Characterising metabolically healthy obesity in weight-discordant monozygotic twins.

AIMS/HYPOTHESIS: Not all obese individuals display the metabolic disturbances commonly associated with excess fat accumulation. Mechanisms maintaining this 'metabolically healthy obesity' (MHO) are as yet unknown. We aimed to study different fat depots and transcriptional pathways in subcutaneous adipose tissue (SAT) as related to the MHO phenomenon. METHODS: Sixteen rare young adult obesity-discordant monozygotic (MZ) twin pairs (intra-pair difference (∆) in BMI ≥ 3 kg/m(2)), aged 22.8-35.8 years, were examined for detailed characteristics of metabolic health (subcutaneous, intra-abdominal and liver fat [magnetic resonance imaging/spectroscopy]), OGTT, lipids, adipokines and C-reactive protein (CRP). Affymetrix U133 Plus 2.0 chips were used to analyse transcriptomics pathways related to mitochondrial function and inflammation in SAT. RESULTS: Based on liver fat accumulation, two metabolically different subgroups emerged. In half (8/16) of the pairs (∆weight 17.1 ± 2.0 kg), the obese co-twin had significantly higher liver fat (∆718%), 78% increase in AUC insulin during OGTT and CRP, significantly more disturbance in the lipid profile and greater tendency for hypertension compared with the lean co-twin. In these obese co-twins, SAT expression of mitochondrial oxidative phosphorylation, branched-chain amino acid catabolism, fatty acid oxidation and adipocyte differentiation pathways were downregulated and chronic inflammation upregulated. In the other eight pairs (∆weight 17.4 ± 2.8 kg), the obese co-twin did not differ from the non-obese co-twin in liver fat (∆8%), insulin sensitivity, CRP, lipids, blood pressure or SAT transcriptomics. CONCLUSIONS/INTERPRETATION: Our results suggest that maintenance of high mitochondrial transcription and lack of inflammation in SAT are associated with low liver fat and MHO.

HbA(1c) in diagnosing and predicting Type 2 diabetes in impaired glucose tolerance: the Finnish Diabetes Prevention Study.

AIMS: We analysed the Finnish Diabetes Prevention Study data in order to evaluate how the new HbA(1c) -based criterion compares with the oral glucose tolerance test in diagnosing Type 2 diabetes among high-risk individuals during a prospective average follow-up of 4 years. METHODS: In the Diabetes Prevention Study, 172 men and 350 women who were overweight and had impaired glucose tolerance at baseline were randomized into an intensive lifestyle intervention or a control group. The oral glucose tolerance test and HbA(1c) measurements were performed annually until the diagnosis of diabetes using the World Health Organization 1985 criteria. RESULTS: The sensitivity of the HbA(1c) ≥ 6.5% (≥ 48 mmol/mol) as a diagnostic criterion for Type 2 diabetes was 35% (95% CI 24%, 47%) in women and 47% (95% CI 31%, 64%) in men compared with diagnosis based on two consecutive oral glucose tolerance tests. The corresponding sensitivities for HbA(1c) ≥ 6.0% (≥ 42 mmol/mol) were 67% (95% CI 55%, 77%) and 68% (95% CI 51%, 82%). The participants with HbA(1c) ≥ 6.5% (≥ 48 mmol/mol) and diabetes based on the oral glucose tolerance test were more obese and had higher fasting glucose and 2-h glucose concentrations than those who had a diabetic oral glucose tolerance test but HbA(1c) < 6.5% (< 48 mmol/mol). There were no differences in the predictive performance of baseline fasting glucose, oral glucose tolerance test and HbA(1c) . CONCLUSIONS: Of those with diabetes diagnosis based on two oral glucose tolerance tests during the Diabetes Prevention Study follow-up, 60% would have remained undiagnosed if diagnosis had been based on HbA(1c) ≥ 6.5% (≥ 48 mmol/mol) criterion.

The metabolic syndrome as a predictor of incident diabetes and cardiovascular events in the Health 2000 Study.

