Incorporating artificial intelligence in urology: Supervised machine learning algorithms demonstrate comparative advantage over nomograms in predicting biochemical recurrence after prostatectomy.

OBJECTIVE: After radical prostatectomy (RP), one-third of patients will experience biochemical recurrence (BCR), which is associated with subsequent metastasis and cancer-specific mortality. We employed machine learning (ML) algorithms to predict BCR after RP, and compare them with traditional regression models and nomograms. METHODS: Utilizing a prospective Uro-oncology registry, 18 clinicopathological parameters of 1130 consecutive patients who underwent RP (2009-2018) were recorded, yielding over 20,000 data points for analysis. The data set was split into a 70:30 ratio for training and validation. Three ML models: Naïve Bayes (NB), random forest (RF), and support vector machine (SVM) were studied, and compared with traditional regression models and nomograms (Kattan, CAPSURE, John Hopkins [JHH]) to predict BCR at 1, 3, and 5 years. RESULTS: Over a median follow-up of 70.0 months, 176 (15.6%) developed BCR, at a median time of 16.0 months (interquartile range [IQR]: 11.0-26.0). Multivariate analyses demonstrated strongest association of BCR with prostate-specific antigen (PSA) (p: 0.015), positive surgical margins (p < 0.001), extraprostatic extension (p: 0.002), seminal vesicle invasion (p: 0.004), and grade group (p < 0.001). The 3 ML models demonstrated good prediction of BCR at 1, 3, and 5 years, with the area under curves (AUC) of NB at 0.894, 0.876, and 0.894, RF at 0.846, 0.875, and 0.888, and SVM at 0.835, 0.850, and 0.855, respectively. All models demonstrated (1) robust accuracy (>0.82), (2) good calibration with minimal overfitting, (3) longitudinal consistency across the three time points, and (4) inter-model validity. The ML models were comparable to traditional regression analyses (AUC: 0.797, 0.848, and 0.862) and outperformed the three nomograms: Kattan (AUC: 0.815, 0.798, and 0.799), JHH (AUC: 0.820, 0.757, and 0.750) and CAPSURE nomograms (AUC: 0.706, 0.720, and 0.749) (p < 0.001). CONCLUSION: Supervised ML algorithms can deliver accurate performances and outperform nomograms in predicting BCR after RP. This may facilitate tailored care provisions by identifying high-risk patients who will benefit from multimodal therapy.

Application of Artificial Intelligence for Diagnosis and Risk Stratification in NAFLD and NASH: The State of the Art.

The diagnosis of nonalcoholic fatty liver disease and associated fibrosis is challenging given the lack of signs, symptoms and nonexistent diagnostic test. Furthermore, follow up and treatment decisions become complicated with a lack of a simple reproducible method to follow these patients longitudinally. Liver biopsy is the current standard to detect, risk stratify and monitor individuals with nonalcoholic fatty liver disease. However, this method is an unrealistic option in a population that affects about one in three to four individuals worldwide. There is an urgency to develop innovative methods to facilitate management at key points in an individual's journey with nonalcoholic fatty liver disease fibrosis. Artificial intelligence is an exciting field that has the potential to achieve this. In this review, we highlight applications of artificial intelligence by leveraging our current knowledge of nonalcoholic fatty liver disease to diagnose and risk stratify NASH phenotypes.

Population susceptibility: A vital consideration in chemical risk evaluation under the Lautenberg Toxic Substances Control Act.

The 2016 Frank Lautenberg Chemical Safety for the 21st Century Act (Lautenberg TSCA) amended the 1976 Toxic Substances Control Act (TSCA) to mandate protection of susceptible and highly exposed populations. Program implementation entails a myriad of choices that can lead to different degrees of public health protections. Well-documented exposures to multiple industrial chemicals occur from air, soil, water, food, and products in our workplaces, schools, and homes. Many hazardous chemicals are associated with or known to cause health risks; for other industrial chemicals, no data exist to confirm their safety because of flaws in 1976 TSCA. Under the 2016 Lautenberg amendments, the United States Environmental Protection Agency (EPA) must evaluate chemicals against risk-based safety standards under enforceable deadlines, with an explicit mandate to identify and assess risks to susceptible and highly exposed populations. Effective public health protection requires EPA to implement the Lautenberg TSCA requirements by incorporating intrinsic and extrinsic factors that affect susceptibility, adequately assessing exposure among vulnerable groups, and accurately identifying highly exposed groups. We recommend key scientific and risk assessment principles to inform health-protective chemical policy such as consideration of aggregate exposures from all pathways and, when data are lacking, the use of health-protective defaults.

