Predicting incident cardiovascular disease among African-American adults: A deep learning approach to evaluate social determinants of health in the Jackson heart study

The present study sought to leverage machine learning approaches to determine whether social determinants of health improve prediction of incident cardiovascular disease (CVD). Participants in the Jackson Heart study with no history of CVD at baseline were followed over a 10-year period to determine first CVD events (i.e., coronary heart disease, stroke, heart failure). Three modeling algorithms (i.e., Deep Neural Network, Random Survival Forest, Penalized Cox Proportional Hazards) were used to evaluate three feature sets (i.e., demographics and standard/biobehavioral CVD risk factors [FS1], FS1 combined with psychosocial and socioeconomic CVD risk factors [FS2], and FS2 combined with environmental features [FS3]) as predictors of 10-year CVD risk. Contrary to hypothesis, overall predictive accuracy did not improve when adding social determinants of health. However, social determinants of health comprised eight of the top 15 predictors of first CVD events. The social determinates of health indicators included four socioeconomic factors (insurance status and types), one psychosocial factor (discrimination burden), and three environmental factors (density of outdoor physical activity resources, including instructional and water activities; modified retail food environment index excluding alcohol; and favorable food stores). Findings suggest that whereas understanding biological determinants may identify who is currently at risk for developing CVD and in need of secondary prevention, understanding upstream social determinants of CVD risk could guide primary prevention efforts by identifying where and how policy and community-level interventions could be targeted to facilitate changes in individual health behaviors.

Removal of chloride ion from drinking water using Ag NPs-Modified bentonite: Characterization and optimization of effective parameters by response surface methodology-central composite design

The presence of chloride ion as an environmental pollutant is having a devastating and irreversible effect on aquatic and terrestrial ecosystems. To ensure safe and clean drinking water, it is vital to remove this substance using non-toxic and eco-friendly methods. This study presents a novel and highly efficient Ag NPs-modified bentonite adsorbent for removing chloride ion, a common environmental pollutant, from drinking water using a facile approach. The surface chemical properties and morphology of the pristine Na-bentonite and Ag NPs-Modified bentonite were characterized by field emission scanning electron microscopy (FESEM) and X-ray spectroscopy (EDX), X-Ray diffraction (XRD), Fourier transform infrared (FTIR), and zeta potential (ζ). To achieve maximum chloride ion removal, the effects of experimental parameters, including adsorbent dosage (1-9 g/L), chloride ion concentration (100-900 mg/L), and reaction time (5-25 h), were examined using the Response Surface Methodology (RSM). The chloride ion removal of 90% was obtained at optimum conditions (adsorbent dosage: 7 g/L, chloride ion concentration: 500 mg/L, and reaction time: 20 h). The adsorption isotherm and kinetics results indicated that the Langmuir isotherm model and pseudo-second-order kinetics were found suitable to chloride ion removal. Additionally, the regeneration and reusability of the Ag NPs-modified bentonite were further studied. In the regeneration and reusability study, the Ag NPs-modified bentonite has shown consistently ≥90% and ≥87% chloride ion removal even up to 2 repeated cycles, separately. Thus, the findings in this study provided convincing evidence for using Ag-NPs modified bentonite as a high-efficiency and promising adsorbent to remove chloride ion from drinking water.

Assessing the effects of therapeutic combinations on SARS-CoV-2 infected patient outcomes: A big data approach

BACKGROUND

The COVID-19 pandemic has demonstrated the need for efficient and comprehensive, simultaneous assessment of multiple combined novel therapies for viral infection across the range of illness severity. Randomized Controlled Trials (RCT) are the gold standard by which efficacy of therapeutic agents is demonstrated. However, they rarely are designed to assess treatment combinations across all relevant subgroups. A big data approach to analyzing real-world impacts of therapies may confirm or supplement RCT evidence to further assess effectiveness of therapeutic options for rapidly evolving diseases such as COVID-19.

METHODS

Gradient Boosted Decision Tree, Deep and Convolutional Neural Network classifiers were implemented and trained on the National COVID Cohort Collaborative (N3C) data repository to predict the patients' outcome of death or discharge. Models leveraged the patients' characteristics, the severity of COVID-19 at diagnosis, and the calculated proportion of days on different treatment combinations after diagnosis as features to predict the outcome. Then, the most accurate model is utilized by eXplainable Artificial Intelligence (XAI) algorithms to provide insights about the learned treatment combination impacts on the model's final outcome prediction.