AIM: The study aimed to examine the role of the metabolic syndrome (MetS) as a predictor of incident cardiovascular disease (CVD) events and incident diabetes, and to compare the various definitions of MetS. METHODS: The population-based Health 2000 Study included 6105 individuals, aged 30-79 years, followed-up for 7 years. CVD during follow-up was defined as coronary death, acute myocardial infarction, coronary revascularization or stroke. MetS was defined according to the International Diabetes Federation (IDF), the 2005 National Cholesterol Education Program-Adult Treatment Panel III (ATP III), the World Health Organization (WHO) and the new Harmonization definitions. The Bayesian information criterion (BIC) was used to compare different Cox proportional-hazards regression models. RESULTS: The highest prevalence estimates of MetS at baseline were observed with the Harmonization definition: 47.8% in men and 40.7% in women. During the follow-up, 238 cases of incident CVD and 172 cases of incident diabetes were observed. All definitions of MetS were significant predictors for incident CVD and diabetes. BIC suggested that the new Harmonization definition of MetS as one entity was a better predictor of the CVD endpoint than the sum of its components, but not for diabetes. Also, the Harmonization definition of MetS was a better predictor of CVD than the Framingham equation in women, but not in men. CONCLUSION: Irrespective of definition, MetS is a significant predictor of incident CVD events and incident diabetes. Also, the new Harmonization definition may be a better predictor of incident CVD than the sum of its components.

Take action to prevent diabetes--the IMAGE toolkit for the prevention of type 2 diabetes in Europe.

When we ask people what they value most, health is usually top of the list. While effective care is available for many chronic diseases, the fact remains that for the patient, the tax payer and the whole of society: prevention is better than cure. Diabetes and its complications are a serious threat to the survival and well-being of an increasing number of people. It is predicted that one in ten Europeans aged 20-79 will have developed diabetes by 2030. Once a disease of old age, diabetes is now common among adults of all ages and is beginning to affect adolescents and even children. Diabetes accounts for up to 18 % of total healthcare expenditure in Europe. The good news is that diabetes is preventable. Compelling evidence shows that the onset of diabetes can be prevented or delayed greatly in individuals at high risk (people with impaired glucose regulation). Clinical research has shown a reduction in risk of developing diabetes of over 50 % following relatively modest changes in lifestyle that include adopting a healthy diet, increasing physical activity, and maintaining a healthy body weight. These results have since been reproduced in real-world prevention programmes. Even a delay of a few years in the progression to diabetes is expected to reduce diabetes-related complications, such as heart, kidney and eye disease and, consequently, to reduce the cost to society. A comprehensive approach to diabetes prevention should combine population based primary prevention with programmes targeted at those who are at high risk. This approach should take account of the local circumstances and diversity within modern society (e.g. social inequalities). The challenge goes beyond the healthcare system. We need to encourage collaboration across many different sectors: education providers, non-governmental organisations, the food industry, the media, urban planners and politicians all have a very important role to play. Small changes in lifestyle will bring big changes in health. Through joint efforts, more people will be reached. The time to act is now.

Quality indicators for the prevention of type 2 diabetes in Europe--IMAGE.

BACKGROUND: The marked increase of type 2 diabetes necessitates active development and implementation of efficient prevention programs. A European level action has been taken by launching the IMAGE project to unify and improve the various prevention management concepts, which currently exist within the EU. This report describes the background and the methods used in the development of the IMAGE project quality indicators for diabetes primary prevention programs. It is targeted to the persons responsible for diabetes prevention at different levels of the health care systems. METHODS: Development of the quality indicators was conducted by a group of specialists representing different professional groups from several European countries. Indicators and measurement recommendations were produced by the expert group in consensus meetings and further developed by combining evidence and expert opinion. RESULTS: The quality indicators were developed for different prevention strategies: population level prevention strategy, screening for high risk, and high risk prevention strategy. Totally, 22 quality indicators were generated. They constitute the minimum level of quality assurance recommended for diabetes prevention programs. In addition, 20 scientific evaluation indicators with measurement standards were produced. These micro level indicators describe measurements, which should be used if evaluation, reporting, and scientific analysis are planned. CONCLUSIONS: We hope that these quality tools together with the IMAGE guidelines will provide a useful tool for improving the quality of diabetes prevention in Europe and make different prevention approaches comparable.

The efficacy of rehabilitation for patients with rheumatoid arthritis: comparison between a 4-week rehabilitation programme in a warm and cold climate.

