SEARCH · Engineering Papers

Results for “regression”

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

SourceResearch connectionFrom yearThrough year

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

At least 1,189 records · Page 66

All sources NASA NTRS DOE OSTI

DOE OSTI2024

Short-Term Load Forecasting Considering EV Charging Loads with Prediction Interval Evaluation

Short-term load forecasting plays a critical role in power system planning and operation. Along with the electrification of various loads, electricity demands are becoming increasingly hard to predict. Notably, the recent rise in electric vehicles (EVs) has further contributed to this unpredictability. To address this issue, this paper proposes a probabilistic load forecasting strategy utilizing Gaussian process regression, structured in a day-ahead manner. While many works focus on deterministic prediction, probabilistic forecasting offers additional insights into variability and uncertainty, enabling more flexible and reliable operation for power systems. To enhance the accuracy of the load forecasting model, the inputs include features related to EV charging habits as well as commonly used weather information. The load forecasting results are evaluated using various metrics, including conventional ones that assess the accuracy of point forecasts, as well as additional metrics that test the reliability of prediction intervals. The proposed load forecasting method is finally tested on real residential power consumption data and EV charging data sampled from real-world sources. The results prove that the new features can greatly improve the performance of the load forecasting method.

electrical vehicle↗

DOE OSTI2019

Data-driven Feature Analysis in Control Design for Series-Compensated Transmission Systems

One challenge in power-system control designs is the gap between numerical model-based analysis and complex real-world power systems. With increased data and measurements being collected from power systems, data-driven analysis (e.g., machine learning) may provide an alternative approach to reveal hidden information through learning from the real system data, and provide insights for better control scheme design during the utility planning process. Data-driven feature analysis was performed to evaluate the relationships between series compensation, power generation, and path flows in a real transmission system, as well as temporal patterns. The main data-driven analysis methods, including statistical cross-correlation, multinomial logistical regression, and classification and regression trees, were integrated for feature selection and developing predictive models of series compensation. Analysis results demonstrated the effectiveness of the proposed methodology in feature analysis and the potential to help improve power-system control scheme design.

42 ENGINEERING↗

DOE OSTI2021

Comparative Analysis of Machine Learning Models for Day-Ahead Photovoltaic Power Production Forecasting

A main challenge for integrating the intermittent photovoltaic (PV) power generation remains the accuracy of day-ahead forecasts and the establishment of robust performing methods. The purpose of this work is to address these technological challenges by evaluating the day-ahead PV production forecasting performance of different machine learning models under different supervised learning regimes and minimal input features. Specifically, the day-ahead forecasting capability of Bayesian neural network (BNN), support vector regression (SVR), and regression tree (RT) models was investigated by employing the same dataset for training and performance verification, thus enabling a valid comparison. The training regime analysis demonstrated that the performance of the investigated models was strongly dependent on the timeframe of the train set, training data sequence, and application of irradiance condition filters. Furthermore, accurate results were obtained utilizing only the measured power output and other calculated parameters for training. Consequently, useful information is provided for establishing a robust day-ahead forecasting methodology that utilizes calculated input parameters and an optimal supervised learning approach. Finally, the obtained results demonstrated that the optimally constructed BNN outperformed all other machine learning models achieving forecasting accuracies lower than 5%.

14 SOLAR ENERGY↗

NASA NTRS2020

Statistical Analysis of Factors Riving Surface Ozone Variability over Continental South Africa

