Environmental data mining and modeling based on machine learning algorithms and geostatistics

The paper presents some contemporary approaches to spatial environmental data analysis. The main topics are concentrated on the decision-oriented problems of environmental spatial data mining and modeling: valorization and representativity of data with the help of exploratory data analysis, spatial predictions, probabilistic and risk mapping, development and application of conditional stochastic simulation models. The innovative part of the paper presents integrated/hybrid model-machine learning (ML) residuals sequential simulations-MLRSS. The models are based on multilayer perceptron and support vector regression ML algorithms used for modeling long-range spatial trends and sequential simulations of the residuals. NIL algorithms deliver non-linear solution for the spatial non-stationary problems, which are difficult for geostatistical approach. Geostatistical tools (variography) are used to characterize performance of ML algorithms, by analyzing quality and quantity of the spatially structured information extracted from data with ML algorithms. Sequential simulations provide efficient assessment of uncertainty and spatial variability. Case study from the Chernobyl fallouts illustrates the performance of the proposed model. It is shown that probability mapping, provided by the combination of ML data driven and geostatistical model based approaches, can be efficiently used in decision-making process. (C) 2003 Elsevier Ltd. All rights reserved.

Performance of noninvasive ventilation algorithms on ICU ventilators during pressure support: a clinical study.

To evaluate the impact of noninvasive ventilation (NIV) algorithms available on intensive care unit ventilators on the incidence of patient-ventilator asynchrony in patients receiving NIV for acute respiratory failure. Prospective multicenter randomized cross-over study. Intensive care units in three university hospitals. Patients consecutively admitted to the ICU and treated by NIV with an ICU ventilator were included. Airway pressure, flow and surface diaphragmatic electromyography were recorded continuously during two 30-min periods, with the NIV (NIV+) or without the NIV algorithm (NIV0). Asynchrony events, the asynchrony index (AI) and a specific asynchrony index influenced by leaks (AIleaks) were determined from tracing analysis. Sixty-five patients were included. With and without the NIV algorithm, respectively, auto-triggering was present in 14 (22%) and 10 (15%) patients, ineffective breaths in 15 (23%) and 5 (8%) (p = 0.004), late cycling in 11 (17%) and 5 (8%) (p = 0.003), premature cycling in 22 (34%) and 21 (32%), and double triggering in 3 (5%) and 6 (9%). The mean number of asynchronies influenced by leaks was significantly reduced by the NIV algorithm (p < 0.05). A significant correlation was found between the magnitude of leaks and AIleaks when the NIV algorithm was not activated (p = 0.03). The global AI remained unchanged, mainly because on some ventilators with the NIV algorithm premature cycling occurs. In acute respiratory failure, NIV algorithms provided by ICU ventilators can reduce the incidence of asynchronies because of leaks, thus confirming bench test results, but some of these algorithms can generate premature cycling.

Use of anisotropic modelling in electrical impedance tomography: description of method and preliminary assessment of utility in imaging brain function in the adult human head.

Electrical Impedance Tomography (EIT) is an imaging method which enables a volume conductivity map of a subject to be produced from multiple impedance measurements. It has the potential to become a portable non-invasive imaging technique of particular use in imaging brain function. Accurate numerical forward models may be used to improve image reconstruction but, until now, have employed an assumption of isotropic tissue conductivity. This may be expected to introduce inaccuracy, as body tissues, especially those such as white matter and the skull in head imaging, are highly anisotropic. The purpose of this study was, for the first time, to develop a method for incorporating anisotropy in a forward numerical model for EIT of the head and assess the resulting improvement in image quality in the case of linear reconstruction of one example of the human head. A realistic Finite Element Model (FEM) of an adult human head with segments for the scalp, skull, CSF, and brain was produced from a structural MRI. Anisotropy of the brain was estimated from a diffusion tensor-MRI of the same subject and anisotropy of the skull was approximated from the structural information. A method for incorporation of anisotropy in the forward model and its use in image reconstruction was produced. The improvement in reconstructed image quality was assessed in computer simulation by producing forward data, and then linear reconstruction using a sensitivity matrix approach. The mean boundary data difference between anisotropic and isotropic forward models for a reference conductivity was 50%. Use of the correct anisotropic FEM in image reconstruction, as opposed to an isotropic one, corrected an error of 24 mm in imaging a 10% conductivity decrease located in the hippocampus, improved localisation for conductivity changes deep in the brain and due to epilepsy by 4-17 mm, and, overall, led to a substantial improvement on image quality. This suggests that incorporation of anisotropy in numerical models used for image reconstruction is likely to improve EIT image quality.

