Search

The escalating integration of machine learning in high-stakes fields such as healthcare raises substantial concerns about model fairness. We propose an interpretable framework - Fairness-Aware Interpretable Modeling (FAIM), to improve model fairness without compromising performance, featuring an interactive interface to identify a "fairer" model from a set of high-performing models and promoting the integration of data-driven evidence and clinical expertise to enhance contextualized fairness. We demonstrated FAIM's value in reducing sex and race biases by predicting hospital admission with two real-world databases, MIMIC-IV-ED and SGH-ED. We show that for both datasets, FAIM models not only exhibited satisfactory discriminatory performance but also significantly mitigated biases as measured by well-established fairness metrics, outperforming commonly used bias-mitigation methods. Our approach demonstrates the feasibility of improving fairness without sacrificing performance and provides an a modeling mode that invites domain experts to engage, fostering a multidisciplinary effort toward tailored AI fairness.

The AutoScore framework can automatically generate data-driven clinical scores in various clinical applications. Here, we present a protocol for developing clinical scoring systems for binary, survival, and ordinal outcomes using the open-source AutoScore package. We describe steps for package installation, detailed data processing and checking, and variable ranking. We then explain how to iterate through steps for variable selection, score generation, fine-tuning, and evaluation to generate understandable and explainable scoring systems using data-driven evidence and clinical knowledge. For complete details on the use and execution of this protocol, please refer to Xie et al. (2020),1 Xie et al. (2022)2, Saffari et al. (2022)3 and the online tutorial https://nliulab.github.io/AutoScore/.

The widespread use of ChatGPT and other emerging technology powered by generative artificial intelligence (GenAI) has drawn much attention to potential ethical issues, especially in high-stakes applications such as healthcare, but ethical discussions are yet to translate into operationalisable solutions. Furthermore, ongoing ethical discussions often neglect other types of GenAI that have been used to synthesise data (e.g., images) for research and practical purposes, which resolved some ethical issues and exposed others. We conduct a scoping review of ethical discussions on GenAI in healthcare to comprehensively analyse gaps in the current research, and further propose to reduce the gaps by developing a checklist for comprehensive assessment and transparent documentation of ethical discussions in GenAI research. The checklist can be readily integrated into the current peer review and publication system to enhance GenAI research, and may be used for ethics-related disclosures for GenAI-powered products, healthcare applications of such products and beyond.

Background: Risk prediction models are useful tools in clinical decision-making which help with risk stratification and resource allocations and may lead to a better health care for patients. AutoScore is a machine learning-based automatic clinical score generator for binary outcomes. This study aims to expand the AutoScore framework to provide a tool for interpretable risk prediction for ordinal outcomes. Methods: The AutoScore-Ordinal framework is generated using the same 6 modules of the original AutoScore algorithm including variable ranking, variable transformation, score derivation (from proportional odds models), model selection, score fine-tuning, and model evaluation. To illustrate the AutoScore-Ordinal performance, the method was conducted on electronic health records data from the emergency department at Singapore General Hospital over 2008 to 2017. The model was trained on 70% of the data, validated on 10% and tested on the remaining 20%. Results: This study included 445,989 inpatient cases, where the distribution of the ordinal outcome was 80.7% alive without 30-day readmission, 12.5% alive with 30-day readmission, and 6.8% died inpatient or by day 30 post discharge. Two point-based risk prediction models were developed using two sets of 8 predictor variables identified by the flexible variable selection procedure. The two models indicated reasonably good performance measured by mean area under the receiver operating characteristic curve (0.785 and 0.793) and generalized c-index (0.737 and 0.760), which were comparable to alternative models. Conclusion: AutoScore-Ordinal provides an automated and easy-to-use framework for development and validation of risk prediction models for ordinal outcomes, which can systematically identify potential predictors from high-dimensional data.

