Systems Engineering for ASPIRE: A Low-Cost, High Risk Parachute Test Project

AimsTo validate a multivariable risk prediction model (Cohorts for Heart and Aging Research in Genomic Epidemiology model for atrial fibrillation (CHARGE-AF)) for 5-year risk of atrial fibrillation (AF) in routinely collected primary care data and to assess CHARGE-AF’s potential for automated, low-cost selection of patients at high risk for AF based on routine primary care data.MethodsWe included patients aged ≥40 years, free of AF and with complete CHARGE-AF variables at baseline, 1 January 2014, in a representative, nationwide routine primary care database in the Netherlands (Nivel-PCD). We validated CHARGE-AF for 5-year observed AF incidence using the C-statistic for discrimination, and calibration plot and stratified Kaplan-Meier plot for calibration. We compared CHARGE-AF with other predictors and assessed implications of using different CHARGE-AF cut-offs to select high-risk patients.ResultsAmong 111 475 patients free of AF and with complete CHARGE-AF variables at baseline (17.2% of all patients aged ≥40 years and free of AF), mean age was 65.5 years, and 53% were female. Complete CHARGE-AF cases were older and had higher AF incidence and cardiovascular comorbidity rate than incomplete cases. There were 5264 (4.7%) new AF cases during 5-year follow-up among complete cases. CHARGE-AF’s C-statistic for new AF was 0.74 (95% CI 0.73 to 0.74). The calibration plot showed slight risk underestimation in low-risk deciles and overestimation of absolute AF risk in those with highest predicted risk. The Kaplan-Meier plot with categories <2.5%, 2.5%–5% and >5% predicted 5-year risk was highly accurate. CHARGE-AF outperformed CHA2DS2-VASc (Cardiac failure or dysfunction, Hypertension, Age >=75 [Doubled], Diabetes, Stroke [Doubled]-Vascular disease, Age 65-74, and Sex category [Female]) and age alone as predictors for AF. Dichotomisation at cut-offs of 2.5%, 5% and 10% baseline CHARGE-AF risk all showed merits for patient selection in AF screening efforts.ConclusionIn patients with complete baseline CHARGE-AF data through routine Dutch primary care, CHARGE-AF accurately assessed AF risk among older primary care patients, outperformed both CHA2DS2-VASc and age alone as predictors for AF and showed potential for automated, low-cost patient selection in AF screening.

Download Full-text

Can wearable technology be used to approximate cardiopulmonary exercise testing metrics?

Perioperative Medicine ◽

10.1186/s13741-021-00180-w ◽

2021 ◽

Vol 10 (1) ◽

Author(s):

Laura Jones ◽

Laura Tan ◽

Suzanne Carey-Jones ◽

Nathan Riddell ◽

Richard Davies ◽

...

Keyword(s):