RESULTS

Gradient Boosted Decision Tree classifiers present the highest prediction accuracy in identifying patient outcomes with area under the receiver operator characteristic curve of 0.90 and accuracy of 0.81 for the outcomes of death or sufficient improvement to be discharged. The resulting model predicts the treatment combinations of anticoagulants and steroids are associated with the highest probability of improvement, followed by combined anticoagulants and targeted antivirals. In contrast, monotherapies of single drugs, including use of anticoagulants without steroid or antivirals are associated with poorer outcomes.

CONCLUSIONS

This machine learning model by accurately predicting the mortality provides insights about the treatment combinations associated with clinical improvement in COVID-19 patients. Analysis of the model's components suggests benefit to treatment with combination of steroids, antivirals, and anticoagulant medication. The approach also provides a framework for simultaneously evaluating multiple real-world therapeutic combinations in future research studies.

Predicting temporal and spatial variability in flood vulnerability and risk of rural communities at the watershed scale

Due to land-use and hydrology changes, people are constantly exposed to floods. The adverse impact of floods is greater on vulnerable populations that disproportionately inhabit flood-prone areas. This paper reports a comprehensive study on flood vulnerability of flood prone areas in residential areas of the Tajan watershed, Iran in two periods before 2006 and after 2006. Flood prone area were determined by the random forest (RF) and K-nearest neighbor (KNN) machine learning methods. To reduce time and cost, the vulnerability was assessed only in areas with very high flood hazard using 4 main criteria (social, policy, economic, infrastructure), 40 items, and 210 questionnaires across 40 villages. Independent t-test, Kruskal-Wallis, and paired t-test were used for statistical analysis of questionnaire data. The results of machine learning models (MLMs) showed that the RF model with AUC = 0.92% is more accurate in determining flood prone areas. The results of paired t-test showed that the three criteria of social (mean P1 = 2.97 and P2 = 3.35), infrastructure (mean P1 = 2.88 and P2 = 3.25), and policy (mean P1 = 3.02 and P2 = 3.50) had significant changes in both periods. The Kruskal-Wallis test also revealed the mean of all four criteria in both periods and all sub-watersheds, except three sub-watersheds 10 (Khalkhil village), 19 (Tellarem and Kerasp villages), and 23 (Dinehsar and Jafarabad), had a significant difference. The results of the t-test also showed a decrease in vulnerability in the second period (before 2006) compared to the first period (after 2006), so the number of sub-watersheds in the very high vulnerability class was more in the first period than in the second period. A vulnerability map was developed using three factors of risk zone area, area of each sub-watershed, and population of each sub-watershed.

An ordinal severity scale for COVID-19 retrospective studies using Electronic Health Record data

Objectives

Although the World Health Organization (WHO) Clinical Progression Scale for COVID-19 is useful in prospective clinical trials, it cannot be effectively used with retrospective Electronic Health Record (EHR) datasets. Modifying the existing WHO Clinical Progression Scale, we developed an ordinal severity scale (OS) and assessed its usefulness in the analyses of COVID-19 patient outcomes using retrospective EHR data.

Materials and Methods

An OS was developed to assign COVID-19 disease severity using the Observational Medical Outcomes Partnership common data model within the National COVID Cohort Collaborative (N3C) data enclave. We then evaluated usefulness of the developed OS using heterogenous EHR data from January 2020 to October 2021 submitted to N3C by 63 healthcare organizations across the United States. Principal component analysis (PCA) was employed to characterize changes in disease severity among patients during the 28-day period following COVID-19 diagnosis.

Results

The data set used in this analysis consists of 2 880 456 patients. PCA of the day-to-day variation in OS levels over the totality of the 28-day period revealed contrasting patterns of variation in disease severity within the first and second 14 days and illustrated the importance of evaluation over the full 28-day period.

Discussion

An OS with well-defined, robust features, based on discrete EHR data elements, is useful for assessments of COVID-19 patient outcomes, providing insights on the progression of COVID-19 disease severity over time.