Background: Risk prediction models are useful tools in clinical decision-making which help with risk stratification and resource allocations and may lead to a better health care for patients. AutoScore is a machine learning–based automatic clinical score generator for binary outcomes. This study aims to expand the AutoScore framework to provide a tool for interpretable risk prediction for ordinal outcomes. Methods: The AutoScore-Ordinal framework is generated using the same 6 modules of the original AutoScore algorithm including variable ranking, variable transformation, score derivation (from proportional odds models), model selection, score fine-tuning, and model evaluation. To illustrate the AutoScore-Ordinal performance, the method was conducted on electronic health records data from the emergency department at Singapore General Hospital over 2008 to 2017. The model was trained on 70% of the data, validated on 10% and tested on the remaining 20%. Results: This study included 445,989 inpatient cases, where the distribution of the ordinal outcome was 80.7% alive without 30-day readmission, 12.5% alive with 30-day readmission, and 6.8% died inpatient or by day 30 post discharge. Two point-based risk prediction models were developed using two sets of 8 predictor variables identified by the flexible variable selection procedure. The two models indicated reasonably good performance measured by mean area under the receiver operating characteristic curve (0.758 and 0.793) and generalized c-index (0.737 and 0.760), which were comparable to alternative models. Conclusion: AutoScore-Ordinal provides an automated and easy-to-use framework for development and validation of risk prediction models for ordinal outcomes, which can systematically identify potential predictors from high-dimensional data.

Interpretable machine learning has been focusing on explaining final models that optimize performance. The current state-of-the-art is the Shapley additive explanations (SHAP) that locally explains variable impact on individual predictions, and it is recently extended for a global assessment across the dataset. Recently, Dong and Rudin proposed to extend the investigation to models from the same class as the final model that are "good enough", and identified a previous overclaim of variable importance based on a single model. However, this method does not directly integrate with existing Shapley-based interpretations. We close this gap by proposing a Shapley variable importance cloud that pools information across good models to avoid biased assessments in SHAP analyses of final models, and communicate the findings via novel visualizations. We demonstrate the additional insights gain compared to conventional explanations and Dong and Rudin's method using criminal justice and electronic medical records data.

INTRODUCTION: Primary care remains an underused venue for prevention and management of paediatric overweight and obesity. A prior trial demonstrated a significant impact of paediatrician/nurse practitioner (Ped/NP)-and registered dietitian (RD)-delivered motivational interviewing (MI) on child body mass index (BMI). The study described here will test the effectiveness of an enhanced version of this primary care-based MI counselling intervention on child BMI. METHODS AND ANALYSIS: This cluster randomised effectiveness trial includes 24 Ped/NPs from 18 paediatric primary care practices that belong to the American Academy of Pediatrics (AAP) national Pediatric Research in Office Settings (PROS) practice-based research network. To date, practices have been randomised (nine to intervention and nine to usual care). Intervention Ped/NPs have been trained in MI, behavioural therapy, billing/coding for weight management and study procedures. Usual care Ped/NPs received training in billing/coding and study procedures only. Children 3- 11 years old with BMI >the 85th percentile were identified via electronic health records (EHRs). Parents from intervention practices have been recruited and enrolled. Over about 2 years, these parents are offered approximately 10 MI-based counselling sessions (about four in person sessions with their childs Ped/NP and up to six telephonic sessions with a trained RD). The primary outcome is change in child BMI (defined as per cent from median BMI for age and sex) over the study period. The primary comparison is between eligible children in intervention practices whose parents enrol in the study and all eligible children in usual care practices. Data sources will include EHRs, billing records, surveys and counselling call notes. ETHICS AND DISSEMINATION: Institutional Review Board approval was obtained from the AAP. All Ped/NPs provided written informed consent, and intervention group parents provided consent and Health Insurance Portability an

Receptor activity-modifying proteins (RAMPs) have been shown to modulate the functions of several G protein-coupled receptors (GPCRs), but potential direct interactions among the three known RAMPs and hundreds of GPCRs have never been investigated. Focusing mainly on the secretin-like family of GPCRs, we engineered epitope-tagged GPCRs and RAMPs, and developed a multiplexed suspension bead array (SBA) immunoassay to detect GPCR-RAMP complexes from detergent-solubilized lysates. Using 64 antibodies raised against the native proteins and 4 antibodies targeting the epitope tags, we mapped the interactions among 23 GPCRs and 3 RAMPs. We validated nearly all previously reported secretin-like GPCR-RAMP interactions, and also found previously unidentified RAMP interactions with additional secretin-like GPCRs, chemokine receptors, and orphan receptors. The results provide a complete interactome of secretin-like GPCRs with RAMPs. The SBA strategy will be useful to search for additional GPCR-RAMP complexes and other interacting membrane protein pairs in cell lines and tissues., QC 20191115

Quantitative changes in leptin concentration lead to alterations in food intake and body weight, but the regulatory mechanisms that control leptin gene expression are poorly understood. Here we report that fat-specific and quantitative leptin expression is controlled by redundant cis elements and trans factors interacting with the proximal promoter together with a long noncoding RNA (lncOb). Diet-induced obese mice lacking lncOb show increased fat mass with reduced plasma leptin levels and lose weight after leptin treatment, whereas control mice do not. Consistent with this finding, large-scale genetic studies of humans reveal a significant association of single-nucleotide polymorphisms (SNPs) in the region of human lncOb with lower plasma leptin levels and obesity. These results show that reduced leptin gene expression can lead to a hypoleptinemic, leptin-responsive form of obesity and provide a framework for elucidating the pathogenic mechanism in the subset of obese patients with low endogenous leptin levels.