Physical Activity ◽

High Risk ◽

Oxygen Uptake ◽

Anaerobic Threshold ◽

Exercise Testing ◽

Low Cost ◽

Cardiopulmonary Exercise Testing ◽

Peak Oxygen Uptake ◽

Wearable Device ◽

Cardiopulmonary Exercise

Abstract Background Consumer wrist-worn wearable activity monitors are widely available, low cost and are able to provide a direct measurement of several markers of physical activity. Despite this, there is limited data on their use in perioperative risk prediction. We explored whether these wearables could accurately approximate metrics (anaerobic threshold, peak oxygen uptake and peak work) derived using formalised cardiopulmonary exercise testing (CPET) in patients undergoing high-risk surgery. Methods Patients scheduled for major elective intra-abdominal surgery and undergoing CPET were included. Physical activity levels were estimated through direct measures (step count, floors climbed and total distance travelled) obtained through continuous wear of a wrist worn activity monitor (Garmin Vivosmart HR+) for 7 days prior to surgery and self-report through completion of the short International Physical Activity Questionnaire (IPAQ). Correlations and receiver operating characteristic (ROC) curve analysis explored the relationships between parameters provided by CPET and physical activity. Device selection Our choice of consumer wearable device was made to maximise feasibility outcomes for this study. The Garmin Vivosmart HR+ had the longest battery life and best waterproof characteristics of the available low-cost devices. Results Of 55 patients invited to participate, 49 (mean age 65.3 ± 13.6 years; 32 males) were enrolled; 37 provided complete wearable data for analyses and 36 patients provided full IPAQ data. Floors climbed, total steps and total travelled as measured by the wearable device all showed moderate correlation with CPET parameters of peak oxygen uptake (peak VO2) (R = 0.57 (CI 0.29–0.76), R = 0.59 (CI 0.31–0.77) and R = 0.62 (CI 0.35–0.79) respectively), anaerobic threshold (R = 0.37 (CI 0.01–0.64), R = 0.39 (CI 0.04–0.66) and R = 0.42 (CI 0.07–0.68) respectively) and peak work (R = 0.56 (CI 0.27–0.75), R = 0.48 (CI 0.17–0.70) and R = 0.50 (CI 0.2–0.72) respectively). Receiver operator curve (ROC) analysis for direct and self-reported measures of 7-day physical activity could accurately approximate the ventilatory equivalent for carbon dioxide (VE/VCO2) and the anaerobic threshold. The area under these curves was 0.89 for VE/VCO2 and 0.91 for the anaerobic threshold. For peak VO2 and peak work, models fitted using just the wearable data were 0.93 for peak VO2 and 1.00 for peak work. Conclusions Data recorded by the wearable device was able to consistently approximate CPET results, both with and without the addition of patient reported activity measures via IPAQ scores. This highlights the potential utility of wearable devices in formal assessment of physical functioning and suggests they could play a larger role in pre-operative risk assessment. Ethics This study entitled “uSing wearable TEchnology to Predict perioperative high-riSk patient outcomes (STEPS)” gained favourable ethical opinion on 24 January 2017 from the Welsh Research Ethics Committee 3 reference number 17/WA/0006. It was registered on ClinicalTrials.gov with identifier NCT03328039.

Download Full-text

Implementing clinical decision support for oncology advanced care planning: A systems engineering framework to optimize the usability and utility of a machine learning predictive model in clinical practice.

Journal of Clinical Oncology ◽

10.1200/jco.2020.39.28_suppl.330 ◽

2021 ◽

Vol 39 (28_suppl) ◽

pp. 330-330

Author(s):

Teja Ganta ◽

Stephanie Lehrman ◽

Rachel Pappalardo ◽

Madalene Crow ◽

Meagan Will ◽

...

Keyword(s):

Machine Learning ◽

High Risk ◽

Predictive Model ◽

Systems Engineering ◽

Care Planning ◽

Learning Models ◽

Predictive Tool ◽

Risk Of Death ◽

The Impact ◽

Machine Learning Models

330 Background: Machine learning models are well-positioned to transform cancer care delivery by providing oncologists with more accurate or accessible information to augment clinical decisions. Many machine learning projects, however, focus on model accuracy without considering the impact of using the model in real-world settings and rarely carry forward to clinical implementation. We present a human-centered systems engineering approach to address clinical problems with workflow interventions utilizing machine learning algorithms. Methods: We aimed to develop a mortality predictive tool, using a Random Forest algorithm, to identify oncology patients at high risk of death within 30 days to move advance care planning (ACP) discussions earlier in the illness trajectory. First, a project sponsor defined the clinical need and requirements of an intervention. The data scientists developed the predictive algorithm using data available in the electronic health record (EHR). A multidisciplinary workgroup was assembled including oncology physicians, advanced practice providers, nurses, social workers, chaplain, clinical informaticists, and data scientists. Meeting bi-monthly, the group utilized human-centered design (HCD) methods to understand clinical workflows and identify points of intervention. The workgroup completed a workflow redesign workshop, a 90-minute facilitated group discussion, to integrate the model in a future state workflow. An EHR (Epic) analyst built the user interface to support the intervention per the group’s requirements. The workflow was piloted in thoracic oncology and bone marrow transplant with plans to scale to other cancer clinics. Results: Our predictive model performance on test data was acceptable (sensitivity 75%, specificity 75%, F-1 score 0.71, AUC 0.82). The workgroup identified a “quality of life coordinator” who: reviews an EHR report of patients scheduled in the upcoming 7 days who have a high risk of 30-day mortality; works with the oncology team to determine ACP clinical appropriateness; documents the need for ACP; identifies potential referrals to supportive oncology, social work, or chaplain; and coordinates the oncology appointment. The oncologist receives a reminder on the day of the patient’s scheduled visit. Conclusions: This workgroup is a viable approach that can be replicated at institutions to address clinical needs and realize the full potential of machine learning models in healthcare. The next steps for this project are to address end-user feedback from the pilot, expand the intervention to other cancer disease groups, and track clinical metrics.

