly detection as a frequency-domain analysis problem. We illustrate that spikes in the amplitude spectrum correspond to the background, and a Gaussian low-pass filter performing on the amplitude spectrum is equivalent to an anomaly detector. The initial anomaly detection map is obtained by the reconstruction with the filtered amplitude and the raw phase spectrum. To further suppress the nonanomaly high-frequency detailed information, we illustrate that the phase spectrum is critical information to perceive the spatial saliency of anomalies. The saliency-aware map obtained by phase-only reconstruction (POR) is used to enhance the initial anomaly map, which realizes a significant improvement in background suppression. In addition to the standard Fourier transform (FT), we adopt the quaternion FT (QFT) for conducting multiscale and multifeature processing in a parallel way, to obtain the frequency domain representation of the hyperspectral images (HSIs). This helps with robust detection performance. Experimental results on four real HSIs validate the remarkable detection performance and excellent time efficiency of our proposed approach when compared to some state-of-the-art anomaly detection methods.

АНАЛІЗ МЕТОДІВ ЗНАХОДЖЕННЯ АНОМАЛІЙ У ЗОБРАЖЕННЯХ

Abstract. Automatic anomaly detection is of great importance in industry, remote sensing, and medicine. It is important to be able to automatically process large amounts of data to detect, for example, chemical objects in multispectral and hyperspectral satellite images, sea mines in side-scan sonar images, or defects in industrial monitoring applications. Automatic detection of anomalous structures on arbitrary images refers to the task of finding inappropriate patterns relative to the normal state of the image. This is a difficult task in computer vision, since there is no clear and straightforward definition of what is normal or not normal for a given arbitrary image. The practical importance is manifested in the development of algorithms and models that can automatically detect unusual or anomalous patterns in images. An analysis of methods for finding anomalies in images from the point of view of the possibility of application to arbitrary images has been carried out. The classification of anomaly detection methods according to the criteria of the involved approaches and models used for modeling the background is presented. Methods that use machine learning, such as one-class support vector method and variational autoencoder, nearest neighbor-based anomaly detection, clustering-based anomaly detection, statistical anomaly detection, spectral anomaly detection, anomaly detection using information theory are discussed. The main attention is paid to the methods classified according to the background modeling approach. Five categories of background modeling methods based on probability density function, global and local homogeneity, sparsity, and self-similarity are considered. For anomaly detection applications, it is recommended to use methods in which the background model best describes the expected anomaly-free background, as this generally results in the best performance. On the basis of research, it was established that an effective universal model for detecting anomalies in arbitrary images should: use only a self-similar or sparse background model; process the residual image as a stochastic process to detect anomalies, such as anomalies in color noise; preprocess the residual image before detecting the anomaly.

IONOLAB-Fusion: Fusion of Radio Occultation into Computerized Ionospheric Tomography

In this study, a 4-D, computerized ionospheric tomography algorithm, IONOLAB-Fusion, is developed to reconstruct electron density using both actual and virtual vertical and horizontal paths for all ionospheric states. The user-friendly algorithm only requires the coordinates of the region of interest and range with the desired spatio-temporal resolutions. The model ionosphere is formed using spherical voxels in a lexicographical order so that a 4-D ionosphere can be mapped to a 2-D matrix. The model matrix is formed automatically using a background ionospheric model with an optimized retrospective or near-real time manner. The singular value decomposition is applied to extract a subset of significant singular values and corresponding signal subspace basis vectors. The measurement vector is filled automatically with the optimized number of ground-based and space-based paths. The reconstruction is obtained in closed form in the least squares sense. When the performance of IONOLAB-Fusion across Europe was compared with ionosonde profiles, a 26.51% and 32.33% improvement was observed over the background ionospheric model for quiet and disturbed days, respectively. When compared with GIM-TEC, the agreement of IONOLAB-Fusion was 37.89% and 31.58% better than those achieved with the background model for quiet and disturbed days, respectively.