Statistical relationships between surface ozone (O3) concentration, precursor species and meteorological conditions in continental South Africa were examined from data obtained from measurement stations in north-eastern South Africa. Three multivariate statistical methods were applied in the investigation, i.e. multiple linear regression (MLR), principal component analysis (PCA) and –regression (PCR), and generalised additive model (GAM) analysis. The daily maximum 8-h moving average O3 concentrations were considered in these statistical models (dependent variable). MLR models indicated that meteorology and precursor species concentrations are able to explain ~50% of the variability in daily maximum O3 levels. MLR analysis revealed that atmospheric carbon monoxide (CO), temperature and relative humidity were the strongest factors affecting the daily O3 variability. In summer, daily O3 variances were mostly associated with relative humidity, while winter O3 levels were mostly linked to temperature and CO. PCA indicated that CO, temperature and relative humidity were not strongly collinear. GAM also identified CO, temperature and relative humidity as the strongest factors affecting the daily variation of O3. Partial residual plots found that temperature, radiation and nitrogen oxides most likely have a non-linear relationship with O3,while the relationship with relative humidity and CO is probably linear. An inter-comparison between O3 levels modelled with the three statistical models compared to measured O3 concentrations showed that the GAM model offered a slight improvement over the MLR model. These findings emphasise the critical role of regional-scale O3 precursors coupled with meteorological conditions in daily variances of O3 levels in continental South Africa.

multiple linear regression (MLR)↗

DOE OSTI2025

Joint Modeling of Wind Speed and Wind Direction Through a Conditional Approach

Atmospheric near surface wind speed and wind direction play an important role in many applications, ranging from air quality modeling, building design, wind turbine placement to climate change research. It is therefore crucial to accurately estimate the joint probability distribution of wind speed and direction. In this work, we develop a conditional approach to model these two variables, where the joint distribution is decomposed into the product of the marginal distribution of wind direction and the conditional distribution of wind speed given wind direction. To accommodate the circular nature of wind direction, a von Mises mixture model is used; the conditional wind speed distribution is modeled as a directional dependent Weibull distribution via a two-stage estimation procedure, consisting of a directional binned Weibull parameter estimation, followed by a harmonic regression to estimate the dependence of the Weibull parameters on wind direction. A Monte Carlo simulation study indicates that our method outperforms two other approaches in estimation efficiency: one that utilizes periodic spline quantile regression and another that generates data from the commonly used Abe-Ley distribution for cylindrical data. We illustrate our method by using the output from a regional climate model to investigate how the joint distribution of wind speed and direction may change under some future climate scenarios. Our method indicates significant changes in the variation of wind speed with respect to some directions.

17 WIND ENERGY↗

DOE OSTI2018

Predicting photovoltaic soiling losses using environmental parameters: An update

This study presents an investigation on the correlations between soiling losses and environmental parameters at 41 locations in the United States, with the aim of analyzing the possibility of predicting soiling losses at a site even when soiling data are not available. The results of this work, which considers the largest pool of soiling data points systematically investigated so far, confirm that a single-variable regression based on particulate matter concentration returns the best correlations with soiling, with adjusted coefficients of determination up to 70%, corresponding to RMSE as low as 0.9%. Among the various particulate matter datasets investigated, a gridded Environment Protection Agency dataset is for the first time found to return correlations similar to those obtained by interpolating particulate matter monitoring station data. We discuss in detail the different interpolation techniques used to process the particulate matter concentrations because they can greatly impact the correlations. Specifically, the correlation coefficients between soiling and particulate matter range between 70% and less than 20%, depending on the interpolation methods and monitoring distance. Spatial interpolation methods based on inverse distance weighting are found to return better correlations than a nearest neighbor or a simple average approach, especially when large distances are considered. Similarly, the effects of different rain thresholds used to calculate the length of the dry periods are examined. An enhanced two-variable regression is found to achieve higher-quality correlations, with adjusted R2 of 90% (RMSE = 0.55%), also suggesting that high and low soiling locations might be differentiated depending on fixed particulate matter or rainfall thresholds.

