China
Africa
Explore chapters and articles related to this topic
The Promise of Artificial Intelligence and Machine Learning
Published in Paul Cerrato, John Halamka, Reinventing Clinical Decision Support, 2020
Although the management of active disease is essential, there is also a place for AI in predicting diabetes in prediabetic patients. The US Centers for Disease Control and Prevention estimates that about 84 million American adults have prediabetes, which translates into more than 1 out of 3 citizens.38 Government officials estimate that between 15% and 30% of these prediabetics will develop Type 2 disease within 3 to 5 years without lifestyle changes.39 A data analysis performed by Allscripts, however, suggests that 80% of at-risk patients will develop active disease.40 With these troubling statistics in mind, it behooves clinicians and patients alike to consider the value of diabetes prevention programs. The National Diabetes Prevention Program, a partnership of public and private organizations, has several practical tools and resources to address this issue, including a simple risk-analysis questionnaire.41
Type 2 Diabetes in the Workplace
Published in Thomas Geisen, Henry Harder, Disability Management and Workplace Integration, 2016
According to the CDA 2009 report (Canadian Diabetes Association 2009b), the total population with T2D has increased from 4.2 per cent in 2000 to 7.3 per cent in 2010 and will increase to 9.9 per cent by 2020. Today, one in four Canadians either has diabetes or prediabetes. Prediabetes is a condition where a person’s blood sugar levels are higher than normal, but not high enough to be diagnosed as T2D (fasting plasma glucose levels of 7.9 mmol/L or higher). The CDA 2009 report (Canadian Diabetes Association 2009b) estimates that nearly six million Canadians today are living with prediabetes. Prediabetes is of concern to clinicians and health care professionals since almost 50 per cent of those with the condition will ultimately develop T2D. A number of studies have mentioned that comorbidities or complications associated with T2D, such as CVD and neuropathy, begin during the prediabetic stage (CDA 2009b; Cohn 2008; Public Health Agency of Canada 2007).
Use of Artificial Intelligence in the Screening and Treatment of Chronic Diseases
Published in Sandeep Reddy, Artificial Intelligence, 2020
Chaitanya Mamillapalli, Daniel J. Fox, Ramanath Bhandari, Ricardo Correa, Vishnu Vardhan Garla, Rahul Kashyap
Currently recommended diabetes laboratory screening tests comprise fasting blood glucose, hemoglobin A1c, and oral glucose tolerance tests. A diabetes diagnosis is confirmed by a 2-hour glucose tolerance test measuring >200 mg/dl or A1c ≥ 6.5% or fasting glucose levels >126 mg/dl. Prediabetes is a precursor for diabetes with glucose levels above normal levels but less than the defined thresholds for diabetes (2-hour glucose levels at 140–200 mg/dl, A1c 5.7%–6.5%, and fasting glucose at 100–126 mg/dl) (Bowen et al., 2018). Approximately 37%–70% of patients with prediabetes progress into type 2 diabetic status within four years of onset at an overall rate of 10%/year (Nathan et al., 2007).