Smoothness: The key factor in well-log information-assisted PINNtomo

Seismic traveltime tomography is popularly used to provide a smooth background model for subsequent depth migration or full waveform inversion (FWI). The core of the traveltime tomography process requires an eikonal solver for seismic traveltime calculation. Most of the conventional eikonal solvers, suffer from first-order convergence errors and difficulties in dealing with irregular topography, which will lead to inaccurate seismic tomography results. To solve these issues, physics-informed neural networks (PINNs) have been developed and demonstrated remarkable success in simultaneous traveltime calculation and velocity construction, which is referred to as PINNtomo. It combines the physics constraint (eikonal equation) and data constraint (traveltime fitting) to form the loss function of the networks. Though PINNtomo can manage to recover a smooth background velocity, it will fail in the inversion for a complicated high-resolution velocity model. We take advantage of the well-log velocity as a prior information to provide an additional model constraint in the loss function to improve the performance of PINNtomo. We found that the well-log information will provide high-resolution components in the inverted velocity model. However, it will destroy the lateral structure of the true model. We propose to use a smoothed version of the well-log velocity as the prior model constraint in PINNtomo. As a result, the accuracy of the inverted background velocity will be improved, which will be beneficial to the subsequent FWI.

CHEX-MATE: Robust reconstruction of temperature profiles in galaxy clusters with XMM-Newton

The “Cluster HEritage project with XMM-Newton: Mass Assembly and Thermodynamics at the End point of structure formation” (CHEX-MATE) is a multi-year heritage program to obtain homogeneous XMM-Newton observations of a representative sample of 118 galaxy clusters. The observations are tuned to reconstruct the distribution of the main thermodynamic quantities of the intra-cluster medium up to R500 and to obtain individual mass measurements, via the hydrostatic-equilibrium equation, with a precision of 15−20%. Temperature profiles are a necessary ingredient for the scientific goals of the project and it is thus crucial to derive the best possible temperature measurements from our data. This is why we have built a new pipeline for spectral extraction and analysis of XMM-Newton data, based on a new physically motivated background model and on a Bayesian approach with Markov chain Monte Carlo methods, which we present in this paper for the first time. We applied this new method to a subset of 30 galaxy clusters representative of the CHEX-MATE sample and show that we can obtain reliable temperature measurements up to regions where the source intensity is as low as 20% of the background, keeping systematic errors below 10%. We compare the median profile of our sample and the best-fit slope at large radii with literature results and we find a good agreement with other measurements based on XMM-Newton data. Conversely, when we exclude the most contaminated regions, where the source intensity is below 20% of the background, we find significantly flatter profiles, in agreement with predictions from numerical simulations and independent measurements with a combination of Sunyaev–Zeldovich and X-ray imaging data.

#2924 Comparison of large language models and traditional natural language processing techniques in predicting arteriovenous fistula failure

Abstract Background and Aims Large language models (LLMs) have gained significant attention in the field of natural language processing (NLP), marking a shift from traditional techniques like Term Frequency-Inverse Document Frequency (TF-IDF). We developed a traditional NLP model to predict arteriovenous fistula (AVF) failure within next 30 days using clinical notes. The goal of this analysis was to investigate whether LLMs would outperform traditional NLP techniques, specifically in the context of predicting AVF failure within the next 30 days using clinical notes. Method We defined AVF failure as the change in status from active to permanently unusable status or temporarily unusable status. We used data from a large kidney care network from January 2021 to December 2021. Two models were created using LLMs and traditional TF-IDF technique. We used “distilbert-base-uncased”, a distilled version of BERT base model [1], and compared its performance with traditional TF-IDF-based NLP techniques. The dataset was randomly divided into 60% training, 20% validation and 20% test dataset. The test data, comprising of unseen patients’ data was used to evaluate the performance of the model. Both models were evaluated using metrics such as area under the receiver operating curve (AUROC), accuracy, sensitivity, and specificity. Results The incidence of 30 days AVF failure rate was 2.3% in the population. Both LLMs and traditional showed similar overall performance as summarized in Table 1. Notably, LLMs showed marginally better performance in certain evaluation metrics. Both models had same AUROC of 0.64 on test data. The accuracy and balanced accuracy for LLMs were 72.9% and 59.7%, respectively, compared to 70.9% and 59.6% for the traditional TF-IDF approach. In terms of specificity, LLMs scored 73.2%, slightly higher than the 71.2% observed for traditional NLP methods. However, LLMs had a lower sensitivity of 46.1% compared to 48% for traditional NLP. However, it is worth noting that training on LLMs took considerably longer than TF-IDF. Moreover, it also used higher computational resources such as utilization of graphics processing units (GPU) instances in cloud-based services, leading to higher cost. Conclusion In our study, we discovered that advanced LLMs perform comparably to traditional TF-IDF modeling techniques in predicting the failure of AVF. Both models demonstrated identical AUROC. While specificity was higher in LLMs compared to traditional NLP, sensitivity was higher in traditional NLP compared to LLMs. LLM was fine-tuned with a limited dataset, which could have influenced its performance to be similar to that of traditional NLP methods. This finding suggests that while LLMs may excel in certain scenarios, such as performing in-depth sentiment analysis of patient data for complex tasks, their effectiveness is highly dependent on the specific use case. It is crucial to weigh the benefits against the resources required for LLMs, as they can be significantly more resource-intensive and costly compared to traditional TF-IDF methods. This highlights the importance of a use-case-driven approach in selecting the appropriate NLP technique for healthcare applications.