14 SOLAR ENERGY↗

DOE OSTI2019

MR texture analysis: potential imaging biomarker for predicting the chemotherapeutic response of patients with colorectal liver metastases

Purpose: The purpose of the study was to determine whether the pre-treated MR texture features of colorectal liver metastases (CRLMs) are predictive of therapeutic response after chemotherapy. Methods: The study included twenty-six consecutive patients (a total of 193 liver metastasis) with unrespectable CRLMs at our institution from August 2014 to February 2016. Lesions were categorized into either responding group or non-responding group according to changes in size. Texture analysis was quantified on T2-weighted images by two radiologists with consensus on regions of interest which were manually drawn on the largest cross-sectional area of the lesions. Five histogram features (mean, variance, skewness, kurtosis, and entropy{sup 1}) and five gray level co-occurrence matrix features (GLCM; angular second moment (ASM), entropy{sup 2}, contrast, correlation, and inverse difference moment (IDM)) were extracted. The texture parameters were statistically analyzed to identify the differences between the two groups, and the potential predictive parameters to differentiate the responding group from the non-responding group were subsequently tested using multivariable logistic regression analysis. Results: A total of 107 responding and 86 non-responding lesions were evaluated. A higher variance, entropy{sup 1}, contrast, entropy{sup 2} and a lower ASM, correlation, IDM were independently (P < 0.05) associated with a good response to chemotherapy with the areas under the ROC curves (AUCs) of 0.602–0.784. Variance (P < 0.001) and ASM (P = 0.001) remained potential predictive values to discriminate responding lesions from non-responding lesions when tested using multivariable logistic regression analysis. The highest AUC of the predictors from the association of variance and ASM was 0.814. Conclusion: MR texture features on pre-treated T2 images have the potential to predict the therapeutic response of colorectal liver metastases.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2019

Predictive role of PI-RADSv2 and ADC parameters in differentiating Gleason pattern 3 + 4 and 4 + 3 prostate cancer

Purpose: To compare the predictive roles of qualitative (PI-RADSv2) and quantitative assessment (ADC metrics), in differentiating Gleason pattern (GP) 3 + 4 from the more aggressive GP 4 + 3 prostate cancer (PCa) using radical prostatectomy (RP) specimen as the reference standard. Methods: We retrospectively identified treatment-naïve peripheral (PZ) and transitional zone (TZ) Gleason Score 7 PCa patients who underwent multiparametric 3T prostate MRI (DWI with b value of 0,1400 and where unavailable, 0,500) and subsequent RP from 2011 to 2015. For each lesion identified on MRI, a PI-RADSv2 score was assigned by a radiologist blinded to pathology data. A PI-RADSv2 score ≤ 3 was defined as “low risk,” a PI-RADSv2 score ≥ 4 as “high risk” for clinically significant PCa. Mean tumor ADC (ADC{sub T}), ADC of adjacent normal tissue (ADC{sub N}), and ADC{sub ratio} (ADC{sub T}/ADC{sub N}) were calculated. Stepwise regression analysis using tumor location, ADC{sub T} and ADC{sub ratio}, b value, low vs. high PI-RADSv2 score was performed to differentiate GP 3 + 4 from 4 + 3. Results: 119 out of 645 cases initially identified met eligibility requirements. 76 lesions were GP 3 + 4, 43 were 4 + 3. ADC{sub ratio} was significantly different between the two GP groups (p = 0.001). PI-RADSv2 score (“low” vs. “high”) was not significantly different between the two GP groups (p = 0.17). Regression analysis selected ADC{sub T} (p = 0.03) and ADC{sub ratio} (p = 0.0007) as best predictors to differentiate GP 4 + 3 from 3 + 4. Estimated sensitivity, specificity, and accuracy of the predictive model in differentiating GP 4 + 3 from 3 + 4 were 37, 82, and 66%, respectively. Conclusions: ADC metrics could differentiate GP 3 + 4 from 4 + 3 PCa with high specificity and moderate accuracy while PI-RADSv2, did not differentiate between these patterns.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2019

False positives in PIRADS (V2) 3, 4, and 5 lesions: relationship with reader experience and zonal location

