Active Learning to Maximize Area Under the ROC Curve

In active learning, a machine learning algorithmis given an unlabeled set of examples U, and is allowed to request labels for a relatively small subset of U to use for training. The goal is then to judiciously choose which examples in U to have labeled in order to optimize some performance criterion, e.g. classification accuracy. We study how active learning affects AUC. We examine two existing algorithms from the literature and present our own active learning algorithms designed to maximize the AUC of the hypothesis. One of our algorithms was consistently the top performer, and Closest Sampling from the literature often came in second behind it. When good posterior probability estimates were available, our heuristics were by far the best.

Area-under-the-curve monitoring of prednisolone for dose optimization in a stable renal transplant population

Background: Renal transplant recipients were noted to appear cushingoid while on low doses of steroid as part of a triple therapy immunosuppression of cyclosporin A (CsA), prednisolone, and azathioprine. Methods: The study group comprised adult renal transplant recipients with stable graft function who had received their renal allograft a minimum of 1 year previously (43 studies undertaken in 22 men and 20 women) with median daily prednisone dose of 7 mg (range 3-10). The control group was healthy nontransplant subjects [median dose 10 mg (10-30)]. Prednisolone bioavailability was measured using a limited 6-hour area under the curve (AUC), with prednisolone measured using specific HPLC assay. Results: The median prednisolone AUC/mg dose for all transplant recipients was significantly greater than the control group by approximately 50% (316 nmol(.)h/L/mg prednisolone versus 218). AUC was significantly higher in female recipients (median 415 versus 297 for men) and in recipients receiving cyclospotin (348 versus 285). The highest AUC was in women on estrogen supplements who were receiving cyclosporin (median 595). A significantly higher proportion of patients on triple therapy had steroid side effects compared with those on steroid and azathioprine (17/27 versus 4/15), more women than men had side effects (14/16 versus 7/22), and the AUC/mg prednisone was greater in those with side effects than without (median 377 versus 288 nmol-h/L/mg). Discussion: The results are consistent with the hypothesis that CsA increases the bioavailability of prednisolone, most likely through inhibition of beta-glycoprotein. The increased exposure to steroid increased the side-effect profile of steroids in the majority of patients. Because the major contributor to AUC is the maximum postdose concentration, it may be possible to use single-point monitoring (2 hours postdose) for routine clinical studies.

Ecological aspects of the sandfly fauna (Diptera, Psychodidae) in an American cutaneous leishmaniasis endemic area under the influence of hydroelectric plants in Paranapanema river, State of Paraná, Brazil

INTRODUCTION: An epidemiological study was undertaken to identify determinant factors in the occurrence of American cutaneous leishmaniasis in areas under the influence of hydroelectric plants in Paranapanema river, State of Paraná, Brazil. The ecological aspects of the phlebotomine fauna were investigated. METHODS: Sandflies were sampled with automatic light traps from February 2004 to June 2006 at 25 sites in the urban and rural areas of Itambaracá, and in Porto Almeida and São Joaquim do Pontal. RESULTS: A total of 3,187 sandflies of 15 species were captured. Nyssomyia neivai predominated (34.4%), followed by Pintomyia pessoai (32.6%), Migonemyia migonei (11.6%), Nyssomyia whitmani (8.8%), and Pintomyia fischeri (2.7%), all implicated in the transmission of Leishmania. Males predominated for Ny. neivai, and females for the other vector species, with significant statistical differences (p < 0.001). Nyssomyia neivai, Pi. pessoai, Ny. whitmani, Brumptomyia brumpti, Mg. migonei, and Pi. fischeri presented the highest values for the Standardized Species Abundance Index (SSAI). The highest frequencies and diversities were found in the preserved forest in Porto Almeida, followed by forests with degradation in São Joaquim do Pontal and Vila Rural. CONCLUSIONS: Sandflies were captured in all localities, with the five vectors predominating. Ny. neivai had its highest frequencies in nearby peridomestic environments and Pi. pessoai in areas of preserved forests. The highest SSAI values of Ny. neivai and Pi. pessoai reflect their wider dispersion and higher frequencies compared with other species, which seems to indicate that these two species may be transmitting leishmaniasis in the area.

Spatial and temporal variability of soil CO2 emission in a sugarcane area under green and slash-and-burn managements

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Measurements of labile Cd, Cu, Ni, Pb, and Zn levels at a northeastern Brazilian coastal area under the influence of oil production with diffusive gradients in thin films technique (DGT)

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Evaluation of single implants placed in the posterior mandibular area under immediate loading: a prospective study

The aim of this study was to evaluate the survival of single dental implants subjected to immediate function. Twelve patients with edentulous areas in the posterior mandible were included in the study. All received at least one regular platform dental implant (3.75mm×11mm or 3.75mm×13mm). Clinical and radiographic parameters were evaluated. The survival rate after 12 months was 83.3%. The implants showed no clinical mobility, had implant stability quotient values (ISQ; Osstell) around 70, bone loss of up to 2mm, and a probing depth of ≤3mm. Although the posterior mandible is an area in which the immediate loading of dental implants should be performed with caution, this treatment presented a good success rate in the present study sample.