Tuberculose miliaire chez une personne âgée: diagnostic inattendu à l'autopsie [Miliary tuberculosis: unexpected autopsy finding in an elderly person].

Miliary tuberculosis is a rare disease that is difficult to diagnose because of its non-specific presentation. It should be suspected in elderly patients who complaint of failure to thrive, unexplained fatigue and weight loss. Using a clinical situation where the diagnosis was made only at autopsy, we briefly review the epidemiology of miliary tuberculosis and propose recommendations for the diagnosis and the prophylaxis of latent tuberculosis. Finally, we discuss criteria to perform epidemiological investigations among close contacts in this situation.

New diagnostic real-time PCR for specific detection of Parachlamydia acanthamoebae DNA in clinical samples

Given the low sensitivity of amoebal coculture, we developed a specific real-time PCR for the detection of Parachlamydia. The analytical sensitivity was high, and the inter- and intrarun variabilities were low. When the PCR was applied to nasopharyngeal aspirates, it was positive for six patients with bronchiolitis. Future studies should assess the role of Parachlamydia in bronchiolitis.

Diagnostic and prognostic value of cerebral 31P magnetic resonance spectroscopy in neonates with perinatal asphyxia.

The impact of depressed neonatal cerebral oxidative phosphorylation for diagnosing the severity of perinatal asphyxia was estimated by correlating the concentrations of phosphocreatine (PCr) and ATP as determined by magnetic resonance spectroscopy with the degree of hypoxic-ischemic encephalopathy (HIE) in 23 asphyxiated term neonates. Ten healthy age-matched neonates served as controls. In patients, the mean concentrations +/- SD of PCr and ATP were 0.99 +/- 0.46 mmol/L (1.6 +/- 0.2 mmol/L) and 0.99 +/- 0.35 mmol/L (1.7 +/- 0.2 mmol/L), respectively (normal values in parentheses). [PCr] and [ATP] correlated significantly with the severity of HIE (r = 0.85 and 0.9, respectively, p < 0.001), indicating that the neonatal encephalopathy is the clinical manifestation of a marred brain energy metabolism. Neurodevelopmental outcome was evaluated in 21 children at 3, 9, and 18 mo. Seven infants had multiple impairments, five were moderately handicapped, five had only mild symptoms, and four were normal. There was a significant correlation between the cerebral concentrations of PCr or ATP at birth and outcome (r = 0.8, p < 0.001) and between the degree of neonatal neurologic depression and outcome (r = 0.7). More important, the outcome of neonates with moderate HIE could better be predicted with information from quantitative 31P magnetic resonance spectroscopy than from neurologic examinations. In general, the accuracy of outcome predictability could significantly be increased by adding results from 31P magnetic resonance spectroscopy to the neonatal neurologic score, but not vice versa. No correlation with outcome was found for other perinatal risk factors, including Apgar score.

Diagnostic accuracy of body mass index, body fat and waist in defining low aerobic fitness in adolescents

Objective: to assess the diagnostic accuracy of different anthropometric markers in defining low aerobic fitness among adolescents. Methods: cross-sectional study on 2,331 boys and 2,366 girls aged 10 - 18 years. Body mass index (BMI) was measured using standardized methods; body fat (BF) was assessed by bioelectrical impedance. Low aerobic fitness was assessed by the 20-meter shuttle run using the FITNESSGRAMR criteria. Waist was measured in a subsample of 1,933 boys and 1,897 girls. Overweight, obesity and excess fat were defined according to the International Obesity Task Force (IOTF) or FITNESSGRAMR criteria. Results: 38.5% of boys and 46.5% of girls were considered as unfit according to the FITNESSGRAMR criteria. In boys, the area under the ROC curve (AUC) and 95% confidence interval were 66.7 (64.1 - 69.3), 67.1 (64.5 - 69.6) and 64.6 (61.9 - 67.2) for BMI, BF and waist, respectively (P<0.02). In girls, the values were 68.3 (65.9 - 70.8), 63.8 (61.3 - 66.3) and 65.9 (63.4 - 68.4), respectively (P<0.001). In boys, the sensitivity and specificity to diagnose low fitness were 13% and 99% for obesity (IOTF); 38% and 86% for overweight + obesity (IOTF); 28% and 94% for obesity (FITNESSGRAMR) and 42% and 81% for excess fat (FITNESSGRAMR). For girls, the values were 9% and 99% for obesity (IOTF); 33% and 82% for overweight + obesity (IOTF); 22% and 94% for obesity (FITNESSGRAMR) and 26% and 90% for excess fat (FITNESSGRAMR). Conclusions: BMI, not body fat or waist, should be used to define low aerobic fitness. The IOTF BMI cut-points to define obesity have a very low screening capacity and should not be used.