#1469 Predicting mortality risk in incident dialysis patients using clinical predictors and patient-reported predictors

Abstract Background and Aims Prognostic prediction models can be used to identify dialysis patients that are at high risk of dying. By doing so, vulnerable patients can be monitored more closely and treatment can be intensified where needed. In addition, individualized predictions can help inform patients about their prognosis. Traditionally, prognostic models primarily rely on known clinical predictors such as age, comorbidities and laboratory markers. Although it is still unchartered territory in the field of nephrology, studies in other fields have shown that patient-reported predictors, like pain and fatigue, can be of use as predictors of survival. Predictive performance of current prognostic models for mortality risk amongst dialysis patients is subpar and we hypothesized that the addition of patient-reported predictors may improve predictions. Therefore, we aimed to develop a novel prognostic model for the prediction of two-year mortality in patients starting dialysis using both clinical predictors and patient-reported predictors. Method Data from the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD) were used to develop two prognostic models for the prediction of 2-year all-cause mortality in incident dialysis patients: a base model consisting of clinical predictors and an updated model containing additional patient-reported predictors. Missing data were handled using a tenfold multiple imputation with fully conditional specification. The imputation model included all candidate predictors as well as the outcome variable of 2-year mortality. Predictors for the base model were selected based on existing mortality models and clinical expertise. Patient-reported predictors to add to the base model were chosen based on availability in NECOSAD, clinical expertise and literature describing patient-reported predictors. The models were developed using multivariate logistic regression and their discriminative abilities were assessed by calculating the area under the curve (AUC). Results A total of 1956 patients from NECOSAD were used for the analyses. Of these, 447 (22.9%) died within two years after baseline (3 months after dialysis initiation). The base model contained the following predictors: age, sex, serum albumin, C-reactive protein, diabetes mellitus, hemoglobin, residual kidney function (GFR at 3 months), body mass index, cardiovascular disease, dialysis modality, malignancy and primary kidney disease. The discriminative abilities of the base model were moderate with a AUC of 0.79 (95% CI: 0.77-0.82). The base model was updated using the following patient-reported predictors: mental component score, physical component score, general health perception, number of symptoms, symptom burden (measured with the KDQOL instrument) depressive symptoms, fatigue and pain. The addition of the patient-reported predictors marginally improved the predictive performance of the model achieving an AUC of 0.81 (95% CI: 0.79-0.84). Conclusion The base model including clinical predictors was enhanced slightly by the addition of patient-reported predictors.

#560 The representation of multimorbidity and frailty in the development and validation of kidney failure risk prediction models – a systematic review

