This paper proposes a method for automatic classification of spectral domain OCT data for the identification of patients with retinal diseases such as Diabetic Macular Edema (DME). We address this issue as an anomaly detection problem and propose a method that not only allows the classification of the OCT volume, but also allows the identification of the individual diseased B-scans inside the volume. Our approach is based on modeling the appearance of normal OCT images with a Gaussian Mixture Model (GMM) and detecting abnormal OCT images as outliers. The classification of an OCT volume is based on the number of detected outliers. Experimental results with two different datasets show that the proposed method achieves a sensitivity and a specificity of 80% and 93% on the first dataset, and 100% and 80% on the second one. Moreover, the experiments show that the proposed method achieves better classification performance than other recently published works.
Ethnicity is an important factor in diabetes care. The understanding of its effect in this country may help to improve diabetes care, glycaemic control and diabetic complication rates. This study was to determine the diabetes control profile in relation to complication rates between the three main ethnics group in Malaysia.
Nasal septum abscesses caused by uncontrolled diabetes mellitus are rare. We report 3 cases. Very few cases have been published concerning non-traumatic nasal septum abscesses. The development of the condition, possible complications, and treatment are discussed.
We present an interesting but high-risk case of an obese male patient aged 56 years with dextrocardia and a left diaphragmatic hernia. Anterior myocardial infarction was diagnosed in 1994, and the patient later presented with a history of unstable angina. The diagnosis for this chronic smoker was triple-vessel disease, impaired left ventricular function, chronic renal failure, chronic bronchitis, impaired lung function, pulmonary hypertension, hypertension, diabetes, and chronic active gastritis (EuroSCORE of 10). The patient underwent successful off-pump coronary artery bypass grafting with 3 saphenous vein grafts to the left anterior descending, obtuse marginal, and right posterior descending arteries. He was discharged home 8 days later.
In 2005, the incidence of renal replacement therapy (RRT) in the United Kingdom was 110 per million of the population (pmp) using the day 0 definition and 103 pmp using the day 90 definition. Relative to the 42 countries reporting data to the USRDS, the day 0 and day 90 rates for RRT incidence in the UK are the 32nd and 35th lowest, respectively. However, the overall incidence for the UK masks higher rates in Scotland, Wales and Northern Ireland (123, 129 and 140 pmp, respectively). Of the six countries with RRT incidence rates comparable with those in the UK (Australia, Finland, Malaysia, New Zealand, Norway and the Netherlands) three had relatively high rates for the age band 20-44, and two had relatively high rates for the age band 45-60. The proportion of incident patients with diabetes as the cause of established renal failure also varied considerably among these six comparator countries from 16% to 40% but rates of peritoneal dialysis utilization were comparable with that in the UK and generally higher than in countries with higher rates of RRT incidence. When transplantation rates were considered alongside prevalence rates for RRT, the UK position appeared relatively high at 46% (11th out of 37 countries), although still considerably lower than in Norway and the Netherlands (72 and 54%, respectively). Although variation in RRT incidence rate exists within the four countries of the UK, the overall RRT incidence, reported for the first time this year, appears similar to that observed in a number of demographically similar countries around the world. Examining the UK alongside the six comparator countries, different patterns of RRT incidence were observed across the age bands and variation in the RRT incidence secondary to diabetes mellitus raised interesting questions. The higher rates of renal transplantation achieved in several of the comparator countries also justifies further analysis.
Background: Hand infection in diabetics is an often ignored but challenging condition. If not addressed effectively, it may result in long term disability, contracture, amputation and even death. Methods: From August 2014 to December 2015, a study was done in our centre, where 49 diabetic hand infection cases were analyzed in two groups, superficial and deep hand infection. Results: Mean age of the patients was 51.63 years. There were 21 superficial infections and 28 deep infections. Cause of infection was unknown or spontaneous in 16 cases, traumatic laceration or crush in 14 patients, following minor prick in 10 cases. Most of the cases were the results of neglected minor wound. Forty-one patients were insulin dependent. Five cases were diagnosed as diabetic at the time of treatment. Four patients were treated conservatively and 45 (92%) cases required operation in the form of incision, drainage and debridement. In 16 (35%) cases, wound was left open and was healed by secondary intention following regular dressing. In five patients, wound was closed secondarily. Partial thickness skin graft was applied in 15 cases. Seven patients were treated with flap coverage. Partial digital or ray amputation were done in 16 cases. All fingers except thumb were amputated in one case and amputation from wrist was done in another patient. Wound swabs were taken, and antibiotics were changed or continued accordingly. But reports of 26 cases were available. No growth was found in four cases, monomicrobial infection was found in 15 patients and polymicrobial in seven cases. Infection resolved with healing in 47 cases. Two patients died during treatment from sepsis, both were insulin dependent, had associated renal failure and from deep infection group. One patient developed severe mental disorder. Conclusions: For diabetic hand infection, early diagnosis and prompt treatment with appropriate antibiotics and emergency surgery with extensile incision is crucial. Primary amputation of the part could be life and limb saving.