L'immunoscintigraphie par les anticorps monoclonaux radiomarqués. Evolution d'une méthode de diagnostic du cancer et espoir d'une nouvelle forme de thérapie [Immunoscintigraphy using radio-labeled monoclonal antibodies. Development of a method of cancer diagnosis and hopes for a new form of therapy].

Personal results are presented to illustrate the development of immunoscintigraphy for the detection of cancer over the last 12 years, from the early experimental results in nude mice grafted with human colon carcinoma to the most modern form of immunoscintigraphy applied to patients, using I123 labeled Fab fragments from monoclonal anti-CEA antibodies detected by single photon emission computerized tomography (SPECT). The first generation of immunoscintigraphy used I131 labeled, immunoadsorbent purified, polyclonal anti-CEA antibodies and planar scintigraphy, as the detection system. The second generation used I131 labeled monoclonal anti-CEA antibodies and SPECT, while the third generation employed I123 labeled fragments of monoclonal antibodies and SPECT. The improvement in the precision of tumor images with the most recent forms of immunoscintigraphy is obvious. However, we think the usefulness of immunoscintigraphy for routine cancer management has not yet been entirely demonstrated. Further prospective trials are still necessary to determine the precise clinical role of immunoscintigraphy. A case report is presented on a patient with two liver metastases from a sigmoid carcinoma, who received through the hepatic artery a therapeutic dose (100 mCi) of I131 coupled to 40 mg of a mixture of two high affinity anti-CEA monoclonal antibodies. Excellent localisation in the metastases of the I131 labeled antibodies was demonstrated by SPECT and the treatment was well tolerated. The irradiation dose to the tumor, however, was too low at 4300 rads (with 1075 rads to the normal liver and 88 rads to the bone marrow), and no evidence of tumor regression was obtained. Different approaches for increasing the irradiation dose delivered to the tumor by the antibodies are considered.

The diagnostic challenge of progressive pseudorheumatoid dysplasia (PPRD): A review of clinical features, radiographic features, and WISP3 mutations in 63 affected individuals.

Progressive pseudorheumatoid dysplasia (PPRD) is a genetic, non-inflammatory arthropathy caused by recessive loss of function mutations in WISP3 (Wnt1-inducible signaling pathway protein 3; MIM 603400), encoding for a signaling protein. The disease is clinically silent at birth and in infancy. It manifests between the age of 3 and 6 years with joint pain and progressive joint stiffness. Affected children are referred to pediatric rheumatologists and orthopedic surgeons; however, signs of inflammation are absent and anti-inflammatory treatment is of little help. Bony enlargement at the interphalangeal joints progresses leading to camptodactyly. Spine involvement develops in late childhood and adolescence leading to short trunk with thoracolumbar kyphosis. Adult height is usually below the 3rd percentile. Radiographic signs are relatively mild. Platyspondyly develops in late childhood and can be the first clue to the diagnosis. Enlargement of the phalangeal metaphyses develops subtly and is usually recognizable by 10 years. The femoral heads are large and the acetabulum forms a distinct "lip" overriding the femoral head. There is a progressive narrowing of all articular spaces as articular cartilage is lost. Medical management of PPRD remains symptomatic and relies on pain medication. Hip joint replacement surgery in early adulthood is effective in reducing pain and maintaining mobility and can be recommended. Subsequent knee joint replacement is a further option. Mutation analysis of WISP3 allowed the confirmation of the diagnosis in 63 out of 64 typical cases in our series. Intronic mutations in WISP3 leading to splicing aberrations can be detected only in cDNA from fibroblasts and therefore a skin biopsy is indicated when genomic analysis fails to reveal mutations in individuals with otherwise typical signs and symptoms. In spite of the first symptoms appearing in early childhood, the diagnosis of PPRD is most often made only in the second decade and affected children often receive unnecessary anti-inflammatory and immunosuppressive treatments. Increasing awareness of PPRD appears to be essential to allow for a timely diagnosis. © 2012 Wiley Periodicals, Inc.