Purpose: To investigate the effect of reader experience and zonal location on the occurrence of false positives (FPs) in PIRADS (V2) 3, 4, and 5 lesions on multiparametric (MP)-MRI of the prostate. Materials and methods: This retrospective study included 139 patients who had consecutively undergone an MP-MRI of the prostate in combination with a transrectal ultrasound MRI fusion-guided biopsy between 2014 and 2017. MRI exams were prospectively read by a group of inexperienced radiologists (cohort 1; 54 patients) and an experienced radiologist (cohort 2; 85 patients). Multivariable logistic regression analysis was performed to determine the association of experience of the radiologist and zonal location with a FP reading. FP rates were compared between readings by inexperienced and experienced radiologists according to zonal location, using Chi-square (χ{sup 2}) tests. Results: A total of 168 lesions in 139 patients were detected. Median patient age was 68 years (Interquartile range (IQR) 62.5–73), and median PSA was 10.9 ng/mL (IQR 7.6–15.9) for the entire patient cohort. According to multivariable logistic regression, inexperience of the radiologist was significantly (P = 0.044, odds ratio 1.927, 95% confidence interval [CI] 1.017–3.651) and independently associated with a FP reading, while zonal location was not (P = 0.202, odds ratio 1.444, 95% CI 0.820–2.539). In the transition zone (TZ), the FP rate of the inexperienced radiologists 59% (17/29) was significantly higher (χ{sup 2}P = 0.033) than that of the experienced radiologist 33% (13/40). Conclusion: Inexperience of the radiologist is significantly and independently associated with a FP reading, while zonal location is not. Inexperienced radiologists have a significantly higher FP rate in the TZ.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2019

Comparison of CT methods for determining graft steatosis in living donor liver transplantation

Purpose: To evaluate and compare the diagnostic performance of non-enhanced computed tomography (NECT) and contrast-enhanced CT (CECT) attenuation indices in the assessment of hepatic steatosis by using biopsy as the reference standard. Materials and methods: This retrospective study was approved by our Institutional Review Board. 55 Potential donors who underwent both NECT and triphasic CECT and core liver biopsy, were included the study. Average attenuation measurements that were obtained from multiple regions in liver, spleen, and psoas muscle on both unenhanced and CECT were used for analysis. Hepatic attenuation measurements were analyzed with and without normalization with the spleen and psoas muscle. Linear regression and receiver operating characteristic (ROC) curve analysis were used to evaluate the statistical association between CT indices and steatosis at histology. Results: Linear regression analysis confirmed the strongest correlation between steatosis and normalized measurements of hepatic attenuation with splenic attenuations on hepatic venous phase of CECT scan (R 0.821; R{sup 2} 0.674 and R 0.816; R{sup 2} 0.665, respectively). The use of ROC curve analysis also demonstrated that normalized measurements of hepatic attenuation with splenic attenuations on hepatic venous phase of CECT showed high diagnostic performance regarding the qualitative distinction of steatosis (AUC values greater than 0.9). Conclusion: Attenuation measurements of liver normalized with spleen on hepatic venous phase may be useful in evaluating steatosis in donor candidates with moderate to severe steatosis who are unacceptable for liver donation. In this manner unnecessary liver biopsy may be avoided in those donor candidates.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2016

MWA Versus RFA for Perivascular and Peribiliary CRLM: A Retrospective Patient- and Lesion-Based Analysis of Two Historical Cohorts

PurposeTo retrospectively analyse the safety and efficacy of radiofrequency ablation (RFA) versus microwave ablation (MWA) in the treatment of unresectable colorectal liver metastases (CRLM) in proximity to large vessels and/or major bile ducts.Method and MaterialsA database search was performed to include patients with unresectable histologically proven and/or {sup 18}F–FDG–PET avid CRLM who were treated with RFA or MWA between January 2001 and September 2014 in a single centre. All lesions that were considered to have a peribiliary and/or perivascular location were included. Univariate logistic regression analysis was performed to assess the distribution of patient, tumour and procedure characteristics. Multivariate logistic regression was used to correct for potential confounders.ResultsTwo hundred and forty-three patients with 774 unresectable CRLM were ablated. One hundred and twenty-two patients (78 males; 44 females) had at least one perivascular or peribiliary lesion (n = 199). Primary efficacy rate of RFA was superior to MWA after 3 and 12 months of follow-up (P = 0.010 and P = 0.022); however, after multivariate analysis this difference was non-significant at 12 months (P = 0.078) and vanished after repeat ablations (P = 0.39). More CTCAE grade III complications occurred after MWA versus RFA (18.8 vs. 7.9 %; P = 0.094); biliary complications were especially common after peribiliary MWA (P = 0.002).ConclusionFor perivascular CRLM, RFA and MWA are both safe treatment options that appear equally effective. For peribiliary CRLM, MWA has a higher complication rate than RFA, with similar efficacy. Based on these results, it is advised to use RFA for lesions in the proximity of major bile ducts.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2017

