scholarly journals The Need for Ethnoracial Equity in Artificial Intelligence for Diabetes Management: Review and Recommendations

10.2196/22320 ◽  
2021 ◽  
Vol 23 (2) ◽  
pp. e22320
Author(s):  
Quynh Pham ◽  
Anissa Gamble ◽  
Jason Hearn ◽  
Joseph A Cafazzo
Keyword(s):  
Artificial Intelligence ◽  
Native American ◽  
Racial Differences ◽  
Diabetes Management ◽  
Secondary Analysis ◽  
Training Data ◽  
Foreign Born ◽  
Increased Risk ◽  
Average Distribution ◽  
Decisive Action

There is clear evidence to suggest that diabetes does not affect all populations equally. Among adults living with diabetes, those from ethnoracial minority communities—foreign-born, immigrant, refugee, and culturally marginalized—are at increased risk of poor health outcomes. Artificial intelligence (AI) is actively being researched as a means of improving diabetes management and care; however, several factors may predispose AI to ethnoracial bias. To better understand whether diabetes AI interventions are being designed in an ethnoracially equitable manner, we conducted a secondary analysis of 141 articles included in a 2018 review by Contreras and Vehi entitled “Artificial Intelligence for Diabetes Management and Decision Support: Literature Review.” Two members of our research team independently reviewed each article and selected those reporting ethnoracial data for further analysis. Only 10 articles (7.1%) were ultimately selected for secondary analysis in our case study. Of the 131 excluded articles, 118 (90.1%) failed to mention participants’ ethnic or racial backgrounds. The included articles reported ethnoracial data under various categories, including race (n=6), ethnicity (n=2), race/ethnicity (n=3), and percentage of Caucasian participants (n=1). Among articles specifically reporting race, the average distribution was 69.5% White, 17.1% Black, and 3.7% Asian. Only 2 articles reported inclusion of Native American participants. Given the clear ethnic and racial differences in diabetes biomarkers, prevalence, and outcomes, the inclusion of ethnoracial training data is likely to improve the accuracy of predictive models. Such considerations are imperative in AI-based tools, which are predisposed to negative biases due to their black-box nature and proneness to distributional shift. Based on our findings, we propose a short questionnaire to assess ethnoracial equity in research describing AI-based diabetes interventions. At this unprecedented time in history, AI can either mitigate or exacerbate disparities in health care. Future accounts of the infancy of diabetes AI must reflect our early and decisive action to confront ethnoracial inequities before they are coded into our systems and perpetuate the very biases we aim to eliminate. If we take deliberate and meaningful steps now toward training our algorithms to be ethnoracially inclusive, we can architect innovations in diabetes care that are bound by the diverse fabric of our society.

scholarly journals The need for ethnoracial equity in artificial intelligence for diabetes management: review and recommendations (Preprint)

2020 ◽  
Author(s):  
Quynh Pham ◽  
Anissa Gamble ◽  
Jason Hearn ◽  
Joseph A Cafazzo
Keyword(s):  
Artificial Intelligence ◽  
Native American ◽  
Racial Differences ◽  
Diabetes Management ◽  
Secondary Analysis ◽  
Training Data ◽  
Foreign Born ◽  
Increased Risk ◽  
Average Distribution ◽  
Decisive Action

UNSTRUCTURED There is clear evidence to suggest that diabetes does not affect all populations equally. Among adults living with diabetes, those from ethnoracial minority communities—foreign-born, immigrant, refugee, and culturally marginalized—are at increased risk of poor health outcomes.Artificial intelligence (AI) is actively being researched as a means of improving diabetes management and care; however, several factors may predispose AI to ethnoracial bias. To better understand whether diabetes AI interventions are being designed in an ethnoracially equitable manner, we conducted a secondary analysis of 141 articles included in a 2018 review by Contreras and Vehi entitled, “Artificial Intelligence for Diabetes Management and Decision Support: Literature Review”. Two members of our research team independently reviewed each article and selected those reporting ethnoracial data for further analysis. Only 10 articles (7.1%) were ultimately selected for secondary analysis in our case study. Of the 131 excluded articles, 118 (90.1%) failed to mention participants’ ethnic and/or racial backgrounds. The included articles reported ethnoracial data under various categories, including race (N = 6), ethnicity (N = 2), race/ethnicity (N = 3), and percentage of Caucasian participants (N = 1). Amongst articles specifically reporting race, the average distribution was 69.5% White, 17.1% Black, and 3.7% Asian. Only two articles reported inclusion of Native American participants. Given the clear ethnic and racial differences in diabetes biomarkers, prevalence and outcomes, the inclusion of ethnoracial training data is likely to improve the accuracy of predictive models. Such considerations are imperative in AI-based tools, which are predisposed to negative biases due to their black-box nature and proneness to distributional shift. Based on our findings, we propose a short questionnaire to assess ethnoracial equity in research describing AI-based diabetes interventions. At this unprecedented time in history, AI can either mitigate or exacerbate disparities in healthcare. Future accounts of the infancy of diabetes AI must reflect our early and decisive action to confront ethnoracial inequities before they are coded into our systems and perpetuate the very biases we aim to eliminate. If we take deliberate and meaningful steps now towards training our algorithms to be ethnoracially inclusive, we can architect innovations in diabetes care that are bound by the diverse fabric of our society.