Radiomics in neuro-oncology: Basics, workflow, and applications.

Over the last years, the amount, variety, and complexity of neuroimaging data acquired in patients with brain tumors for routine clinical purposes and the resulting number of imaging parameters have substantially increased. Consequently, a timely and cost-effective evaluation of imaging data is hardly feasible without the support of methods from the field of artificial intelligence (AI). AI can facilitate and shorten various time-consuming steps in the image processing workflow, e.g., tumor segmentation, thereby optimizing productivity. Besides, the automated and computer-based analysis of imaging data may help to increase data comparability as it is independent of the experience level of the evaluating clinician. Importantly, AI offers the potential to extract new features from the routinely acquired neuroimages of brain tumor patients. In combination with patient data such as survival, molecular markers, or genomics, mathematical models can be generated that allow, for example, the prediction of treatment response or prognosis, as well as the noninvasive assessment of molecular markers. The subdiscipline of AI dealing with the computation, identification, and extraction of image features, as well as the generation of prognostic or predictive mathematical models, is termed radiomics. This review article summarizes the basics, the current workflow, and methods used in radiomics with a focus on feature-based radiomics in neuro-oncology and provides selected examples of its clinical application.

Implementing organ-on-chip in a next-generation risk assessment of chemicals: a review.

Organ-on-chip (OoC) technology is full of engineering and biological challenges, but it has the potential to revolutionize the Next-Generation Risk Assessment of novel ingredients for consumer products and chemicals. A successful incorporation of OoC technology into the Next-Generation Risk Assessment toolbox depends on the robustness of the microfluidic devices and the organ tissue models used. Recent advances in standardized device manufacturing, organ tissue cultivation and growth protocols offer the ability to bridge the gaps towards the implementation of organ-on-chip technology. Next-Generation Risk Assessment is an exposure-led and hypothesis-driven tiered approach to risk assessment using detailed human exposure information and the application of appropriate new (non-animal) toxicological testing approaches. Organ-on-chip presents a promising in vitro approach by combining human cell culturing with dynamic microfluidics to improve physiological emulation. Here, we critically review commercial organ-on-chip devices, as well as recent tissue culture model studies of the skin, intestinal barrier and liver as the main metabolic organ to be used on-chip for Next-Generation Risk Assessment. Finally, microfluidically linked tissue combinations such as skin-liver and intestine-liver in organ-on-chip devices are reviewed as they form a relevant aspect for advancing toxicokinetic and toxicodynamic studies. We point to recent achievements and challenges to overcome, to advance non-animal, human-relevant safety studies.

Biosafety during a pandemic: shared resource laboratories rise to the challenge.

Biosafety has always been an important aspect of daily work in any research institution, particularly for cytometry Shared Resources Laboratories (SRLs). SRLs are common-use spaces that facilitate the sharing of knowledge, expertise, and ideas. This sharing inescapably involves contact and interaction of all those within this working environment on a daily basis. The current pandemic caused by SARS-CoV-2 has prompted the re-evaluation of many policies governing the operations of SRLs. Here we identify and review the unique challenges SRLs face in maintaining biosafety standards, highlighting the potential risks associated with not only cytometry instrumentation and samples, but also the people working with them. We propose possible solutions to safety issues raised by the COVID-19 pandemic and provide tools for facilities to adapt to evolving guidelines and future challenges.

Evolving Science and Practice of Risk Assessment.

Managing public health risks from environmental contaminants has historically relied on a risk assessment process defined by the regulatory context in which these risks are assessed. Risk assessment guidance follows a straightforward, chemical-by-chemical approach to inform regulatory decisions around the question "what is the risk-based concentration protective of human and ecological health outcomes?" Here we briefly summarize regulatory risk assessment in the context of innovative risk assessment approaches based on an evolving understanding of the underlying scientific disciplines that support risk analysis more broadly. We discuss scientific versus regulatory tensions in the application of these approaches for future risk assessments, and challenges in translating our improved understanding of the underlying scientific complexity to the regulatory landscape to better inform decision making that extends beyond conventional regulatory mandates.

Echocardiographic Assessments for Peripartum Cardiac Events in Pregnant Women with Low-Risk Congenital Heart Disease.