OBJECTIVES: To investigate the long-term effect (week 16) of a 4-week rehabilitation programme for patients with rheumatoid arthritis (RA) and to compare the effect of this intervention given in a Mediterranean or a Norwegian climate. METHODS: A randomized, controlled, parallel group design, where 124 RA patients applying for rehabilitation were randomized to a rehabilitation programme either in Norway or in a Mediterranean climate. The participants were examined clinically immediately before (week 0) and after (week 4) the rehabilitation period as well as in week 16 and answered a mailed questionnaire in week 28. The 28-Joint Disease Activity Score (DAS28), American College of Rheumatology (ACR) response and physical tests were used to measure clinical response. RESULTS: The baseline DAS28 value 4.45 (1.16) was reduced by -0.95 (1.05) in the Mediterranean climate and the baseline DAS28 value 4.18 (1.17) was reduced by -0.37 (0.92) in the Norwegian climate at week 16 (p = 0.003). An ACR20 improvement was achieved in 25% of the patients treated in the Mediterranean climate and in 15% of those treated in the Norwegian climate. Sustained improvement in all ACR core components at week 16 and in patient's assessment of health status at week 28 was found in the patients treated in the Mediterranean climate only. Tests of physical function, the 6-Minute Walk Test (6MWT) and the Timed Up and Go (TUG), showed comparable improvements in patients treated in both climates. CONCLUSIONS: RA patients showed immediate positive effects with regard to disease activity, physical function, and symptoms during a 4-week rehabilitation programme. The effects on disease activity and symptoms were larger and better maintained at least 3 months after rehabilitation in a warm rather than in a cold climate.

Long-term prognosis after coronary artery bypass surgery.

OBJECTIVE: To analyse the risk of coronary heart disease (CHD) events and total mortality among patients who had coronary artery bypass graft (CABG) surgery during 1988-1992. METHODS: A population-based myocardial infarction (MI) register included data on invasive cardiac procedures among residents of the study area. The subjects aged 35-64 years were followed-up for 12 years for non-fatal and fatal CHD events and all-cause mortality, excluding events within 30 days of the CABG operation. CABG was performed on 1158 men and 215 women. RESULTS: The overall survival of men who underwent CABG was similar to the survival of the corresponding background population for about ten years but started to worsen after that. At twelve years of follow-up, 23% (n=266, 95% CI 234-298) of the men who had undergone the operation had died, while the expected proportion, based on mortality in the background population, was 20% (n=231, 95% CI 226-237). The CHD mortality of men who had undergone the operation was clearly higher than in the background population. Among women, the mortality after CABG was about twice the expected mortality in the corresponding background population. In Cox proportional hazards models age, smoking, history of MI, body mass index and diabetes were significant predictors of mortality. CONCLUSIONS: The prognosis of male CABG patients did not differ from the prognosis of the corresponding background population for about ten years, but started to deteriorate after that. History of MI prior to CABG and major cardiovascular risk factors was a predictor of an adverse outcome.

Level A in vitro-in vivo correlation (IVIVC) model with Bayesian approach to formulation series.

In vitro-in vivo correlation (IVIVC) models for formulation series are useful in drug development, but the current models are limited by their inability to include data variability in the predictions. Our goal was to develop a level A IVIVC model that provides predictions with probabilities. The Bayesian approach was used to describe uncertainty related to the model and the data. Three bioavailability studies of levosimendan were used to develop IVIVC model. Dissolution was tested at pH 5.8 with basket. The IVIVC model with Bayesian approach consisted of prior and observed data. All observed data were fitted to the one-compartment model together with prior data. Probability distributions of pharmacokinetic parameters and concentration time profiles were obtained. To test the external predictability of IVIVC model, only dissolution data of formulations E and F were used. The external predictability was good. The possibility to utilize all observed data when constructing IVIVC model, can be considered as a major strength of Bayesian approach. For levosimendan capsule data traditional IVIVC model was not predictable. The usefulness of IVIVC model with Bayesian approach was shown with our data, but the same approach can be used more widely for formulation optimization and for dissolution based biowaivers.

Myocardial infarction in diabetic and non-diabetic persons with and without prior myocardial infarction: the FINAMI Study.

AIMS/HYPOTHESIS: We compared the risk of acute coronary events in diabetic and non-diabetic persons with and without prior myocardial infarction (MI), stratified by age and sex. METHODS: A Finnish MI-register study known as FINAMI recorded incident MIs and coronary deaths (n=6988) among people aged 45 to 74 years in four areas of Finland between 1993 and 2002. The population-based FINRISK surveys were used to estimate the numbers of persons with prior diabetes and prior MI in the population. RESULTS: Persons with diabetes but no prior MI and persons with prior MI but no diabetes had a markedly greater risk of a coronary event than persons without diabetes and without prior MI. The rate of recurrent MI among non-diabetic men with prior MI was higher than the incidence of first MI among diabetic men aged 45 to 54 years. The rate ratio was 2.14 (95% CI 1.40-3.27) among men aged 50. Among elderly men, diabetes conferred a higher risk than prior MI. Diabetic women had a similar risk of suffering a first MI as non-diabetic women with a prior MI had for suffering a recurrent MI. CONCLUSIONS/INTERPRETATION: Both persons with diabetes but no prior MI, and persons with a prior MI but no diabetes are high-risk individuals. Among men, a prior MI conferred a higher risk of a coronary event than diabetes in the 45-54 year age group, but the situation was reversed in the elderly. Among diabetic women, the risk of suffering a first MI was similar to the risk that non-diabetic women with prior MI had of suffering a recurrent MI.