scholarly journals THE RELATIONSHIP BETWEEN INFLAMMATORY DIETARY PATTERN AND INCIDENCE OF PERIODONTITIS

2021 ◽  
pp. 1-36
Author(s):  
Ahmed A. Alhassani ◽  
Frank B. Hu ◽  
Bernard A. Rosner ◽  
Fred K. Tabung ◽  
Walter C. Willett ◽  
...  
Keyword(s):  
Confidence Interval ◽  
Periodontal Disease ◽  
Secondary Analysis ◽  
Hazard Ratio ◽  
Reduced Rank Regression ◽  
Rank Regression ◽  
Increased Risk ◽  
Cox Proportional Hazard Models ◽  
Study Population

ABSTRACT The long-term inflammatory impact of diet could potentially elevate the risk of periodontal disease through modification of systemic inflammation. The aim of the present study was to prospectively investigate the associations between a food based, reduced rank regression (RRR) derived, empirical dietary inflammatory pattern (EDIP) and incidence of periodontitis. The study population was composed of 34,940 men from the Health Professionals Follow-Up Study, who were free of periodontal disease and major illnesses at baseline (1986). Participants provided medical and dental history through mailed questionnaires every 2 years, and dietary data through validated semi-quantitative food frequency questionnaires every 4 years. We used Cox proportional hazard models to examine the associations between EDIP scores and validated self-reported incidence of periodontal disease over a 24-year follow-up period. No overall association between EDIP and the risk of periodontitis was observed; the hazard ratio comparing the highest EDIP quintile (most proinflammatory diet) to the lowest quintile was 0.99 (95% confidence interval: 0.89 -1.10, p-value for trend = 0.97). A secondary analysis showed that among obese non-smokers (i.e. never and former smokers at baseline), the hazard ratio for periodontitis comparing the highest EDIP quintile to the lowest was 1.39 (95% confidence interval: 0.98 -1.96, p-value for trend = 0.03). In conclusion, no overall association was detected between EDIP and incidence of self-reported periodontitis in the study population. From the subgroups evaluated EDIP was significantly associated with increased risk of periodontitis only among nonsmokers who were obese. Hence, this association must be interpreted with caution.

The Role of Disordered Eating in Type 2 Diabetes: A Pilot Study

2021 ◽  
pp. 155982762110024
Author(s):  
Alyssa M. Vela ◽  
Brooke Palmer ◽  
Virginia Gil-Rivas ◽  
Fary Cachelin
Keyword(s):  
Physical Activity ◽  
Type 2 Diabetes ◽  
Disordered Eating ◽  
Lifestyle Intervention ◽  
Eating Behaviors ◽  
Diabetes Management ◽  
Behavioral Therapy ◽  
Exercise Stress ◽  
Increased Risk

Rates of type 2 diabetes mellitus continue to rise around the world, largely due to lifestyle factors such as poor diet, overeating, and lack of physical activity. Diet and eating is often the most challenging aspect of management and, when disordered, has been associated with increased risk for diabetes-related complications. Thus, there is a clear need for accessible and evidence-based interventions that address the complex lifestyle behaviors that influence diabetes management. The current study sought to assess the efficacy and acceptability of a pilot lifestyle intervention for women with type 2 diabetes and disordered eating. The intervention followed a cognitive behavioral therapy guided-self-help (CBTgsh) model and included several pillars of lifestyle medicine, including: diet, exercise, stress, and relationships. Ten women completed the 12-week intervention that provided social support, encouraged physical activity, and addressed eating behaviors and cognitions. Results indicate the lifestyle intervention was a feasible treatment for disordered eating behaviors among women with type 2 diabetes and was also associated with improved diabetes-related quality of life. The intervention was also acceptable to participants who reported satisfaction with the program. The current CBTgsh lifestyle intervention is a promising treatment option to reduce disordered eating and improve diabetes management.