Ecological aspects of the sandfly fauna (Diptera, Psychodidae) in an American cutaneous leishmaniasis endemic area under the influence of hydroelectric plants in Paranapanema river, State of Parana, Brazil

Introduction: An epidemiological study was undertaken to identify determinant factors in the occurrence of American cutaneous leishmaniasis in areas under the influence of hydroelectric plants in Paranapanema river, State of Parana, Brazil. The ecological aspects of the phlebotomine fauna were investigated. Methods: Sandflies were sampled with automatic light traps from February 2004 to June 2006 at 25 sites in the urban and rural areas of Itambaraca, and in Porto Almeida and Sao Joaquim do Pontal. Results: A total of 3,187 sandflies of 15 species were captured. Nyssomyia neivai predominated (34.4%), followed by Pintomyia pessoai (32.6%), Migonemyia migonei (11.6%), Nyssomyia whitmani (8.8%), and Pintomyia fischeri (2.7%), all implicated in the transmission of Leishmania. Males predominated for Ny. neivai, and females for the other vector species, with significant statistical differences (p < 0.001). Nyssomyia neivai, Pi. pessoai, Ny. whitmani, Brumptomyia brumpti, Mg. migonei, and Pi. fischeri presented the highest values for the Standardized Species Abundance Index (SSAI). The highest frequencies and diversities were found in the preserved forest in Porto Almeida, followed by forests with degradation in Sao Joaquim do Pontal and Vila Rural. Conclusions: Sandflies were captured in all localities, with the five vectors predominating. Ny. neivai had its highest frequencies in nearby peridomestic environments and Pi. pessoai in areas of preserved forests. The highest SSAI values of Ny. neivai and Pi. pessoai reflect their wider dispersion and higher frequencies compared with other species, which seems to indicate that these two species may be transmitting leishmaniasis in the area.

Ecological aspects of the sandfly fauna (Diptera, Psychodidae) in an American cutaneous leishmaniasis endemic area under the influence of hydroelectric plants in Paranapanema river, State of Paraná, Brazil

INTRODUCTION: An epidemiological study was undertaken to identify determinant factors in the occurrence of American cutaneous leishmaniasis in areas under the influence of hydroelectric plants in Paranapanema river, State of Paraná, Brazil. The ecological aspects of the phlebotomine fauna were investigated. METHODS: Sandflies were sampled with automatic light traps from February 2004 to June 2006 at 25 sites in the urban and rural areas of Itambaracá, and in Porto Almeida and São Joaquim do Pontal. RESULTS: A total of 3,187 sandflies of 15 species were captured. Nyssomyia neivai predominated (34.4%), followed by Pintomyia pessoai (32.6%), Migonemyia migonei (11.6%), Nyssomyia whitmani (8.8%), and Pintomyia fischeri (2.7%), all implicated in the transmission of Leishmania. Males predominated for Ny. neivai, and females for the other vector species, with significant statistical differences (p < 0.001). Nyssomyia neivai, Pi. pessoai, Ny. whitmani, Brumptomyia brumpti, Mg. migonei, and Pi. fischeri presented the highest values for the Standardized Species Abundance Index (SSAI). The highest frequencies and diversities were found in the preserved forest in Porto Almeida, followed by forests with degradation in São Joaquim do Pontal and Vila Rural. CONCLUSIONS: Sandflies were captured in all localities, with the five vectors predominating. Ny. neivai had its highest frequencies in nearby peridomestic environments and Pi. pessoai in areas of preserved forests. The highest SSAI values of Ny. neivai and Pi. pessoai reflect their wider dispersion and higher frequencies compared with other species, which seems to indicate that these two species may be transmitting leishmaniasis in the area.

The limitations of policy coordination in the euro area under the European Semester. Bruegel Policy Contribution 2015/19, November 2015

The European Semester is a yearly process of the European Union to improve economic policy coordination and ensure the implementation of the EU’s economic rules. Each Semester concludes with recommendations for the euro area as a whole and for each EU member state. We show that implementation of recommendations was poor at the beginning of the Semester in 2011, and has deteriorated since. The European Semester is not particularly effective at enforcing even the EU’s fiscal and macroeconomic imbalance rules. We find that euro-area recommendations with tangible economic goals are not well reflected in the recommendations issued to member states. Finally, we review various proposals to improve the efficiency of the European Semester and conclude that while certain steps could be helpful, policy coordination will likely continue to have major limitations.