Five-year risk of developing clinical diabetes after first myocardial infarction; the FINAMI study.

AIM: To investigate the incidence of clinical diabetes as determined by the incidence of diabetes drug reimbursements within a 5-year period after the first myocardial infarction (MI) in patients who were non-diabetic at the time of their first MI. RESEARCH DESIGN AND METHODS: A population-based MI register, FINMONICA/FINAMI, recorded all coronary events in persons of 35-64 years of age between 1988 and 2002 in four study areas in Finland. These records were used to identify subjects sustaining their first MI (n = 2632). Participants of the population-based risk factor survey FINRISK (surveys 1987, 1992, 1997 and 2002), who did not have diabetes or a history of MI, served as the control group (n = 7774). The FINMONICA/FINAMI study records were linked with the National Social Security Institute's drug reimbursement records, which include diabetes medications, using personal identification codes. The records were used to identify subjects who developed diabetes during the 5-year follow-up period (n = 98 in the MI group and n = 79 in the control group). RESULTS: Sixteen per cent of men and 20% of women sustaining their first MI were known to have diabetes and thus were excluded from this analysis. Non-diabetic men having a first MI were at more than twofold {hazard ratio (HR) 2.3 [95% confidence interval (CI) 1.6-3.4]}, and women fourfold [HR 4.3 (95% CI 2.4-7.5)], risk of developing diabetes mellitus during the next 5 years compared with the control population without MI. CONCLUSIONS: Many patients who do not have diabetes at the time of their first MI develop diabetes in the following 5 years.

Serum homocysteine, creatinine, and glucose as predictors of the severity and extent of coronary artery disease in asymptomatic members of high-risk families.

BACKGROUND: There has been no previous study to determine the severity and extent of coronary artery disease (CAD) in subjects with no diagnosis or symptoms of CAD at the time of the angiography. METHODS: Fifty-three subjects, who were siblings of patients with early onset CAD, underwent coronary angiography. Indices to describe per-patient characteristics of CAD were calculated, based on computer-aided quantitative coronary angiography. Clinical and laboratory characteristics were correlated to the angiographic parameters. RESULTS: Serum total homocysteine (rho = 0.29, P < 0.05) and creatinine (rho = 0.47, P = 0.001) levels were related to the global atheroma burden index. The median of the atheroma burden index was two times higher in the top homocysteine quartile compared to the lowest quartile. The overall atheroma burden index correlated significantly with the fasting blood glucose level in all subjects. Diabetes, especially when albuminuria was present, was a powerful risk factor. In a multivariate analysis, only age and sex were independent predictors of atheroma burden. CONCLUSIONS: Serum homocysteine and creatinine concentrations, and diabetes with albuminuria were found to be markers of the severity and extent of CAD in subjects of high-risk families without symptoms of CAD.

Cholesterol efflux capacity in vitro predicts the severity and extent of coronary artery disease in patients with and without type 2 diabetes.

OBJECTIVE: To investigate the relation between severity and extent of coronary artery disease (CAD) and in vitro cholesterol efflux capacity. DESIGN: This study consisted of 46 type 2 diabetic, and 42 nondiabetic men undergoing coronary angiography. Quantitative coronary angiography was used to estimate the severity, extent, and overall "atheroma burden" of CAD. The capacity of patient plasma to induce cholesterol efflux from cultured Fu5AH rat hepatoma cells was measured in vitro. RESULTS: In the combined study population (n = 88), there was a significant inverse correlation between efflux and global atheroma burden (r = -0.23, p < 0.05). In the diabetic group, the global atheroma burden index was independently associated both with cholesterol efflux and with LpA-I levels. However, in the nondiabetic CAD group this association was lost when efflux and LpA-I levels were included in the same model. CONCLUSION: The present study demonstrated that efflux capacity was inversely associated with the severity and extent of CAD. In the diabetic group this association was independent of LpA-I levels, suggesting impaired antiatherogenic potential of these particles in type 2 diabetic patients.