scholarly journals FRI0421 RATES OF PROGRESSION DIFFER BETWEEN STRUCTURAL PHENOTYPES OF KNEE OSTEOARTHRITIS: A SECONDARY ANALYSIS FROM THE FNIH COHORT

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 808.1-809
Author(s):  
F. Roemer ◽  
J. Collins ◽  
T. Neogi ◽  
M. Crema ◽  
A. Guermazi
Keyword(s):  
Subchondral Bone ◽  
Radiographic Progression ◽  
Conditional Logistic Regression ◽  
Secondary Analysis ◽  
Sensitivity Analyses ◽  
Case Group ◽  
Composite Outcome ◽  
Bone Phenotype ◽  
Increased Risk ◽  
Inflammatory Phenotype

Background:Imaging plays an important role in determining structural disease severity and potential suitability of patients recruited to disease-modifying osteoarthritis drug (DMOAD) trials. It has been suggested that there may be three main structural phenotypes in OA, i.e., inflammation, meniscus/cartilage and subchondral bone. These may progress differently and may represent distinct tissue targets for DMOAD approaches.Objectives:To stratify the Foundation for National Institutes of Health Osteoarthritis Biomarkers Consortium (FNIH) cohort, a well-defined subsample of the larger Osteoarthritis Initiative (OAI) study, into distinct structural phenotypes based on semiquantitative MRI assessment and to determine their risk for progression over 48 months.Methods:The FNIH was designed as a case-control study with knees showing either 1) radiographic and pain progression (i.e., “composite” cases), 2) radiographic progression only (“JSL”), 3) pain progression only, and 4) neither radiographic nor pain progression. MRI of both knees was performed on 3 T systems at the four OAI clinical sites. Two musculoskeletal radiologists read the baseline MRIs according to the MOAKS scoring system. Knees were stratified into subchondral bone, meniscus/cartilage and inflammatory phenotypes1. A secondary, less stringent definition for inflammatory and meniscus/cartilage phenotype was used for sensitivity analyses. The relation of each phenotype to risk of being in the JSL or composite case group compared to those not having that phenotype was determined using conditional logistic regression. Only KL2 and 3 and those without root tears were included.Results:485 knees were included. 362 (75%) did not have any phenotype, while 95 (20%) had the bone phenotype, 22 (5%) the cartilage/meniscus phenotype and 19 (4%) the inflammatory phenotype. The bone phenotype was associated with a higher risk of the JSL and composite outcome (OR 1.81;[95%CI 1.14,2.85] and 1.65; 95%CI [1.04,2.61]) while the inflammatory (OR 0.96 [95%CI 0.38,2.42] and 1.25; 95%CI [0.48,3.25]) and the meniscus/cartilage phenotypes were not (OR 1.30 95%CI [0.55,3.07] and 0.99; 95%CI [0.40,2,49]).In sensitivity analyses, the bone phenotype and having two phenotypes (vs. none) were both associated with increased risk of experiencing the composite outcome (bone: OR 1.65; 95% CI 1.04, 2.61; 2 phenotypes: OR 1.87; 95% CI 1.11, 3.16.Conclusion:The bone phenotype was associated with increased risk of having both radiographic and pain progression together, or radiographic progression alone, whereas the inflammatory phenotype or meniscus/cartilage phenotype each individually were not associated with either outcome. Phenotypic stratification appears to provide insights into risk for structural or composite structure plus pain progression, and therefore may be useful to consider when selecting patients for inclusion in clinical trials.References:[1]Roemer FW, Collins J, Kwoh CK, et al. MRI-based screening for structural definition of eligibility in clinical DMOAD trials: Rapid OsteoArthritis MRI Eligibility Score (ROAMES). Osteoarthritis Cartilage 2020;28(1):71-81Disclosure of Interests:Frank Roemer: None declared, Jamie Collins Consultant of: Boston Imaging Core Lab (BICL), LLC., Tuhina Neogi Grant/research support from: Pfizer/Lilly, Consultant of: Pfizer/Lilly, EMD-Merck Serono, Novartis, Michel Crema: None declared, Ali Guermazi Consultant of: AventisGalapagos, Pfizer, Roche, AstraZeneca, Merck Serono, and TissuGene

scholarly journals Artificial Intelligence, 3D Documentation, and Rock Art—Approaching and Reflecting on the Automation of Identification and Classification of Rock Art Images