Purpose: To investigate: 1) if branched-chain amino acid supplementation (BCAA) enhances recovery from acute, eccentric resistance exercise as measured by indirect markers of muscle damage, and 2) the impact of acute, eccentric resistance exercise on markers of autophagy and heat shock protein during recovery. Methods: Twenty resistance-trained males were randomly assigned to eight days of BCAA or PLCB and adhered to a diet consisting of 1.2 g/kg/d protein. On day five, all subjects completed 10x8 eccentric squats at 70% one repetition maximum (1RM), followed by 5x20 split jumps. Plasma creatine kinase (CK), erythrocyte glutathione, vertical jump (VJ), maximal voluntary isometric contraction (MVIC), jump squat (JS), perceived soreness were measured as indirect markers of muscle damage. Autophagy (LC3-I, LC3-II, p62) and heat shock protein (HSP70) responses were assessed via peripheral blood mononuclear cells. Variables were measured immediately before, immediately post (IPE), as well as 1, 2, 4, 24, 48, and 72 hours (hr) post-exercise. Results: Plasma CK levels were significantly lower for the BCAA group at 48 hr post-exercise; however, no significant group-by-time effect was detected. Erythrocyte glutathione (GSSG/tGSH) was significantly elevated 1, 2, and 4 hr in both BCAA and PLCB groups compared to pre-exercise; however, no significant group-by-time effect was detected at any time-point. The BCAA group reported significantly less soreness compared to PLCB at 48 and 72 hr post-exercise. MVIC was significantly higher for the BCAA group at 24 hr post-exercise, but no significant group-by-time effect was observed. No significant difference between groups was detected for VJ or JS. No differences between groups were detected for protein expression of LC3-II, LC3 II/ I, p62, or HSP70 at any time-point. When combining groups, LC3-II decreased significantly 2 and 4 hr post-exercise. p62 decreased significantly IPE, 2, and 4 hr post-exercise and significantly increase

Though site specific, these data provide critical information to local CBOs and can be used as a tool for comparison with national data and Healthy People 2010 goals.

Reduced partial pressure of oxygen impairs exercise performance at altitude. Acute nitrate supplementation, at sea level may reduce oxygen cost during submaximal exercise in hypoxia. Therefore, we investigated the metabolic response during exercise at altitude following acute nitrate consumption. Ten well-trained (61.01±7.37 ml/kg/min) males (age 28±7 yr) completed 3 experimental trials (T1, T2, T3). T1 included baseline demographics, a maximal aerobic capacity test (VO2max) and five submaximal intensity cycling determination bouts at an elevation of 1600m. A 4-day dietary washout, minimizing consumption of nitrate rich foods, preceded T2 and T3. In a randomized, double-blind, placebo-controlled, crossover fashion, subjects consumed a placebo (PL) or nitrate rich (NR) beverage 2.5 hours prior to T2 and T3. Exercise at 3500m (T2 and T3) consisted of a 5-min warm-up (25% VO2max) and four 5-min cycling bouts (40, 50, 60, 70% of VO2max) each separated a 4-min rest period. Cycling RPM and watts for each submaximal bout were determined during T1. Pre-exercise plasma nitrite was elevated following NR consumption compared to PL (p<0.05). Oxygen consumption, respiratory exchange ratio, oxygen saturation, heart rate and rating of perceived exertion were not different at any submaximal intensity between NR and PL treatments. Blood lactate, however, was reduced following NR consumption compared to PL at 40 and 60% of VO2max (p<.0.05). Following exercise, compared to rest, catalase and 8-isoprostane increased at 3500m in both the placebo and nitrate rich groups (p<0.05). There was no difference between groups (placebo vs nitrate rich) for either of these markers of oxidative stress. Our findings suggest that acute nitrate supplementation prior to exercise at 3500m does not reduce oxygen cost or oxidative stress, but may reduce lactate production at lower intensity workloads.