Low Cerebral Blood Volume Identifies Poor Outcome in Stent Retriever Thrombectomy

BackgroundMechanical thrombectomy (MT) is an efficient treatment of acute stroke caused by large-vessel occlusion. We evaluated the factors predicting poor clinical outcome (3-month modified Rankin Scale, mRS >2) although MT performed with modern stent retrievers.MethodsWe prospectively collected the clinical and imaging data of 105 consecutive anterior circulation stroke patients who underwent MT after multimodal CT imaging. Patients with occlusion of the internal carotid artery and/or middle cerebral artery up to the M2 segment were included. We recorded baseline clinical, procedural and imaging variables, technical outcome, 24-h imaging outcome and the clinical outcome. Differences between the groups were studied with appropriate statistical tests and binary logistic regression analysis.ResultsLow cerebral blood volume Alberta stroke program early CT score (CBV-ASPECTS) was associated with poor clinical outcome (median 7 vs. 9, p = 0.01). Lower collateral score (CS) significantly predicted poor outcome in regression modelling with CS = 0 increasing the odds of poor outcome 4.4-fold compared to CS = 3 (95% CI 1.27–15.5, p = 0.02). Lower CBV-ASPECTS significantly predicted poor clinical outcome among those with moderate or severe stroke (OR 0.82, 95% CI 0.68–1, p = 0.05) or poor collateral circulation (CS 0–1, OR 0.66, 95% CI 0.48–0.90, p = 0.009) but not among those with mild strokes or good collaterals.ConclusionsCBV-ASPECTS estimating infarct core is a significant predictor of poor clinical outcome among anterior circulation stroke patients treated with MT, especially in the setting of poor collateral circulation and/or moderate or severe stroke.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2018

The Effects of Alloying Elements on the Continuous Cooling Transformation Behavior of 2¼Cr-1Mo Steels

The continuous cooling transformation (CCT) behavior of eight 2¼Cr-1Mo steels from a statistically designed matrix was determined. These steels contained two levels of carbon (0.07 and 0.16 wt.%), manganese (0.35 and 0.85%), chromium (1.5 and 2.8%), and molybdenum (0.3 and 1.25%). Each steel was tested in a quenching dilatometer at five to six cooling rates between 725 and 1.2 °C/min. For each CCT sample, the change in length, microstructure, and macrohardness were determined. The ferrite content was also measured for samples cooled near the ferrite nose. Pearson correlation and multiple regression analyses were performed for various CCT diagram parameters. The correlation analysis showed that carbon and chromium contents significantly affected the critical temperatures and the bainite and martensite transformation temperatures. Increasing carbon content significantly increased the hardness for the bainite and martensite range of cooling rates, but hardness at slower cooling rates was unaffected by alloying elements. Regression equations were obtained for the critical temperatures and the ferrite nose cooling rate.