Abstract Background and Aims Prognostic models that identify individuals with chronic kidney disease (CKD) at highest risk of developing kidney failure help clinicians to make decisions and deliver precision medicine. People with CKD are very likely to have multiple long-term health conditions (multimorbidity) and often experience frailty. These factors impact progression of kidney disease and influence the risk of other outcomes such as death. It is unclear to what extent prognostic models, that estimate the risk of kidney failure, consider the impact of multimorbidity and frailty and whether these models are valid in such sub-populations. This systematic review (CRD42022347295) describes and evaluates the representation of multimorbidity and frailty within cohorts used to develop and/or validate prognostic models assessing the risk of kidney failure in individuals with CKD. The review aims to determine if multimorbidity or frailty has been considered in relation to prediction of kidney failure and if reliable prognostic models exist for use in this population. Method We included studies that described the derivation, validation or update of a kidney failure prognostic model (outcome assessed at ≥2 years) for use in adults with CKD and reported at least one measure of either discrimination or calibration. The primary outcome for the review was the representation of multimorbidity or frailty in these models. An electronic search for published peer-reviewed articles involved MEDLINE, CINAHL Plus and the Cochrane Library – CENTRAL and was supplemented with manual review of references from previous systematic reviews and clinical guidelines. Results A total of 97 studies, reporting 121 different kidney failure prognostic models were identified. A total of 2 925 413 participants and 149 380 kidney failure events were included across all studies. Included participants had a mean age of 58.9 years (SD 9.6), 44.4% were female, with a mean eGFR of 47.5 ml/min/1.73 m2 (SD 10.7). Only two studies reported multimorbidity, measured by Elixhauser Comorbidity Index and Charlson Co-morbidity Index (CCI), and a single study reported a self-reported proxy frailty measure. The rates of specific co-morbidities were reported in a greater proportion of studies: 67.0% (n = 65) studies reported baseline data on diabetes, 54.6% (n = 53) reported hypertension, and 39.2% (n = 38) reported cardiovascular disease. No studies included frailty in model development and only one study considered multimorbidity as a predictor variable (via CCI). No studies assessed model performance in populations with multimorbidity. A single study assessed clinical utility of a model and referral algorithms considering frailty and the risk of kidney failure and death. Conclusion There is a paucity of kidney failure risk prediction models that consider the impact of multimorbidity and/or frailty in adults with CKD, resulting in a lack of clear evidence-based practice and guidance for multimorbid or frail individuals. These knowledge gaps should be explored to help clinicians know whether these models can be used for patients who experience multimorbidity and/or frailty.

#1944 How the patient characteristics and the prescription can be accounted for in patient-specific predictive HD models?

Abstract Background and Aims Compartmental models in HemoDialysis (HD) allow for understanding and predicting the patient-specific response to the treatment. These models, after training with consistent clinical data, can be adapted to each single patient through the estimation of a set of parameters. This work explores the relationship between the sets of estimated parameters and the anamneses of the enrolled patients, together with the settings of the HD sessions, and their time evolution. The finding of similar relationships will help properly set the models even avoiding the training phase. Method The data acquired during the INTERREG InterACTIVE-HD 2.0 observational longitudinal study, in partnership with 5 Italian and Switzerland hospitals (ASSTs Lariana, SetteLaghi, Valtellina e Alto Lario, Regional Hospital of Lugano, and Kantonsspital Graubünden of Chur Dialysis Units), were used (576 sessions). Clinical prescription, machine settings, and blood composition at the start, end, and every hour during the HDs were monitored over 7 months. The parametric multi-compartmental, multi-solute model optimized during the study was used. The patient-specific parameters are: Their default values (literature-based) are: Patient-specific parameters have been identified for each patient, by training the model with 3 session's data. Then the model can be used to predict the patient's response by knowing the clinical data at the beginning of the HD and setting the parameters equal to the mean of the 3 identified values. A correlation analysis between the patient-dependent parameters and the anamneses of the enrolled patients, together with the settings of the HD sessions, has been performed. The optimization orders have been also considered. Results k(K+), k(Urea), and k(Crea) are the first optimized and remain generally of the same magnitude. k(Na+), k(Cl-), k(HCO3-), k(Mg2+), and k(Glu), correlate with the UF rate, the therapy technique, or the concentrations in dialysate bags. They are later optimized and characterized by wider variability in values and optimization orders. k(Ca2+) remains at the default in 98% of the sessions, being optimized last. More convective therapies seem to induce patient-specific variations in active and passive cellular transport for specific solutes. η relevant correlations refer to the dialysate bag composition and the baseline conditions of the patients. They are in general late optimized, in HDF before than in SHD. η(Urea) and η(Crea) in 99% of the cases do not move from the default, suggesting a lower interaction of the non-ionic species with the filter membrane. They are optimized only in non-hypertensive insulin-dependent diabetic patients. No correlation between η and the filter material or the initial serum albumin was instead observed. ρ is related to the therapy; in HDF sessions it is over the range boundaries ([0,10]). For females undergoing SHD, it assumes the highest values (&amp;gt;9). ρ seems also to be affected by the use of beta-blockers (p &amp;lt; 0.01) and antihyperglycemic ((p &amp;lt; 0.01). Ca_buff coefficients for females (higher values) and males are statistically different. Diabetics using anti-hyperglycemia drugs are characterized by lower values. Conclusion Compartmental patient-specific models, if successfully trained become a useful support tool for therapy customization, mainly for ESRD patients with multiple comorbidities. The observed correlations represent the first step to causally relate the values of the modulating parameters with the patient characteristics implying the reduction or amplification of a specific response to the treatment. The next step will be an analysis of combined effects. The final goal is to be able to correctly set up the models even before the training phase is completed.