Background: Commercially available dietary products advertised to promote weight loss are an under researched but heavily purchased commodity in the United States. Despite only limited evidence, interest in dietary supplements continues to increase. Obesity is an increasingly prevalent and preventable morbidity with multiple behavioral, surgical and pharmacological interventions currently available. Commercially available dietary supplements are often advertised to stimulate metabolism and cause rapid weight and/or fat loss, although few well controlled studies have demonstrated such effects. This work uniquely summarizes the current evidence evaluating the efficacy of several over-the-counter thermogenic products for their effects on resting energy expenditure. Additionally, this work outlines the important therapeutic benefits provided by dietary stimulators of metabolism and peroxisome proliferator-activated receptor gamma co-activator 1 (PGC-1) in skeletal muscle. Our work uniquely describes the effects of a commercially available dietary supplement on resting metabolic rate in humans as well as the metabolic and biochemical effects in vitro. Methods: Human rhabdomyosarcoma cells (RD) and mouse myoblasts (C2C12) were cultured under standard conditions and treated with various doses of a commercially available supplement (RF) for various durations and assessed for changes in metabolism, metabolic gene expression and mitochondrial content. Additionally, human subjects ingested either placebo or RF in a double-blind placebo controlled fashion and metabolic rate and blood pressure were measured at 3 time points for 3 hours post-ingestion. Results: RF enhanced metabolism, metabolic gene expression, and mitochondrial content in both cell models. RF also enhanced energy expenditure in human male subjects without altering substrate utilization. RF also significantly increased systolic blood pressure. Conclusion: RF appears to increase metabolism immediately following in

Concerned with the management of complex long-term engineering projects, this important volume, of great interest to postgraduate students of business, technology management and engineering, reports on a set of rich, novel and unique findings concerning the conduct and management of three high profile and complex projects. The major investments which constitute complex long-term projects represent an increasingly important source of economic activity, often with particularly significant consequences for economic growth and public policy. This informative volume expertly contributes to broader debates concerning new organizational forms, knowledge management and organizational learning and the management of innovation in project-based settings.

Obesity is one of the most prevalent maladies in the United States and is a major cause of preventable death. Weight loss supplements frequently claim uncoupling as a mechanism of action. Uncoupling agents could be used for weight loss because they disrupt mitochondrial metabolism thereby reducing adenosine triphosphate (ATP) yield. Consequently, metabolic efficiency diminishes increasing basal metabolic rate. 2,4-Dinitrophenol (DNP) successfully uncouples but was banned in 1938 because of a narrow window between efficacy and toxicity. PURPOSE: To measure the ability of a blend that reportedly contains 17-dihydroxy-delta-5-etiocholane-7-one and p-methylcarbonylethylphenol and other substances (A7E), a purported uncoupling agent, to interfere with oxidative phosphorylation in Saccharomyces cerevisiae as evidenced by lower ATP production. METHODS: Timed culture studies of S. cerevisiae were performed using two separate agents, A7E (a purported uncoupling agent) and DNP (a known uncoupling agent) at three doses (DNP: Low, Moderate, High; A7E: Low, Moderate, High ), and an ethanol-treated control to detect interference with mitochondrial coupling. Microbial staining was used to ascertain cell viability and any changes in cell densities. ATP production was estimated by measuring luminosity generated in culture supernatants using the ATP Bioluminescence Kit (Sigma St. Louis, MO.). RESULTS: Luminosity measurements estimating ATP production revealed statistically lower ATP in agent-treated supernatants than in control supernatants (p < 0.001) except for low dose A7E versus control (p > .05), suggesting that both A7E and DNP acted by a mechanism of uncoupling. Luminosity values were measured in relative luminosity units (RLU). Treatments with A7E at Low, Moderate, and High doses generated group mean luminosity values of 24,596, 16,038, and 6,969, respectively. Treatments with DNP at Low, Moderate, and High doses generated group mean luminosity values of 17,191, 11,901, 767 R