Determinants of the severity and extent of coronary artery disease in patients with type-2 diabetes and in nondiabetic subjects.

BACKGROUND: Factors predicting the anatomic distribution and the severity and extent of coronary atherosclerosis in patients with clinically manifest coronary artery disease (CAD) for type-2 diabetic patients could be different than those for nondiabetic patients. OBJECTIVE: To study the determinants of severity and extent of CAD in consecutive patients with type 2 diabetes mellitus, compared with those for matched nondiabetic patients, undergoing clinically indicated coronary angiography. METHODS: Coronary angiograms of 48 men and seven women with type-2 diabetes and an equal number of nondiabetic subjects were analyzed quantitatively. Scores reflecting severity and extent of CAD were compared with potential risk factors using univariate correlation analyses and multivariate regression models. RESULTS: For the diabetics, a global coronary atheroma burden index was independently and directly related to age (P = 0.022) and to level of intermediate-density lipoprotein cholesterol (P = 0.055), and inversely to level of particles of a subtype of high-density lipoprotein (P = 0.022). Several angiographic indexes were related to the duration of diabetes and control of glycemia. For the nondiabetic group, global atheroma burden was independently related to age (P = 0.028), a history of hypertension (P = 0.028), and concentration of low-density lipoprotein (P = 0.013), and inversely to level of apolipoprotein A-I (P = 0.008). The duration of coronary disease and a history of smoking were also predictive of severe coronary atherosclerosis among nondiabetic patients. CONCLUSIONS: Classical risk factors are strong predictors of the severity and extent of coronary atherosclerosis in nondiabetic patients, but the most important determinants for type-2 diabetic patients are levels of triglyceride-rich lipoproteins and apolipoprotein A-I-containing particles of high-density lipoprotein, and factors directly related to diabetes.

Neurologic sequelae after caesarean section.

BACKGROUND: Because pregnancy increases the sensitivity of nervous tissue to local anaesthetics, pregnant patients may be at higher risk of developing neurologic deficits after spinal block than non-pregnant patients. Therefore, we evaluated prospectively the incidence and type of neurologic symptoms after spinal anaesthesia with hyperbaric bupivacaine for caesarean section. METHODS: In this prospective follow-up study we recorded neurologic complications during anaesthesia and postoperatively until discharge from the hospital of 219 patients, who underwent caesarean section under spinal anaesthesia with hyperbaric bupivacaine (5 mg/ml, mean 13 mg). The patients filled in a questionnaire on the first and fifth postoperative days. In the case of complaints typical of neurologic symptoms they were checked first by the anaesthesiologist and, in the case of persistent symptoms, afterwards additionally by a neurologist. RESULTS: Twenty-six of 219 patients were not included in the further evaluation because of incomplete return of their questionnaires. Seventeen mothers (8.8%) complained of transient neurologic symptoms (TNSs), lasting mostly 1-2 days, in the buttocks and/or legs during the first three postoperative days. Eleven patients (5.7%) complained of postdural puncture headache. Two patients (emergency caesarean section because of protracted labour in one and elective caesarean section because of previous caesarean section in the other) complained of persisting pain or sensory abnormalities. Neither of them felt paraesthesia during lumbar puncture. CONCLUSION: Women after caesarean section under a spinal block seem to suffer more often from TNSs than non-pregnant women. The conclusions are, however, uncertain since we had no control group operated on under other than spinal anaesthesia. The persisting neurologic symptoms in two patients might also be due to the obstetric procedure itself. To find out about the validity and possible underlying causes of our results, we need randomised studies with control groups receiving epidural or general anaesthesia.

Angiographic severity and extent of coronary artery disease in patients with type 1 diabetes mellitus.

Studies of the characteristics of coronary artery disease (CAD) in diabetic patients have shown conflicting results. Only 2 studies exploring the severity of CAD, specifically in type 1 diabetes, have been published, and neither of them has used computer-aided quantitative coronary angiography. This retrospective study comprised 64 (24 women and 40 men) type 1 diabetic patients and nondiabetic control subjects. To estimate the severity, extent, and overall "atheroma burden" of CAD, we used quantitative coronary angiographic-based segmental analysis of coronary angiograms. Type 1 diabetic patients had greater global severity (p < 0.001), global extent (p < 0.001), and global atheroma burden (p < 0.001) indexes than nondiabetic control subjects. Quantitative coronary angiographic-derived indexes of CAD were, on average, 1.4- to 4.3-fold higher in diabetic than in nondiabetic patients. These differences were particularly marked in women. We found that type 1 diabetic patients with a clinical indication for coronary angiography, especially women, have more severe, extensive, and distal type of CAD than individually matched nondiabetic control patients. Our findings, including a loss of sex difference for CAD among type 1 diabetic patients and a marked impact of type 1 diabetes in women, are not explained by established risk factors.