Download Full-text

Development of Low Cost Vehicle Concept for United States Using Systems Engineering Principles

10.4271/2011-01-1019 ◽

2011 ◽

Cited By ~ 2

Author(s):

Hussain Tajmahal ◽

Shantanu Ranadive

Keyword(s):

United States ◽

Systems Engineering ◽

Low Cost ◽

Vehicle Concept

Download Full-text

Cries in the Dark: Reconstruction after Hurricane Mitch in Honduras

Open House International ◽

10.1108/ohi-01-2006-b0004 ◽

2006 ◽

Vol 31 (1) ◽

pp. 31-38

Author(s):

Kurt Rhyner

Keyword(s):

High Risk ◽

Low Cost ◽

Construction Materials ◽

Natural Phenomenon ◽

Police Chief ◽

Hurricane Mitch ◽

The Public ◽

Risk Areas ◽

Mountainous River

Disasters are always caused by a combination of factors, and the natural phenomenon that brings them on is usually just a catalyst. The underlying cause of most disasters is poverty as mostly the poor segments of the population usually live in high risk areas where their shelter all too often cannot withstand even light winds, small inundations or medium earthquakes. When Hurricane Mitch hit Central America in October 1998, all countries were ill prepared. A few weeks earlier, the authorities of the Honduran capital, Tegucigalpa, had attempted to simulate an evacuation, but it had met with a great degree of resistance from the public. When Mitch hit, unprecedented masses of water raced down the mountainous river beds. People were taken by surprise, as no efficient organisation existed. Everybody ran for their lives. Houses slid down hillsides, rivers swept bridges, houses and people with them. Six years later, Tegucigalpa looks very similar to the days before Mitch. The steep hillsides are covered with a potpourri of dwellings, from miserable huts to solid upmarket houses. Regulations were passed in the year 2002 to prohibit construction in high risk areas; however, enforcement is difficult, especially when existing buildings are renovated and even enlarged. Theoretically it is possible to evacuate high risk areas. Nonetheless, such drastic measures are virtually impossible to implement, as no mayor or police chief would survive such an action in office. The paper presents a case study which shows that the underlying problems of poverty and the non-availability of suitable land for people to relocate from high risk areas can usually not be overcome by post-disaster reconstruction programmes. A mitigation strategy is thus to empower inhabitants of high risk areas to improve their own situation by affordable access to information, advice and suitable low cost construction materials through “Building Advisory Services” and Ecomaterials producers within the neighbourhoods.

Download Full-text

Clinical Implementation of a Parent Questionnaire to Identify Seizures in High-Risk Children

Journal of Child Neurology ◽

10.1177/0883073820911505 ◽

2020 ◽

Vol 35 (7) ◽

pp. 485-491

Author(s):

Celia Greenlaw ◽

Sarah Nuss ◽

Cristina Camayd-Muñoz ◽

Rinat Jonas ◽

Julie Vanier Rollins ◽

...

Keyword(s):

High Risk ◽

Low Cost ◽

Numerical Data ◽

Primary Care Providers ◽

Autism Spectrum ◽

Categorical Variables ◽

Care Providers ◽

Clinical Implementation ◽

Parent Questionnaire ◽

High Risk Children

Background: This study evaluated the effectiveness of a parent-completed questionnaire for detecting seizures in high-risk children. Methods: A 2-part seizure screen for children up to 12 years of age with suspected autism spectrum disorder, developmental delay, or seizure, was implemented in 12 Massachusetts clinics serving populations with high health disparities. Primary care providers and developmental behavioral pediatricians administered part 1, a brief highly sensitive screen. If the result was positive, a research assistant administered part 2, a more detailed screen with higher specificity. Positive part 2 results prompted a specialized assessment by a pediatric neurologist. Screening data were evaluated for detection of seizures or other diagnoses, reason for conducting the screen, and appointment outcomes. Data analysis included chi-squared tests, percentages for categorical variables, and means for numerical data. Results: Of 207 administered seizure questionnaires, 78% of children screened positive on part 1. Of those, 94% of families completed part 2 by telephone, and 64 individuals screened positive. The screen helped to detect 15 new seizure diagnoses and 35 other neurologic diagnoses. Average time to first scheduled appointment was 23.8 days. The no-show rate was 7%. Conclusions: The seizure questionnaire effectively identified seizures and other disorders in a diverse population of high-risk children. Broader use of this low-cost screening tool could improve access to care for children with suspected seizures, increase seizure recognition, and help allocate resources more effectively.

Download Full-text