2021 ◽  
Author(s):  
Christian Horn ◽  
Oscar Ivarsson ◽  
Cecilia Lindhé ◽  
Rich Potter ◽  
Ashely Green ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Bronze Age ◽  
Data Augmentation ◽  
Rock Art ◽  
Training Data ◽  
Future Research ◽  
Rock Surface ◽  
Southern Scandinavia ◽  
Bounding Boxes

AbstractRock art carvings, which are best described as petroglyphs, were produced by removing parts of the rock surface to create a negative relief. This tradition was particularly strong during the Nordic Bronze Age (1700–550 BC) in southern Scandinavia with over 20,000 boats and thousands of humans, animals, wagons, etc. This vivid and highly engaging material provides quantitative data of high potential to understand Bronze Age social structures and ideologies. The ability to provide the technically best possible documentation and to automate identification and classification of images would help to take full advantage of the research potential of petroglyphs in southern Scandinavia and elsewhere. We, therefore, attempted to train a model that locates and classifies image objects using faster region-based convolutional neural network (Faster-RCNN) based on data produced by a novel method to improve visualizing the content of 3D documentations. A newly created layer of 3D rock art documentation provides the best data currently available and has reduced inscribed bias compared to older methods. Several models were trained based on input images annotated with bounding boxes produced with different parameters to find the best solution. The data included 4305 individual images in 408 scans of rock art sites. To enhance the models and enrich the training data, we used data augmentation and transfer learning. The successful models perform exceptionally well on boats and circles, as well as with human figures and wheels. This work was an interdisciplinary undertaking which led to important reflections about archaeology, digital humanities, and artificial intelligence. The reflections and the success represented by the trained models open novel avenues for future research on rock art.

scholarly journals An Artificial-Intelligence-Discovered Functional Ingredient, NRT_N0G5IJ, Derived from Pisum sativum, Decreases HbA1c in a Prediabetic Population

Nutrients ◽  
2021 ◽  
Vol 13 (5) ◽  
pp. 1635
Author(s):  
Sweeny Chauhan ◽  
Alish Kerr ◽  
Brian Keogh ◽  
Stephanie Nolan ◽  
Rory Casey ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Pisum Sativum ◽  
Cost Effective ◽  
Glycated Haemoglobin ◽  
Glucose Regulation ◽  
Functional Ingredient ◽  
Effective Manner ◽  
Human Skeletal Muscle Cells ◽  
Increased Risk

The prevalence of prediabetes is rapidly increasing, and this can lead to an increased risk for individuals to develop type 2 diabetes and associated diseases. Therefore, it is necessary to develop nutritional strategies to maintain healthy glucose levels and prevent glucose metabolism dysregulation in the general population. Functional ingredients offer great potential for the prevention of various health conditions, including blood glucose regulation, in a cost-effective manner. Using an artificial intelligence (AI) approach, a functional ingredient, NRT_N0G5IJ, was predicted and produced from Pisum sativum (pea) protein by hydrolysis and then validated. Treatment of human skeletal muscle cells with NRT_N0G5IJ significantly increased glucose uptake, indicating efficacy of this ingredient in vitro. When db/db diabetic mice were treated with NRT_N0G5IJ, we observed a significant reduction in glycated haemoglobin (HbA1c) levels and a concomitant benefit on fasting glucose. A pilot double-blinded, placebo controlled human trial in a population of healthy individuals with elevated HbA1c (5.6% to 6.4%) showed that HbA1c percentage was significantly reduced when NRT_N0G5IJ was supplemented in the diet over a 12-week period. Here, we provide evidence of an AI approach to discovery and demonstrate that a functional ingredient identified using this technology could be used as a supplement to maintain healthy glucose regulation.

scholarly journals Overt acts of perceived discrimination reported by British working-age adults with and without disability

2019 ◽  
Author(s):  
Eric Emerson ◽  
Allison Milner ◽  
Zoe Aitken ◽  
Lauren Krnjacki ◽  
Cathy Vaughan ◽  
...  
Keyword(s):  
Public Health ◽  
Negative Impact ◽  
Policy Research ◽  
Secondary Analysis ◽  
People With Disabilities ◽  
Public Health Problem ◽  
Health Staff ◽  
Cross Sectional ◽  
Increased Risk ◽  
The Uk