This retrospective cohort study aimed to explore the relationship between temporal changes in the cardiac function and peripartum cardiac events in pregnant women with low-risk congenital heart disease.We performed echocardiography at early and late pregnancy and postpartum in 76 pregnant women with low-risk congenital heart disease, and compared echocardiographic parameters between subjects with and without peripartum cardiac events. Median age at delivery was 27 (range, 24-31) years. The ZAHARA and CARPREG II scores suggested that most women were found to be at low-risk for pregnancy. Fifteen subjects had cardiac events that included heart failure in 10, arrhythmia in 4, and pulmonary hypertension in one subject. The left ventricular and atrial volumes significantly increased from early pregnancy toward late pregnancy, and the E/A ratio and global longitudinal strain significantly decreased from early pregnancy toward late pregnancy. The left atrial volume (67 [53-79] versus 45 [35-55] mL, P = 0.002) and plasma brain natriuretic peptide level (58 [36-123] versus 34 [18-48] pg/mL, P = 0.026) at late pregnancy were significantly higher in subjects with cardiac events than in those without cardiac events.An increase in the left atrial volume followed by mild left ventricular diastolic dysfunction is related to peripartum cardiac events in women with congenital heart disease who are at low risk for cardiac events during pregnancy.

The Known and Unknown about the Environmental Safety of Nanomaterials in Commerce.

The widespread nanomaterial use in commercial products has fed significant concern over environmental health and safety ramifications. Initially, little was known as to how these highly reactive particulates interacted with biological systems. Nanomaterials have introduced complexities not normally considered in traditional safety assessments of chemicals and therefore have generated uncertainty in the reliability of standard tests of safety. Advances in understanding the potential impacts of nanomaterials have occurred since their introduction, particularly for those used in the greatest quantities in commerce. The impact of characteristics such as charge, size, surface functionalization, chemical composition, and certain transformations on the potential effect of nanomaterials in the environment continue to move the field forward. However, generalizations of risk based on any one factor across nanomaterials is not possible. Estimating risk also remains difficult due to the introduction of materials that are new and more complex, minimal information on the specific molecular interactions of nanomaterials and organisms, and the need for more tools for measuring the dynamics of nanomaterial state and fate in complex matrices. Finally, exposure estimates are difficult due to difficulty of environmental monitoring which may be exacerbated by lack of information on nanomaterials in products and new uses in the marketplace.

Associations of Advance Directive Knowledge, Attitudes, and Barriers/Benefits With Preferences for Advance Treatment Directives Among Patients With Heart Failure and Their Caregivers.

BACKGROUND: Patients with heart failure (HF) have not been considered as major beneficiaries of advance directives (ADs). We analyzed factors affecting the preferences for the adoption of ADs by patients with HF and their caregivers. METHODS AND RESULTS: Seventy-one patient (mean age: 68 years)-caregiver (mean age: 55 years) dyads were enrolled during clinic visits for routine care at a single institution and completed questionnaires during in-person visits. Cohen's kappa coefficients and generalized estimating equation models were used to analyze the data. The agreement on dyadic perspectives for aggressive treatments was poor or fair, whereas agreement relative to hospice care was moderate (k = 0.42, 95% confidence interval = 0.087-0.754). Both patients and caregivers demonstrated poor knowledge of ADs and similar levels of perceived benefits and barriers to advance care planning. However, the caregivers had more positive attitudes toward ADs than patients. Patients and caregivers who were older and/or males had greater odds of preferring aggressive treatments and/or hospice care. Further, those with depressive symptoms had lower odds of preferring hospice care. CONCLUSION: The dyadic agreement was moderately high only for hospice care preferences. Both patients and caregivers demonstrated knowledge of shortfalls regarding ADs. Timely AD discussions could increase dyadic agreement and enhance informed and shared decision-making regarding medical care.

The unsteady state and inertia of chemical regulation under the US Toxic Substances Control Act.

After 40 years, the 1976 US Toxic Substances Control Act (TSCA) was revised under the Frank R. Lautenberg Chemical Safety for the 21st Century Act. Its original goals of protecting the public from hazardous chemicals were hindered by complex and cumbersome administrative burdens, data limitations, vulnerabilities in risk assessments, and recurring corporate lawsuits. As a result, countless chemicals were entered into commercial use without toxicological information. Few chemicals of the many identified as potential public health threats were regulated or banned. This paper explores the factors that have worked against a comprehensive and rational policy for regulating toxic chemicals and discusses whether the TSCA revisions offer greater public protection against existing and new chemicals.

A national population-based cohort study to investigate inequalities in maternal mortality in the United Kingdom, 2009-17.