Conclusions

The OS provides a framework that can facilitate better understanding of the course of acute COVID-19, informing clinical decision-making and resource allocation.

447. An Ordinal Scale Assessing SARS-CoV-2 Infected Patient Outcomes Using Electronic Health Records

Background

A major challenge to identifying effective treatments for COVID-19 has been the conflicting results offered by small, often underpowered clinical trials. The World Health Organization (WHO) Ordinal Scale (OS) has been used to measure clinical improvement among clinical trial participants and has the benefit of measuring effect across the spectrum of clinical illness. We modified the WHO OS to enable assessment of COVID-19 patient outcomes using electronic health record (EHR) data.

Methods

Employing the National COVID Cohort Collaborative (N3C) database of EHR data from 50 sites in the United States, we assessed patient outcomes, April 1,2020 to March 31, 2021, among those with a SARS-CoV-2 diagnosis, using the following modification of the WHO OS: 1=Outpatient, 3=Hospitalized, 5=Required Oxygen (any), 7=Mechanical Ventilation, 9=Organ Support (pressors; ECMO), 11=Death. OS is defined over 4 weeks beginning at first diagnosis and recalculated each week using the patient’s maximum OS value in the corresponding 7-day period. Modified OS distributions were compared across time using a Pearson Chi-Squared test.

Results

The study sample included 1,446,831 patients, 54.7% women, 14.7% Black, 14.6% Hispanic/Latinx. Pearson Chi-Sq P< 0.0001 was obtained comparing the distribution of 2^nd Quarter 2020 OS with the distribution of later time points for Week 4.

Table 1. OS at week 1 and 4 by quarter

The study sample included 1,446,831 patients, 54.7% women, 14.7% Black, 14.6% Hispanic/Latinx. Pearson Chi-Sq P< 0.0001 was obtained comparing the distribution of 2nd Quarter 2020 OS with the distribution of later time points for Week 4.

Conclusion

All Week 4 OS distributions significantly improved from the initial period (April-June 2020) compared with subsequent months, suggesting improved management. Further work is needed to determine which elements of care are driving the improved outcomes. Time series analyses must be included when assessing impact of therapeutic modalities across the COVID pandemic time frame.

Disclosures

Sally L. Hodder, M.D., Gilead (Advisor or Review Panel member)Merck (Grant/Research Support, Advisor or Review Panel member)Viiv Healthcare (Grant/Research Support, Advisor or Review Panel member)

Multi-objective design optimization of dental implant geometrical parameters

In-silico investigations are becoming an integral part of the development of novel biomedical devices, including dental implants. Using computer simulations can streamline the process by tuning different geometrical and structural features, emphasizing the osseointegration of the implant design a priori, leading to the optimal designs in preparation for in-vivo trails. This research aims to elucidate the interrelationship between 12 geometrical variables that holistically define the shape of the implant. The approach to achieve optimality hinged on coupling the finite element analysis results with the fractional factorial design method. The latter was used to determine the most influential variables during the screening process, followed by the parameter optimization process using the response surface method, regarding four different objectives, namely: bone-implant contact area, volume of trabecular bone dead cells, volume of cortical bone dead cells, and axial displacement. This resulted in reducing the number of virtual experiments and substantially decreasing the computational cost without compromising the accuracy of the solution. It was found that the optimized values improved the performance significantly. The validity of all models was verified by comparing optimized responses with simulation results. A sensitivity analysis was performed on all five optimized models to address the effect of friction coefficient on the implant-bone joint interaction. It was shown that the mechanical behavior of implant-bone would be independent in higher friction coefficients. The significance of this study is demonstrated in determining the most effective and optimized values of all possible geometrical parameters considering their singular or interactive effects.