36 MATERIALS SCIENCE↗

DOE OSTI2019

Postoperative weight loss followed by radical cystectomy predicts poor prognosis in patients with muscle-invasive bladder cancer

We aimed to investigate the impact of postoperative weight loss following radical cystectomy (RC) on patients’ prognoses. RC and urinary diversion were performed in 512 consecutive patients with muscle-invasive bladder cancer at our hospitals between May 1996 and July 2018. Demographic clinical information, pre- and postoperative serum albumin, hemoglobin, and weight were evaluated retrospectively at 1 month. We also evaluated the association of weight loss with complications and overall survival (OS) as estimated using the Kaplan–Meier method and compared using the log-rank test. Risk factors for poor OS were determined by Inverse Probability of Treatment Weighted (IPTW)-adjusted Cox regression analysis. In 385 patients who met the study search criteria, median postoperative weight loss from baseline at 1 week and 1 month was 1.1 (− 1.8%) and 3.2 (− 5.4%) kg, respectively. Patients with significant weight loss (defined as ≥ 7.5% at 1 month) had higher-grade complications within 1 month and had significantly shorter OS than those with weight loss of < 7.5%. Type of urinary diversion, loss of serum albumin, and loss of hemoglobin were not significantly associated with weight loss. IPTW-adjusted Cox regression analysis showed that such significant weight loss was an independent risk factor for poor OS. Weight loss followed by radical cystectomy was significantly associated with poor prognosis in patients with muscle-invasive bladder cancer.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2024

Computationally efficient and error aware surrogate construction for numerical solutions of subsurface flow through porous media

Limiting the injection rate to restrict the pressure below a threshold at a critical location can be an important goal of simulations that model the subsurface pressure between injection and extraction wells. The pressure is approximated by the solution of Darcy’s partial differential equation for a given permeability field. The subsurface permeability is modeled as a random field since it is known only up to statistical properties. This induces uncertainty in the computed pressure. Solving the partial differential equation for an ensemble of random permeability simulations enables estimating a probability distribution for the pressure at the critical location. These simulations are computationally expensive, and practitioners often need rapid online guidance for real-time pressure management. An ensemble of numerical partial differential equation solutions is used to construct a Gaussian process regression model that can quickly predict the pressure at the critical location as a function of the extraction rate and permeability realization. The Gaussian process surrogate analyzes the ensemble of numerical pressure solutions at the critical location as noisy observations of the true pressure solution, enabling robust inference using the conditional Gaussian process distribution. Our first novel contribution is to identify a sampling methodology for the random environment and matching kernel technology for which fitting the Gaussian process regression model scales as O ( n log n ) instead of the typical O ( n 3 ) rate in the number of samples n used to fit the surrogate. The surrogate model allows almost instantaneous predictions for the pressure at the critical location as a function of the extraction rate and permeability realization. Our second contribution is a novel algorithm to calibrate the uncertainty in the surrogate model to the discrepancy between the true pressure solution of Darcy’s equation and the numerical solution. Finally, although our method is derived for building a surrogate for the solution of Darcy’s equation with a random permeability field, the framework broadly applies to solutions of other partial differential equations with random coefficients.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2019

Advantage of multi-band solar-induced chlorophyll fluorescence to derive canopy photosynthesis in a temperate forest

When solar-induced chlorophyll fluorescence (SIF) is developed as a novel approach to quantify gross primary production (GPP), narrow SIF emission spectrum around the atmospheric oxygen absorption windows (˜ 761 nm and ˜687 nm) has been widely used to represent SIF and to derive GPP. As SIF is a continuous spectrum, deriving the full broadband SIF emission spectrum over 640–850 nm provides an opportunity to fully explore the potential of SIF in estimating GPP. Using high-frequency measurements of canopy carbon flux and SIF emissions at the atmospheric absorption bands, here we reconstruct the full SIF spectrum from SIF signals at the absorption bands, and then analyze correlations between the GPP and the selected single SIF bands and their combinations using both linear regression (LR) and Gaussian processes regression (GPR). Our results indicate that (1) the red SIF bands (640–700 nm) shows low correlation with GPP due to the strong (re)absorption of red SIF emissions by leaf chlorophyll; (2) the individual bands in near-infrared area (at 720 nm, 740 nm, and 761 nm) can determine about 60% and 62% of the variance in GPP with hourly scale by LR and GPR, respectively, and the combination of those SIF bands provide increased predictive power, explaining 66% and 76% variance in GPP with hourly scale by LR and GPR, respectively; (3) the solar radiation saturation, fraction of direct solar radiation, air temperature and leaf area index may have negative impacts on the SIF-GPP correlations when they are beyond optimal thresholds; and (4) the temporal aggregation of SIF-GPP (daily scale) enhances the correlations as compared with the hourly scale: the daily combination of SIF bands (at 687 nm, 720 nm, and 761 nm) can account for 80% and 93% of variance in daily daytime GPP by LR and GPR, respectively, suggesting that the combination of these three bands of SIF on the daily scale is the best proxy for GPP. Our results provide a new approach to analyze the SIF-GPP correlations using the ground-based full broadband SIF emission, suggesting that multi-bands SIF has a stronger capacity in predicting plant GPP than traditionally used signal band SIF data.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2015