BACKGROUND: Disparities have been documented in maternal mortality rates between women from different ethnic, age and socio-economic groups in the UK. It is unclear whether there are differential changes in these rates amongst women from different groups over time. The objectives of this analysis were to describe UK maternal mortality rates in different age, ethnic and socio-economic groups between 2009 and 2017, and to identify whether there were changes in the observed inequalities, or different trends amongst population subgroups. METHODS: Maternal mortality rates with 95% confidence intervals (CI) in specific age, deprivation and ethnic groups were calculated using numbers of maternal deaths as numerator and total maternities as denominator. Relative risks (RR) with 95% CI were calculated and compared using ratios of relative risk. Change over time was investigated using non-parametric tests for trend across ordered groups. RESULTS: Women from black and Asian groups had a higher mortality rate than white women in most time periods, as did women aged 35 and over and women from the most deprived quintile areas of residence. There was evidence of an increasing trend in maternal mortality amongst black women and a decrease in mortality amongst women from the least deprived areas, but no trends over time in any of the other ethnic, age or IMD groups were seen. There was a widening of the disparity between black and white women (RR 2.59 in 2009-11 compared with 5.27 in 2015-17, ratio of the relative risks 2.03, 95% CI 1.11, 3.72). CONCLUSIONS: The clear differences in the patterns of maternal mortality amongst different ethnic, age and socio-economic groups emphasise the importance of research and policies focussed specifically on women from black and minority ethnic groups, together with other disadvantaged groups, to begin to reduce maternal mortality in the UK.

Nonclinical safety assessment of epigenetic modulatory drugs: Current status and industry perspective.

Pharmaceutic products designed to perturb the function of epigenetic modulators have been approved by regulatory authorities for treatment of advanced cancer. While the predominant effort in epigenetic drug development continues to be in oncology, non-oncology indications are also garnering interest. A survey of pharmaceutical companies was conducted to assess the interest and concerns for developing small molecule direct epigenetic effectors (EEs) as medicines. Survey themes addressed (1) general levels of interest and activity with EEs as therapeutic agents, (2) potential safety concerns, and (3) possible future efforts to develop targeted strategies for nonclinical safety assessment of EEs. Thirteen companies contributed data to the survey. Overall, the survey data indicate the consensus opinion that existing ICH guidelines are effective and appropriate for nonclinical safety assessment activities with EEs. Attention in the framework of study design should, on a case by case basis, be considered for delayed or latent toxicities, carcinogenicity, reproductive toxicity, and the theoretical potential for transgenerational effects. While current guidelines have been appropriate for the nonclinical safety assessments of epigenetic targets, broader experience with a wide range of epigenetic targets will provide information to assess the potential need for new or revised risk assessment strategies for EE drugs.

Trends in the prevalence of cardiometabolic disease and cardiovascular events by body mass index category in adults from 1999 to 2016.

OBJECTIVES: An increasing percentage of the US population is obese. Cardiometabolic risk in the population increases with body mass index (BMI), but whether this correlation changes over time is unknown. We analysed the National Health and Nutrition Examination Survey (NHANES) database from 1999 to 2016 to determine if the prevalence of cardiometabolic disease and cardiovascular events within each BMI category is changing over time as the BMI of the population is increasing. STUDY DESIGN: For each of the nine survey cycles covering this period, we divided the population by BMI category (normal, overweight, class 1 obesity, class ≥2 obesity) and subsequently by the presence of cardiovascular events or cardiometabolic disease. NHANES participants are a group of 5000 individuals/cycle selected to be representative of the US population. We used the weighted data sets to perform trend analyses for each risk/BMI group adjusted for relevant confounders. RESULTS: The distribution of the highest risk category (cardiovascular event) has not changed over time within any BMI category. The distribution of the lowest risk category (cardiometabolically healthy) increased significantly over time in all BMI categories. This was noted in the 18- to 45-year subgroup but not in the group aged >45 years. CONCLUSIONS: The increase in the prevalence of overweight and obese individuals might be associated with a 'healthy obesity' phenotype in those <45 years; however, individuals >45 years showed a proportional increase in associated cardiometabolic risk.

Risk assessment for prolonged sickness absence due to musculoskeletal disorders: protocol for a prospective cohort study.