Abstract Background Exposure to discrimination can have a negative impact on health. There is little robust evidence on the prevalence of exposure of people with disabilities to discrimination, the sources and nature of discrimination they face, and the personal and contextual factors associated with increased risk of exposure. Methods Secondary analysis of de-identified cross-sectional data from the three waves of the UK’s ‘Life Opportunities Survey’. Results In the UK (i) adults with disabilities were over three times more likely than their peers to be exposed to discrimination, (ii) the two most common sources of discrimination were strangers in the street and health staff and (iii) discrimination was more likely to be reported by participants who were younger, more highly educated, who were unemployed or economically inactive, who reported financial stress or material hardship and who had impairments associated with hearing, memory/speaking, dexterity, behavioural/mental health, intellectual/learning difficulties and breathing. Conclusions Discrimination faced by people with disabilities is an under-recognised public health problem that is likely to contribute to disability-based health inequities. Public health policy, research and practice needs to concentrate efforts on developing programs that reduce discrimination experienced by people with disabilities.

Prediction Models for Truck Accidents at Freeway Ramps in Washington State Using Regression and Artificial Intelligence Techniques

1998 ◽  
Vol 1635 (1) ◽  
pp. 30-36 ◽  
Author(s):  
Wael H. Awad ◽  
Bruce N. Janson
Keyword(s):  
Artificial Intelligence ◽  
Neural Networks ◽  
Linear Regression ◽  
Prediction Models ◽  
Washington State ◽  
Training Data ◽  
Coefficient Of Determination ◽  
Training Process ◽  
Truck Accidents ◽  
High Level

Three different modeling approaches were applied to explain truck accidents at interchanges in Washington State during a 27-month period. Three models were developed for each ramp type including linear regression, neural networks, and a hybrid system using fuzzy logic and neural networks. The study showed that linear regression was able to predict accident frequencies that fell within one standard deviation from the overall mean of the dependent variable. However, the coefficient of determination was very low in all cases. The other two artificial intelligence (AI) approaches showed a high level of performance in identifying different patterns of accidents in the training data and presented a better fit when compared to the regression model. However, the ability of these AI models to predict test data that were not included in the training process showed unsatisfactory results.

Sexual minority status and psychotic symptoms: findings from the Netherlands Mental Health Survey and Incidence Studies (NEMESIS)

2013 ◽  
Vol 44 (2) ◽  
pp. 421-433 ◽  
Author(s):  
M. J. Gevonden ◽  
J. P. Selten ◽  
I. Myin-Germeys ◽  
R. de Graaf ◽  
M. ten Have ◽  
...  
Keyword(s):  
Mental Health ◽  
Psychotic Symptom ◽  
Sexual Minorities ◽  
Mental Health Problems ◽  
Psychotic Symptoms ◽  
Minority Status ◽  
Foreign Born ◽  
Cross Sectional Survey ◽  
Increased Risk ◽  
Incidence Studies

BackgroundEthnic minority position is associated with increased risk for psychotic outcomes, which may be mediated by experiences of social exclusion, defeat and discrimination. Sexual minorities are subject to similar stressors. The aim of this study is to examine whether sexual minorities are at increased risk for psychotic symptoms and to explore mediating pathways.MethodA cross-sectional survey was performed assessing cumulative incidence of psychotic symptoms with the Composite International Diagnostic Interview in two separate random general population samples (NEMESIS-1 and NEMESIS-2). Participants were sexually active and aged 18–64 years (n = 5927, n = 5308). Being lesbian, gay or bisexual (LGB) was defined as having sexual relations with at least one same-sex partner during the past year. Lifetime experience of any psychotic symptom was analysed using logistic regression, adjusted for gender, educational level, urbanicity, foreign-born parents, living without a partner, cannabis use and other drug use.ResultsThe rate of any psychotic symptom was elevated in the LGB population as compared with the heterosexual population both in NEMESIS-1 [odds ratio (OR) 2.56, 95% confidence interval (CI) 1.71–3.84] and NEMESIS-2 (OR 2.30, 95% CI 1.42–3.71). Childhood trauma, bullying and experience of discrimination partly mediated the association.ConclusionsThe finding that LGB orientation is associated with psychotic symptoms adds to the growing body of literature linking minority status with psychosis and other mental health problems, and suggests that exposure to minority stress represents an important mechanism.

Sign in / Sign up

Export Citation Format

Share Document