Quantitative risk assessment of accidental points in the area of Shahid Beheshti field to Abuzar Square in Yazd with the method of accidents rate - severity index

Introduction: Accidents are one of the biggest public health problems in the world. At the same time, most young victims are in good health before the accident. World Health Organization studies indicate that in 2020 road accidents will be one of three of the three leading causes of death globally. Materials and Methods: Due to selecting the inner-city route, this study identified the accident hot zones by the number-accident intensity index method in one year. The table of coefficients proposed by the state of Georgia was used to weigh the severity of the accidents (injuries, deaths, and damages). Results: In this study, six spatial units were identified as hot zones ( 9/1% ), and 19 spatial units were identified as yellow zones ( 28/8% ); as. Finally, 41 spatial units ( 4100 meters from 6600 meters ) were identified as cold zones ( 62/1% ). Conclusion: The most important ways to reduce traffic accidents include the review of traffic signs in terms of the number, size, location, height, and the installation of the speed breaker in the spatial units as hot zones, the construction of an underpass, and the structure of pedestrian bridges with embedding the escalator in some important routs ( hot zones ), developing a comprehensive and long– term training program for improving the driving culture, etc.

Efficient optical trapping with cylindrical vector beams

Radially and azimuthally polarized beams can create needle-like electric and magnetic fields under tight focusing conditions, respectively, and thus have been highly recommended for optical manipulation. There have been reports on the superiority of these beams over the conventional Gaussian beam for providing a larger optical force in single beam optical trap. However, serious discrepancies in their experimental results prevent one from concluding this superiority. Here, we theoretically and experimentally study the impact of different parameters - such as spherical aberration, the numerical aperture of the focusing lens, and the particles' size - on optical trapping stiffness of radially, azimuthally, and linearly polarized beams. The result of calculations based on generalized Lorenz-Mie theory, which is in good agreement with the experiment, reveals that the studied parameters determine which polarization state has the superiority for optical trapping. Our findings play a crucial role in the development of optical tweezers setups and, in particular, in biophysical applications when laser-induced heating in the optical tweezers applications is the main concern.

Deconvolution based photoacoustic reconstruction with sparsity regularization

In most photoacoustic tomography (PAT) reconstruction approaches, it is assumed that the receiving transducers have omnidirectional response and can fully surround the region of interest. These assumptions are not satisfied in practice. To deal with these limitations, we present a novel deconvolution based photoacoustic reconstruction with sparsity regularization (DPARS) technique. The DPARS algorithm is a semi-analytical reconstruction approach in which the projections of the absorber distribution derived from a deconvolution-based method are computed and used to generate a large linear system of equations. In these projections, computed over limited viewing angles, the directivity effect of the transducer is taken into account. The distribution of absorbers is computed using a sparse representation of absorber coefficients obtained from the discrete cosine transform. This sparse representation helps improve the numerical conditioning of the system of equations and reduces the computation time of the deconvolution-based approach by one order of magnitude relative to Tikhonov regularization. The algorithm has been tested in simulations, and using two-dimensional and three-dimensional experimental data obtained with a conventional ultrasound transducer. The results show that DPARS, when evaluated using contrast-to-noise ratio and root-mean-square errors, outperforms the conventional delay-and-sum (DAS) reconstruction method.

Does calcium score in great pelvic vessels predict colorectal anastomotic leakage? A prospective study of one hundred anastomoses

BACKGROUND

Anastomotic leakage is one of the most severe surgical complications following surgery. This prospective study was designed to investigate an association between the calcification in the descending aorta and its major branches using a calcium-scoring software and colorectal anastomotic leakage.

METHODS

From January 2012 to March 2013, one hundred patients underwent surgeries involving colorectal anastomosis procedures. Calcium score in descending aorta and great pelvic vessels was measured using the Syngo-CT 2006G-W software. A questionnaire was completed containing demographic and underlying risk factors suspicious to be associated with anastomotic leakage, in addition to surgical characteristics data.

RESULTS

55 males and 45 females entered the study with the mean age of 63.70±7.17 years. The average duration time of surgery was 149.30±20.24 minutes. The type of surgery was elective for 90 patients and emergency for 10 others. The mean calcium score was higher in greater arteries as in abdominal aorta and common iliac arteries in comparison to the other pelvic vessels. Comparing two groups of patients with and without anastomotic leakage, the calcium score was higher in descending aorta and all great pelvic vessels of patients with colorectal anastomotic leakage (P<0.001). One patient (1%) died due to postoperative anastomotic leakage two weeks after the surgery.

CONCLUSIONS

Atherosclerotic calcification in the descending aorta and its major branches can be considered as a risk factor in the development of colorectal anastomotic leakage. (Acta gastroenterol. belg., 2016, 79, 415-420).