The lithospheric structure of Greenland from a stepwise forward and inverse modelling approach

Summary Greenland’s tectonic history is complex, and the resulting lithospheric structure is, although extensively researched, not well constrained. In this study, we model the lithospheric structure of Greenland in a consistent, integrated framework with three steps. First, we build a lithospheric background model by forward modelling, adjusted to gravity gradient data and shear wave velocities from a regional tomography model. Subsequently, we jointly invert for the upper crustal density and susceptibility structure by minimizing the gravity residuals and magnetic total field anomaly misfit. The last modelling step searches for upper crustal thermal parameters to fit our model to the most recent geothermal heat flow predictions for Greenland. Finally, we present 3D models of the density, temperature and velocity structure for the lithosphere as well as thermal parameters and susceptibilities for the upper crust. Our model also includes the depth of the Moho and LAB in Greenland. A comparison between inverted crustal parameters and surface geology shows a clear correlation. The novelty of our model is that all these results are consistent with each other and simultaneously explain a wide range of observed data.

Adaptive design of experiments to fit surrogate Gaussian process regression models allows fast sensitivity analysis of the input waveform for patient-specific 3D CFD models of liver radioembolization

Background and objectivePatient-specific 3D computational fluid dynamics (CFD) models are increasingly being used to understand and predict transarterial radioembolization procedures used for hepatocellular carcinoma treatment. While sensitivity analyses of these CFD models can help to determine the most impactful input parameters, such analyses are computationally costly. Therefore, we aim to use surrogate modelling to allow relatively cheap sensitivity analysis. As an example, we compute Sobol's sensitivity indices for three input waveform shape parameters. MethodsWe extracted three characteristic shape parameters from our input mass flow rate waveform (peak systolic mass flow rate, heart rate, systolic duration) and defined our 3D input parameter space by varying these parameters within 75 %-125 % of their nominal values. To fit our surrogate model with a minimal number of costly CFD simulations, we developed an adaptive design of experiments (ADOE) algorithm. The ADOE uses 100 Latin hypercube sampled points in 3D input space to define the initial design of experiments (DOE). Subsequently, we re-sample input space with 10,000 Latin Hypercube sampled points and cheaply estimate the outputs using the surrogate model. In each of 27 equivolume bins which divide our input space, we determine the most uncertain prediction of the 10,000 points, compute the true outputs using CFD, and add these points to the DOE. For each ADOE iteration, we calculate Sobol's sensitivity indices, and we continue to add batches of 27 samples to the DOE until the Sobol indices have stabilized. ResultsWe tested our ADOE algorithm on the Ishigami function and showed that we can reliably obtain Sobol's indices with an absolute error <0.1. Applying ADOE to our waveform sensitivity problem, we found that the first-order sensitivity indices were 0.0550, 0.0191 and 0.407 for the peak systolic mass flow rate, heart rate, and the systolic duration, respectively. ConclusionsAlthough the current study was an illustrative case, the ADOE allows reliable sensitivity analysis with a limited number of complex model evaluations, and performs well even when the optimal DOE size is a priori unknown. This enables us to identify the highest-impact input parameters of our model, and other novel, costly models in the future.