A Church-Based Diabetes Risk Factor Prevention Program Improves Psychosocial Factors and Food-Related Behaviors
Published in Journal of Hunger & Environmental Nutrition, 2020
Joel Gittelsohn, Elizabeth T. Anderson Steeves, Jessica J Ho, Ahyoung Shin, Harmony Farner, Amber Summers
In 2014, 29.1 million adults in the United States were living with diabetes, and another 86 million were living with pre-diabetes.1 African Americans (AA) are disproportionately affected by diabetes, with 9.5% of African Americans diagnosed with diabetes as compared to 5.8% of Whites in 2014,2 and low-income African Americans are especially negatively affected.3 In addition, more than 27% of AA adults have hypertension, one of the risk factors for type-2 diabetes.4,5 The prevention of diabetes is strongly associated with improvements in lifestyle behaviors, such as diet and physical activity.6 Moreover, increased consumption of healthy foods such as fruits and vegetables as well as increased physical activity are protective against obesity, a major diabetes risk factor.7–9
Ensemble Architecture for Prediction of Grading of Diabetic Retinopathy
Published in Cybernetics and Systems, 2022
Shruti Jain, Sanket Saxena, Shivam Sinha
More than 37 million people in the United States have diabetes. Over one third adults, i.e., 96 million US adults have pre-diabetes, where 8 in 10 of them don’t know they have it. Now a day’s Diabetes is the seventh leading cause of death in the United States (Centre for Disease Control and Prevention). A lot of sugar in the blood can prompt the blockage of the minuscule veins that support the retina, removing its blood flexibly. There are two types of DR namely Early diabetic retinopathy (EDR) & Advanced diabetic retinopathy (ADR). EDR is the progressive basic structure called no proliferative diabetic retinopathy (NPDR). In EDR, fresh blood vessels are not developing (proliferating). When having NPDR, the dividers of the veins in the retina are debilitated (Muangnak et al. 2015). The bigger retinal vessels start widening and get unpredictable in measurement. NPDR can advance from gentle to serious, as more veins become blocked and the nerve filaments in the retina start growing. In ADR, DR advances to the extreme sort, known as proliferative diabetic retinopathy. In this sort, harmed veins close off, causing the development of new, unusual veins in the retina, and can spill into the reasonable, jam-like substance that fills the focal point of the eye (Bhardwaj, Jain, and Sood 2021a). Based on ADR and EDR, DR is divided into different severity levels (as shown in Figure 2) namely non-diseased (severity level 0), mild (severity level 1), moderate (severity level 2), severe (severity level 3), and proliferative DR (severity level 4). The danger of building up the eye condition can increment because of high blood pressure, pregnancy, high cholesterol, tobacco use, glucose level.
Artificial Intelligence-based Prediction of Diabetes and Prediabetes Using Health Checkup Data in Korea
Published in Applied Artificial Intelligence, 2022
Hyeonseop Yuk, Juhui Gim, Jung Kee Min, Jaesuk Yun, Tae-Young Heo
Furthermore, we identified prediction factors and consolidated them into the physiological status of prediabetes and T2D, respectively. First, FBG and HbA1c levels were the most significant factors for the early prediction of both prediabetes and T2D. These results are expected to be based on the pathophysiology of diabetes mellitus. FBG or HbA1c is also regarded as the top-ranked factor by Boruta, SelectKBest, and Lasso method. These results suggest that all methods in our study showed similar feature selection performance. However, we demonstrated that FBG had the strongest effect on future prediabetes, even in the normal blood glucose range. High BFP, abnormal liver and thyroid function, and electrolyte imbalance are expected to have significant effects on the development of prediabetes in future. These abnormal conditions may be related to metabolic disturbances and endocrine homeostasis. Therefore, high glucose levels within the normal range may be associated with prediabetes development in individuals with the abnormal physiological conditions mentioned above. Similarly, altered kidney function and inflammation/infection are prediction factors for the future development of T2D. Surprisingly, symptoms of kidney malfunction, as evidenced by BUN and creatinine tests, were demonstrated even before T2D development. Generally, poor management of diabetes induces kidney failure due to pathological vascular alterations (Braunwald 2019). However, our study showed that kidney dysfunction in prediabetic individuals may develop before T2D. In addition, inflammation, as evidenced by CRP and eosinophil counts within prediabetes, is a prediction factor for future diabetes. According to aprevious report (Tariq et al. 2020), eosinophilia is associated with inflammation in kidney diseases. These results suggest that the progression from prediabetes to diabetes can be predicted in individuals with altered kidney function. Odds ratio analysis showed that HbA1c had the highest value for prediabetes and diabetes progression as 3.76 and 5.77, respectively. Particularly, even in the normal stage where the FBG value is less than 100, HbA1c was identified as the most significant risk factor for prediabetes.