scholarly journals Designing COVID-19 mortality predictions to advance clinical outcomes: Evidence from the Department of Veterans Affairs

2021 ◽  
Vol 28 (1) ◽  
pp. e100312
Author(s):  
Christos A Makridis ◽  
Tim Strebel ◽  
Vincent Marconi ◽  
Gil Alterovitz
Keyword(s):  
Artificial Intelligence ◽  
Risk Factors ◽  
Medical History ◽  
Clinical Outcomes ◽  
Mortality Rates ◽  
Veterans Affairs ◽  
Department Of Veterans Affairs ◽  
The Usa ◽  
Precision Recall Curve ◽  
Recall Curve

Using administrative data on all Veterans who enter Department of Veterans Affairs (VA) medical centres throughout the USA, this paper uses artificial intelligence (AI) to predict mortality rates for patients with COVID-19 between March and August 2020. First, using comprehensive data on over 10 000 Veterans’ medical history, demographics and lab results, we estimate five AI models. Our XGBoost model performs the best, producing an area under the receive operator characteristics curve (AUROC) and area under the precision-recall curve of 0.87 and 0.41, respectively. We show how focusing on the performance of the AUROC alone can lead to unreliable models. Second, through a unique collaboration with the Washington D.C. VA medical centre, we develop a dashboard that incorporates these risk factors and the contributing sources of risk, which we deploy across local VA medical centres throughout the country. Our results provide a concrete example of how AI recommendations can be made explainable and practical for clinicians and their interactions with patients.

scholarly journals Colorectal cancer in the world: incidence, mortality and risk factors

2017 ◽  
Vol 4 (10) ◽  
pp. 1656 ◽  
Author(s):  
Hamidreza Sadeghi Gandomani ◽  
Seyed Majid Yousefi ◽  
Mohammad Aghajani ◽  
Abdollah Mohammadian-Hafshejani ◽  
Abed Asgari Tarazoj ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Risk Factors ◽  
Medical History ◽  
Peer Review ◽  
Search Strategy ◽  
Mortality Rates ◽  
Incidence Rates ◽  
Screening Programs ◽  
The World ◽  
Peer Reviewers

A rapid literature search strategy was conducted for all English language literature published before July 2017. The search was conducted using the electronic databases PubMed, Scopus and Web of Science. The search strategy included the keywords ‘colorectal cancer’, ‘epidemiology’, ‘incidence’, ‘mortality’, ‘risk factor’, and ‘world’. In 2012, the highest CRC incidence rates were observed in the Republic of Korea, Slovakia and Hungary while the lowest incidence rates were seen in Singapore, Serbia and Japan. The highest CRC mortality rates in both sexes were seen in Central and Eastern Europe and the lowest mortality rates were found in Middle Division of Africa. The main risk factors for CRC include nutritional factors, past medical history, smoking, socioeconomic status, and family medical history. According to the increasing trend of CRC incidence and mortality in the world, implementation of prevention programs such as screening programs, diet modification, and healthy lifestyle education is necessary. Peer Review Details Peer review method: Single-Blind (Peer-reviewers: 02) Peer-review policy Plagiarism software screening?: Yes Date of Original Submission: 26 August 2017 Date accepted: 20 Sept 2017 Peer reviewers approved by: Dr. Lili Hami Editor who approved publication: Dr. Phuc Van Pham  

scholarly journals 1882. Tuberculosis in the Department of Veterans Affairs: Missed Opportunities for Prevention

2019 ◽  
Vol 6 (Supplement_2) ◽  
pp. S50-S51
Author(s):  
Gina Oda ◽  
Cynthia Lucero-Obusan ◽  
Patricia Schirmer ◽  
Mark Holodniy
Keyword(s):  
Risk Factors ◽  
Risk Factor ◽  
Preventive Treatment ◽  
Veterans Affairs ◽  
Case Definition ◽  
Nucleic Acid Amplification ◽  
Department Of Veterans Affairs ◽  
Directly Observed Therapy ◽  
Local Public Health ◽  
Ltbi Testing