CoolClot, a novel hemostatic agent for controlling life-threatening arterial bleeding

BACKGROUND:

Uncontrolled bleeding is the first leading cause of preventable death in the battlefield and the 2nd cause of mortality in civil accidents. Incompressible hemorrhage control is among the interventions that drastically increase the survival rate in wound individuals. We have previously shown that a certain mixture of bentonite and zeolite minerals can significantly decrease the bleeding in rats.

METHODS:

In this study, nine healthy hybrid dogs were selected and after induction of anesthesia with ether, either arterial puncture by a needle or arteriotomy was performed on both groin regions of the dogs. For control arteries (either the right or left femoral artery), only pressure by sterilized gauze was performed, while for the femoral arteries of the opposite side, our invented hemostatic agent, namely CoolClot, was topically used before applying the pressure. In the second stage of the study, to assess the coagulation time, blood samples were collected from 10 volunteer students.

RESULTS:

CoolClot significantly decreased the bleeding time in animals whose femoral arteries were cut or punctured. In the human phase of the study, the mean coagulation time in control blood samples was 253.4±44.1 seconds, whereas it was 149.5±50.0, 162.3±74.6 and 143.4±114.6 seconds, respectively in blood samples treated with bentonite, zeolite and CoolClot (P<0.05).

CONCLUSIONS:

As controlling bleeding after a life-threatening arterial damage is critical for increasing the chance of survival, the results obtained in this study indicate the significant efficacy of CoolClot in shortening the bleeding time. Our experiments also indicate that CoolClot can significantly reduce the clotting time in human blood samples.

Corrosion behavior of pulse electrodeposited nanostructure Ni-SiC composite coatings

Ni-SiC nano-composite coatings with various contents of SiC were prepared by pulse electrodeposition from a modified Watts bath containing SiC nano-particles. The effect of SiC concentration, current density, duty cycle and pulse frequency on the corrosion behavior of the coatings was investigated by means of potentiodynamic polarization and Electrochemical Impedance Spectroscopy (EIS) methods. It has been found that the Ni-SiC composite coatings show better corrosion resistance in a 3.5% NaCI solution than pure nickel. The corrosion resistance of the coatings increases by increasing the amount of embedded SiC particles. This improvement can be attributed to the morphology and the crystallographic texture of the coatings.

DXRaySMCS: a user-friendly interface developed for prediction of diagnostic radiology X-ray spectra produced by Monte Carlo (MCNP-4C) simulation

In this work, the general purpose Monte Carlo N-particle radiation transport computer code (MCNP-4C) was used for the simulation of X-ray spectra in diagnostic radiology. The electron's path in the target was followed until its energy was reduced to 10 keV. A user-friendly interface named 'diagnostic X-ray spectra by Monte Carlo simulation (DXRaySMCS)' was developed to facilitate the application of MCNP-4C code for diagnostic radiology spectrum prediction. The program provides a user-friendly interface for: (i) modifying the MCNP input file, (ii) launching the MCNP program to simulate electron and photon transport and (iii) processing the MCNP output file to yield a summary of the results (relative photon number per energy bin). In this article, the development and characteristics of DXRaySMCS are outlined. As part of the validation process, output spectra for 46 diagnostic radiology system settings produced by DXRaySMCS were compared with the corresponding IPEM78. Generally, there is a good agreement between the two sets of spectra. No statistically significant differences have been observed between IPEM78 reported spectra and the simulated spectra generated in this study.

Progression of metastatic human prostate cancer to androgen independence in immunodeficient SCID mice

Prostate cancer mortality results from metastasis to bone and hormone-independent tumor growth. Models to study these progressive changes are lacking. Here we describe the propagation of advanced human prostate cancer by direct transfer of surgical samples from patients into immune-deficient male SCID mice. Explants from six of eight patients formed prostate tumors and two showed unique cytogenetic, biologic and molecular features that were retained through six or more passages. One grew in an androgen-independent fashion, whereas the second formed tumors that regressed following castration then regrew. Micrometastatic disease was detected in the hematopoietic tissues of half of the recipient mice. Thus selected specimens of advanced human prostate cancer can be propagated in SCID mice in a manner that recapitulates the clinical transition from androgen-sensitive to androgen-independent growth, accompanied by micrometastasis.