BLZF1 expression is of prognostic significance in hepatocellular carcinoma

BLZF1, a member of b-ZIP family, has been implicated in epigenetic regulation and Wnt/β-catenin signaling. Its expression and clinical significance in human cancers remain largely unknown. In this study, we showed that BLZF1 expression was reduced in hepatocellular carcinoma (HCC) tissues, compared to the paracarcinoma tissues, at both mRNA and protein levels. Results of immunohistochemistry revealed that BLZF1 was presented in both nuclear and cytoplasm. Decreased expression of nuclear and cytosolic BLZF1 in HCC was depicted in 68.2% and 79.2% of the 634 cases. Nuclear BLZF1 expression was significantly associated with tumor multiplicity (P = 0.048) and tumor capsule (P = 0.028), while cytosolic BLZF1 expression was correlated with serum AFP level (P = 0.017), tumor differentiation (P = 0.001) and tumor capsule (P = 0.003). Kaplan–Meier analysis indicated both nuclear and cytosolic BLZF1 expression was associated with poor overall survival. Low nuclear BLZF1 also indicated unfavorable disease-free survival and high tendency of tumor recurrence. Furthermore, multiple Cox regression analysis revealed nuclear BLZF1 as an independent factor for overall survival (Hazard Ratio (HR) = 0.827, 95% confident interval (95%CI): 0.697–0.980, P = 0.029). The prognostic value of BLZF1 was further confirmed by stratified analyses. Collectively, our data suggest BLZF1 is a novel unfavorable biomarker for prognosis of patients with HCC. - Highlights: • BLZF1 expression was much lower in HCC tissues. • Low BLZF1 expression was associated with poor outcomes in a cohort of 634 HCC patients. • Multiple Cox regression analysis indicated nuclear BLZF1 as an independent predictor for overall survival.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2022

When physics-informed data analytics outperforms black-box machine learning: A case study in thickness control for additive manufacturing

Aerosol jet printing (AJP) has emerged as a promising noncontact additive manufacturing method for high-resolution printing for a wide range of material systems. A key challenge limiting the broader adoption of AJP in the material science community is the lack of methods to precisely control thickness. Herein, we develop a model-based design of experiment (MBDoE) framework that integrates physics-informed models, nonlinear regression, and information criteria to postulate, select and calibrate the best model to describe and optimize the AJP manufacturing process. Starting with already available data from system commissioning (e.g., prior single variable sensitivity analysis), four candidate physics-informed models are postulated and trained. MBDoE identifies a single additional optimal experiment to validate these predictive models with quantified uncertainties, which are then used to determine the best experimental conditions to control printed film thickness. As a comparative benchmark, the analysis is repeated using the same dataset with nonparametric Gaussian process regression (GPR) model that does not incorporate physical information. Using MBDoE principles, we find that only five experiments are necessary to calibrate the nonlinear physics-informed parametric model, and with said limited data, this model outperforms the black-box machine learning GPR model. This key result underscores an emerging trend in the data science community: incorporating physical information into predictive models often drastically reduces the data requirements. Leveraging MBDoE further increased the data efficiency. By design, the proposed data science framework is general in nature and can be easily extended to other experimental and additive manufacturing systems beyond AJP.

Aerosol jet printing↗