Background - Smoking rates are projected to increase substantially in developing countries such as South Africa. Purpose - The aim of this study was to test the efficacy of two contrasting approaches to school-based smoking prevention in South African youth compared to the standard health education program. One experimental program was based on a skills training/peer resistance model and the other on a harm minimization model. Method - Thirty-six public schools from two South African provinces, KwaZulu-Natal and the Western Cape, were stratified by socioeconomic status and randomized to one of three groups. Group 1 (comparison) schools (n = 12) received usual tobacco use education. Group 2 schools (n = 12) received a harm minimization curriculum in grades 8 and 9. Group 3 schools (n = 12) received a life skills training curriculum in grades 8 and 9. The primary outcome was past month use of cigarettes based on a self-reported questionnaire. Result - Five thousand two hundred sixty-six students completed the baseline survey. Of these, 4,684 (89%) completed at least one follow-up assessment. The net change in 30-day smoking from baseline to 2-year follow-up in the control group was 6% compared to 3% in both harm minimization (HM) and life skills training (LST) schools. These differences were not statistically significant. Intervention response was significantly moderated by both gender and race. The HM intervention was more effective for males, whereas the life skills intervention was more effective for females. For black African students, the strongest effect was evident for the HM intervention, whereas the strongest intervention effect for “colored” students was evident for the LST group. Conclusion - The two experimental curricula both produced similar overall reductions in smoking prevalence that were not significantly different from each other or the control group. However, the impact differed by gender and race, suggesting a need to tailor tobacco and drug use p

The role of perceptual frames and reframing in the context of the shifting demand for goods and services is explored. The case of Alstom Transport's building of the Pendolino tilting train for Virgin Trains' West Coast Mainline route is used as an example of how framing and reframing can be a key management tool in contexts where customers are driving new strategic directions in the delivery of goods and services. We see the successful management of transitional projects (projects which depart from existing knowledge and capabilities to deliver transformed goods and services) as dependent on the ability of project managers, and other key actors, to reshape and realign existing perceptual frames. The paper examines the difficulties of so doing.

The article considers knowledge management issues from the client's perspective. In the example presented, a sludge treatment centre procured by Northumbrian Water Ltd (NWL), the task faced by the client was to manage knowledge in a context where the core technology being procured was new and resulted in the need for new knowledge to be created and shared both pre- and post-delivery. In exploring these issues, the article reveals the problems of (and some solutions to) managing knowledge across the project life-cycle and between different groups, where the motivation for generating and sharing knowledge was not the same for all participants.

This paper explores some basic issues relevant to the use of public funds to support the production and delivery of cultural services, such as the policy goals of federal assistance to the arts and how the way in which support is provided help in the attainment of these goals.

The purpose of this paper is to provide an economic perspective on the problem of transporting people to Gateway National Recreation Area located in New York City and northeastern New Jersey. While it does not contain any detailed empirical calculations for the solution to this complex issue, it is hoped that some of the suggestions might be useful input into the planning process, and might open the way to more detailed research and analysis.

While there are insufficient data to reach any detailed planning and development conclusions concerning Delaware Water Gap National Recreation Area (DWGNRA), an examination of some of the problems from the perspective of economic theory can lead to some insights concerning the provision of recreation services at DWGNRA.

The value of noise attenuation policies depends upon the costs that noise imposes. There is little or no information on the medical or psychological impact that noise levels as low as those experienced near expressways (between 65 and 75 dBA) have on human health and well-being. Instead, the nuisance cost of noise was estimated by determining the decline in property values that could be attributed to the level of noise in a neighborhood. Noise levels were measured at 233 locations in the city of Chicago outside homes whose sale price was known. The noise index used corresponds to the new EPA measure of Equivalent Sound Level. The results of the analysis suggest that there is no appreciable nuisance value to noise levels below 50 dBA and that the level of damages are related to the price of the home.

The report is intended primarily as a discussion of the appropriate way to measure the benefits of recreational open space. Topics discussed include the following: Public responsibility for urban recreation; Placing a value on recreational facilities; Predicting facility use; The estimation of the demand for recreation; The use of land values to value recreation sites; A synthesis: cost-benefit analysis of an urban park.

Dicogr.-filmogr. p. 289-333. Index, L'auteur rappelle les étapes de la vie de Karajan et montre surtout sa façon de travailler, sa conception de la direction d'orchestre et de la musique. Un ouvrage très fouillé et intéressant.

More than 10 years on since the launch of the Internet, there are clear differential levels of adoption of the Internet for purchasing leisure travel across countries. In some countries, such as Portugal, only a minority of travel purchasing is conducted over the Internet. This paper aims to contribute to a greater understanding of adoption of purchasing over the Internet by evaluating the determinants of intention to adopt the Internet for purchasing leisure travel. A number of variables are used to assess which factors influence intention, including the attributes of purchasing over the Internet (relative advantage and complexity), involvement, and the stage in the e-commerce adoption path. Kruskal-Wallis and logistic regression were employed and the results indicated that relative advantage, involvement and stage in the e-commerce adoption path were important predictors of the intention to purchase leisure travel over the Internet, while complexity was not.