BACKGROUND: Musculoskeletal disorders are the leading cause of sickness absence and disability pension in Norway. There is strong evidence that long-term sickness absence due to musculoskeletal disorders are associated with a reduced probability of return to work (RTW). A way to meet the economic and resource-demanding challenges related to individual follow-up of this group is to identify and treat those individuals with a high risk of prolonged sickness. The overall purposes of this project are 1) to determine the most accurate screening tool to identify people at a high risk of prolonged sickness absence due to an musculoskeletal disorder, and 2) to investigate severity of musculoskeletal health, health-related quality-of-life, health care utilization, and costs across different risk profiles in people on sick leave due to a musculoskeletal disorder. METHODS: People older than 18 years of age on sick leave for at least 4 weeks due to a musculoskeletal disorder will be invited to participate in this prospective observational cohort study conducted within the Norwegian Welfare and Labor Administration (NAV) system in collaboration with OsloMet - Oslo Metropolitan University. The main outcome is sickness absence, obtained from the NAV registry. Data on sickness absence will be retrieved prospectively in the period from study inclusion to 12 months follow-up, and retrospectively 12 months before inclusion in the study. Possible risk factors will be self-reported by the participants at inclusion while health care utilization will be retrieved from registry data. To conduct analyses including 15 to 20 predictor variables, we aim at including 500-600 people on sick leave due to musculoskeletal disorders. DISCUSSION: This study may provide tools that can be used to identify individuals with high risk of prolonged sickness absence and may thus be important from both a socioeconomic and individual perspective. Further, the study may give valuable insight into identification of sickness absence profiles and the associations between these profiles and musculoskeletal health status, health-related quality of life and costs. TRIAL REGISTRATION: Retrospectively registered in ClinicalTrials.gov (NCT04196634, 27.11.2019).

Improving Health Risk Assessment as a Basis for Public Health Decisions in the 21st Century.

One-fifth of the way through the 21st century, a commonality of factors with those of the last 50 years may offer the opportunity to address unfinished business and current challenges. The recommendations include: (1) Resisting the tendency to oversimplify scientific assessments by reliance on single disciplines in lieu of clear weight-of-evidence expressions, and on single quantitative point estimates of health protective values for policy decisions; (2) Improving the separation of science and judgment in risk assessment through the use of clear expressions of the range of judgments that bracket protective quantitative levels for public health protection; (3) Use of comparative risk to achieve the greatest gains in health and the environment; and (4) Where applicable, reversal of the risk assessment and risk management steps to facilitate timely and substantive improvements in public health and the environment. Lessons learned and improvements in the risk assessment process are applied to the unprecedented challenges of the 21st century such as, pandemics and climate change. The beneficial application of the risk assessment and risk management paradigm to ensure timely research with consistency and transparency of assessments is presented. Institutions with mandated stability and leadership roles at the national and international levels are essential to ensure timely interdisciplinary scientific assessment at the interface with public policy as a basis for organized policy decisions, to meet time sensitive goals, and to inform the public.

Constructing Inpatient Pressure Injury Prediction Models Using Machine Learning Techniques.

The incidence rate of pressure injury is a critical nursing quality indicator in clinic care; consequently, factors causing pressure injury are diverse and complex. The early prevention of pressure injury and monitoring of these complex high-risk factors are critical to reduce the patients' pain, prevent further surgical treatment, avoid prolonged hospital stay, decrease the risk of wound infection, and lower associated medical costs and expenses. Although a number of risk assessment scales of pressure injury have been adopted in various countries, their criteria are set for specific populations, which may not be suitable for the medical care systems of other countries. This study constructs three prediction models of inpatient pressure injury using machine learning techniques, including decision tree, logistic regression, and random forest. A total of 11 838 inpatient records were collected, and 30 sets of training samples were adopted for data analysis in the experiment. The experimental results and evaluations of the models suggest that the prediction model built using random forest had most favorable classification performance of 0.845. The critical risk factors for pressure injury identified in this study were skin integrity, systolic blood pressure, expression ability, capillary refill time, and level of consciousness.

Madness, sex, and risk: A poststructural analysis.

The body of the one deemed mad often remains a sexual body with sexual needs. Mental health services respond to these demands of the body in various ways, including constructing rules around physical movement. In this context, we were interested in how mental health clinicians problematized the sexual needs and practices of residents of a long-stay mental health rehabilitation facility and how solutions were constructed in relation to the residents' sexual desires. This paper reports findings from mental health clinicians, as part of a case study where we addressed this question. Mental health clinicians responded to residents' sexuality from a discourse of risk. The notion of the engagement with each resident as a sexual subject was subordinated to the paternalistic notion of protecting the patient from risk. The resident became an object to be monitored and protected rather than a subject with sexual desire and agency. This paternalism also showed itself in relation to a 'no-sex on-site' rule that allowed for a shift of risk from the organization to the resident. Residents, rather than having a relatively safe place to have sexual relations, were required to find a place elsewhere, potentially unsafe, outside the facility.

Treatment of latent infection to achieve tuberculosis elimination in low-incidence countries.

In a Perspective for the Tuberculosis Special Issue, Kevin Schwartzman and colleagues discuss the choices and implications for personal versus public health benefits when pursuing tuberculosis elimination in low-incidence countries.