Ursodeoxycholic acid and endothelial-dependent, nitric oxide-independent vasodilatation of forearm resistance arteries in patients with coronary heart disease.

AIMS: Ursodeoxycholic acid (UDCA) has cholesterol lowering and anti-inflammatory effects and bile acids are reported to exert vasodilator effects; all of these properties might be considered desirable in a drug used in the treatment of patients with coronary heart disease. We investigated a hypothesis that UDCA may dilate arteries and the mechanism of action. METHODS: We evaluated effects of a 6-week treatment with UDCA in 11 coronary heart disease patients on endothelium-dependent (acetylcholine-induced) and -independent (nitroprusside-induced) vasodilatations in forearm vasculature by strain-gauge plethysmography. Healthy individuals (n=14) served as baseline controls. RESULTS: The percentage increase by acetylcholine in the flow of the infused arm relative to the non-infused arm of coronary heart disease patients during the trial remained unaltered, but vasodilatation to NG-monomethyl-l-arginine+acetylcholine was improved by 161+/-27% with UDCA vs 83+/-22% with placebo (mean difference 91% [95% CI 35%, 147%], P=0.016). CONCLUSIONS: Six weeks' UDCA therapy improved endothelium-dependent nitric oxide-independent vasodilatation, which might maintain arterial flow in coronary heart disease patients under conditions of impaired nitric oxide production.

Quantitative comparison of angiographic characteristics of coronary artery disease in patients with noninsulin-dependent diabetes mellitus compared with matched nondiabetic control subjects.

The angiographic characteristics of coronary artery disease (CAD) in noninsulin-dependent diabetes mellitus (NIDDM) patients were studied by quantitative coronary angiography (QCA). Fifty-seven consecutive NIDDM patients undergoing clinically indicated elective coronary angiography and 57 nondiabetic coronary artery disease (CAD) patients were individually matched for sex, age, and body mass index. Technically adequate coronary angiograms, available for 55 subjects in each group, were analyzed with third-generation QCA software. To evaluate the anatomic severity and extent of CAD, several QCA-derived parameters were incorporated into indexes describing various per-patient features of CAD. These measures reflect CAD severity, extent, and overall "atheroma burden," and were calculated separately for different coronary segments (i.e., left main, proximal, mid, and distal segments), for the different coronary arterial territories (i.e., left main, left anterior descending, left circumflex, and right), and for the entire coronary tree. No significant differences were found between the NIDDM and nondiabetic groups (global severity index, 51 +/- 14 vs 54 +/- 13, p = NS; global extent index, 34 +/- 13 vs 32 +/- 12, p = NS; global atheroma burden index, 27 +/- 16 vs 24 +/- 12, p = NS). We also found no between-group differences in proximal, mid, or distal segments, in separate vessel territories, or in left ventricular function. Our data suggest that CAD patients, with and without NIDDM, who have similar symptoms at a given age, have similar severity and extent of CAD.

On neural blind separation with noise suppression and redundancy reduction.

Noise is an unavoidable factor in real sensor signals. We study how additive and convolutive noise can be reduced or even eliminated in the blind source separation (BSS) problem. Particular attention is paid to cases in which the number of sensors is larger than the number of sources. We propose various methods and associated adaptive learning algorithms for such an extended BSS problem. Performance and validity of the proposed approaches are demonstrated by extensive computer simulations.

Least-squares methods for blind source separation based on nonlinear PCA.

In standard blind source separation, one tries to extract unknown source signals from their instantaneous linear mixtures by using a minimum of a priori information. We have recently shown that certain nonlinear extensions of principal component type neural algorithms can be successfully applied to this problem. In this paper, we show that a nonlinear PCA criterion can be minimized using least-squares approaches, leading to computationally efficient and fast converging algorithms. Several versions of this approach are developed and studied, some of which can be regarded as neural learning algorithms. A connection to the nonlinear PCA subspace rule is also shown. Experimental results are given, showing that the least-squares methods usually converge clearly faster than stochastic gradient algorithms in blind separation problems.