BACKGROUND & AIMS: Diabetes is a risk factor for colorectal cancer. We studied the association between markers of glucose metabolism and metabolic syndrome and the presence of colorectal adenomas in a large number of asymptomatic men and women attending a health screening program in South Korea. We also investigated whether these associations depend on adenoma location.
METHODS: In a cross-sectional study, we measured fasting levels of glucose, insulin, hemoglobin A1c, and C-peptide and calculated homeostatic model assessment (HOMA) values (used to quantify insulin resistance) for 19,361 asymptomatic South Korean subjects who underwent colonoscopy examinations from January 2006 to June 2009. Participants completed a standardized self-administered health questionnaire and a validated semiquantitative food frequency questionnaire. Blood samples were collected on the day of the colonoscopy; fasting blood samples were also collected. Robust Poisson regression was used to model the associations of glucose markers with the prevalence of any adenoma.
RESULTS: Using detailed multivariable-adjusted dose-response models, the prevalence ratios (aPR, 95% confidence interval [CI]) for any adenoma, comparing the 90th with the 10th percentile, were 1.08 (1.00-1.16; P = .04) for fasting glucose, 1.07 (0.99-1.15; P = .10) for insulin, 1.09 (1.02-1.18, P = .02) for HOMA, 1.09 (1.01-1.17; P = .02) for hemoglobin A1c, and 1.14 (1.05-1.24; P = .002) for C-peptide. The corresponding ratios for nonadvanced adenomas were 1.11 (0.99-1.25; P = .08), 1.10 (0.98-1.24; P = .12), 1.15 (1.02-1.29; P = .02), 1.14 (1.01-1.28; P = .03), and 1.20 (1.05-1.37; P = .007), respectively. The corresponding ratios for advanced adenomas were 1.32 (0.94-1.84; P = .11), 1.23 (0.87-1.75; P = .24), 1.30 (0.92-1.85; P = .14), 1.13 (0.79-1.61; P = .50), and 1.67 (1.15-2.42; P = .007), respectively. Metabolic syndrome was associated with the prevalence of any adenoma (aPR, 1.18; 95% CI, 1.13-1.24; P < .001), nonadvanced adenoma (aPR, 1.30; 95% CI, 1.20-1.40; P < .001), and advanced adenoma (aPR, 1.42; 95% CI, 1.14-1.78; P = .002). Associations were similar for adenomas located in the distal versus proximal colon.
CONCLUSIONS: Increasing levels of glucose, HOMA values, levels of hemoglobin A1c and C-peptide, and metabolic syndrome are significantly associated with the prevalence of adenomas. Adenomas should be added to the list of consequences of altered glucose metabolism.
The current hypertension (HTN) guidelines recommend the assessment of other cardiovascular disease (CVD) risk factors in individuals with HTN for further management. Few studies in Asian populations have been published to identify the outcome of individuals with HTN and other CVD risk factors. This study aims to assess the effect of HTN alone, and in combination with other CVD risk factors on all-cause and CVD mortality. Three cross-sectional studies carried out in Singapore (baseline 1982--1995) consisting of 5830 persons were grouped by the absence or presence of HTN and CVD risk factors. They were followed-up (mean 14.1 years) by linkage with the National Death Register. Cox's proportional hazards model was used to obtain adjusted hazard ratios (HRs) for risk of mortality. HTN individuals with either <2 CVD risk factors (adjusted HR 1.4; 95% confidence interval (CI) 1.0-1.8) or > or =2 CVD risk factors (adjusted HR 2.3; 95% CI 1.9-3.0) were at increased risk of all-cause mortality compared to normotensive individuals. The findings were similar for CVD mortality. HTN individuals who also smoked or had diabetes were at highest risk of all-cause mortality, whereas those with elevated total cholesterol/high-density lipoprotein cholesterol, smoked or diabetes had the highest risk for CVD mortality. These findings show that in HTN individuals it is important to assess the presence of other CVD risk factors and manage accordingly.