Abstract Background US and global elimination of tuberculosis (TB) is an important goal. Despite decreased incidence, CDC predicts elimination of TB in the US will not occur in the 21st century without improved detection and treatment of latent TB infection (LTBI). We describe the current burden of active TB infection and LTBI testing and treatment among patients within the Department of Veterans Affairs (VA). Methods Using the 2009 CDC case definition for laboratory-confirmed TB, we queried VA data sources from January 2010 to December 2018 for Mycobacterium tuberculosis detected via culture or nucleic acid amplification test (NAAT) from specimens from all body sites. For all TB patients, we extracted demographic, ICD-9 and ICD-10 risk factor, and LTBI testing and treatment data. Results Between 2010 and 2018, the average annual incidence of TB was 1.7 cases per 100,000 unique users of VA care (ranging from a high of 2.8 in 2010 to low of 0.8 in 2018). For 899 identified cases, demographic factors associated with highest TB rates were age between 45 and 64, Asian race, and residence in District of Columbia (Table 1). The most frequently occurring risk factors were substance abuse, diabetes, and homelessness. Of 90 patients with susceptibility documentation, 14 (15%) had resistance to 1 or more anti-TB drug (1 with multi-drug-resistant TB). Fifteen patients (1.7%) died within 7 days of their TB diagnosis; in all but 2 cases, TB was the primary cause of death (Table 2). Figure 1 depicts screening and treatment for LTBI among patients with TB. Only 228/899 (25.4%) TB patients had LTBI screening ≥ 3 months prior to diagnosis. Of the 347 TB patients never screened for LTBI, 264 (76%) had ≥ 1 documented TB risk factor. Among 228 patients screened for LTBI >3 months prior to active disease, 69 (30%) screened positive; however, only 24 (35%) had LTBI treatment initiated. Conclusion Although rates of TB infection are decreasing, VHA providers would benefit from education on recognizing patients with risk factors which place them at high risk for TB who should be screened for LTBI. CDC recommends preventive treatment of patients who screen positive for LTBI, and provider collaboration with local public health departments to provide directly observed therapy in cases where adherence may be in question. Disclosures All Authors: No reported Disclosures.

scholarly journals Explainable artificial intelligence (XAI) for exploring spatial variability of lung and bronchus cancer (LBC) mortality rates in the contiguous USA

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Zia U. Ahmed ◽  
Kang Sun ◽  
Michael Shelly ◽  
Lina Mu
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Risk Factors ◽  
Mortality Rates ◽  
Gradient Boosting ◽  
Ensemble Machine Learning ◽  
Gradient Boosting Machine ◽  
Explainable Artificial Intelligence ◽  
The Impact ◽  
Bronchus Cancer

AbstractMachine learning (ML) has demonstrated promise in predicting mortality; however, understanding spatial variation in risk factor contributions to mortality rate requires explainability. We applied explainable artificial intelligence (XAI) on a stack-ensemble machine learning model framework to explore and visualize the spatial distribution of the contributions of known risk factors to lung and bronchus cancer (LBC) mortality rates in the conterminous United States. We used five base-learners—generalized linear model (GLM), random forest (RF), Gradient boosting machine (GBM), extreme Gradient boosting machine (XGBoost), and Deep Neural Network (DNN) for developing stack-ensemble models. Then we applied several model-agnostic approaches to interpret and visualize the stack ensemble model's output in global and local scales (at the county level). The stack ensemble generally performs better than all the base learners and three spatial regression models. A permutation-based feature importance technique ranked smoking prevalence as the most important predictor, followed by poverty and elevation. However, the impact of these risk factors on LBC mortality rates varies spatially. This is the first study to use ensemble machine learning with explainable algorithms to explore and visualize the spatial heterogeneity of the relationships between LBC mortality and risk factors in the contiguous USA.

scholarly journals Ethical Applications of Artificial Intelligence: Evidence from Veterans’ Health Research (Preprint)

2021 ◽  
Author(s):  
Christos Makridis ◽  
Seth Hurley ◽  
Mary Klote ◽  
Gil Alterovitz
Keyword(s):  
Artificial Intelligence ◽  
Veterans Affairs ◽  
Vulnerable Groups ◽  
Veterans Health ◽  
Department Of Veterans Affairs ◽  
Subject Matter Experts ◽  
Technology Governance ◽  
Widespread Agreement ◽  
Potential Benefits

UNSTRUCTURED There is widespread agreement that, while artificial intelligence offers significant potential benefits for individuals and society at large, there are also serious challenges to overcome with respect to its governance. Recent policymaking has focused on establishing principles for the trustworthy use of AI. Adhering to these principles is especially important to adhere to protect vulnerable groups and ensure their confidence in the technology and its uses. Using the Department of Veterans Affairs as a case study, we focus on three principles of particular interest: (i) designing, developing, acquiring, and using AI where the benefits of use significantly outweigh the risks and the risks are assessed and managed, (ii) ensuring that the application of AI occurs in well-defined domains and are accurate, effective, and fit for intended purposes, and (iii) ensure the operations and outcomes of AI applications are sufficiently interpretable and understandable by all subject matter experts, users, and others. We argue that these principles and applications apply to vulnerable groups more generally and that adherence to them can allow the VA and other organizations to continue modernizing its technology governance, leveraging the gains of AI and managing its risks.

Sign in / Sign up

Export Citation Format

Share Document