Economic Policy Coordination in the Euro Area under the European Semester. CEPS Special Report No. 123, December 2015

The implementation record of the Country Specific Recommendations (CSRs) has declined over time, as financial turbulence lessened and the economic outlook started to improve. Urgency for reforms seemingly receded to leave room to request member states to move towards more accommodative stances. It is mainly the small countries that implement, at least partially, the recommendations addressed to them. Unfortunately, there is little that the EU can do to change the status quo. Yet, the President of the Eurogroup could be held accountable for the implementation of the recommendations addressed to the euro area. The creation of National Competitiveness Boards risks making the European Semester even more complex and likely to have little impact in the countries that need them most, namely large countries and those with poor governance. To make it effective, a procedure would be needed to make national wage norms consistent at the euro-area level, which may be a very difficult objective to achieve.

Advancing Bag-of-visual-words Representations For Lesion Classification In Retinal Images.

Diabetic Retinopathy (DR) is a complication of diabetes that can lead to blindness if not readily discovered. Automated screening algorithms have the potential to improve identification of patients who need further medical attention. However, the identification of lesions must be accurate to be useful for clinical application. The bag-of-visual-words (BoVW) algorithm employs a maximum-margin classifier in a flexible framework that is able to detect the most common DR-related lesions such as microaneurysms, cotton-wool spots and hard exudates. BoVW allows to bypass the need for pre- and post-processing of the retinographic images, as well as the need of specific ad hoc techniques for identification of each type of lesion. An extensive evaluation of the BoVW model, using three large retinograph datasets (DR1, DR2 and Messidor) with different resolution and collected by different healthcare personnel, was performed. The results demonstrate that the BoVW classification approach can identify different lesions within an image without having to utilize different algorithms for each lesion reducing processing time and providing a more flexible diagnostic system. Our BoVW scheme is based on sparse low-level feature detection with a Speeded-Up Robust Features (SURF) local descriptor, and mid-level features based on semi-soft coding with max pooling. The best BoVW representation for retinal image classification was an area under the receiver operating characteristic curve (AUC-ROC) of 97.8% (exudates) and 93.5% (red lesions), applying a cross-dataset validation protocol. To assess the accuracy for detecting cases that require referral within one year, the sparse extraction technique associated with semi-soft coding and max pooling obtained an AUC of 94.2 ± 2.0%, outperforming current methods. Those results indicate that, for retinal image classification tasks in clinical practice, BoVW is equal and, in some instances, surpasses results obtained using dense detection (widely believed to be the best choice in many vision problems) for the low-level descriptors.

Evaluation Of Soft-tissue Lesions With (18)f-fdg Pet/ct: Initial Results Of A Prospective Trial.

Although MRI is utilized for planning the resection of soft-tissue tumors, it is not always capable of differentiating benign from malignant lesions. The risk of local recurrence of soft-tissue sarcomas is increased when biopsies are performed before resection and by inadequate resections. PET associated with computed tomography using fluorodeoxyglucose labeled with fluorine-18 ((18)F-FDG PET/CT) may help differentiate between benign and malignant tumors, thus avoiding inadequate resections and making prior biopsies unnecessary. The purpose of this study was to evaluate the usefulness of (18)F-FDG PET/CT in differentiating benign from malignant solid soft-tissue lesions. Patients with solid lesions of the limbs or abdominal wall detected by MRI were submitted to (18)F-FDG PET/CT. The maximum standardized uptake value (SUVmax) cutoff was determined to differentiate malignant from benign tumors. Regardless of the (18)F-FDG PET/CT results all patients underwent biopsy and surgery. MRI was performed in 54 patients, and 10 patients were excluded because of purely lipomatose or cystic lesions. (18)F-FDG PET/CT was performed in the remaining 44 patients. Histopathology revealed 26 (59%) benign and 18 (41%) malignant soft-tissue lesions. A significant difference in SUVmax was observed between benign and malignant soft-tissue lesions. The SUVmax cutoff of 3.0 differentiated malignant from benign lesions with 100% sensitivity, 83.3% specificity, 89.6% accuracy, 78.3% positive predictive value, and 100% negative predictive value. (18)F-FDG PET/CT seems to be able to differentiate benign from malignant soft-tissue lesions with good accuracy and very high negative predictive value. Incorporating (18)F-FDG PET/CT into the diagnostic algorithm of these patients may prevent inadequate resections and unnecessary biopsies.

Evaluation Of Red Cell And Reticulocyte Parameters As Indicative Of Iron Deficiency In Patients With Anemia Of Chronic Disease.

The purpose of this study was to evaluate the effectiveness of mature red cell and reticulocyte parameters under three conditions: iron deficiency anemia, anemia of chronic disease, and anemia of chronic disease associated with absolute iron deficiency. Peripheral blood cells from 117 adult patients with anemia were classified according to iron status, and inflammatory activity, and the results of a hemoglobinopathy investigation as: iron deficiency anemia (n=42), anemia of chronic disease (n=28), anemia of chronic disease associated with iron deficiency anemia (n=22), and heterozygous β thalassemia (n=25). The percentage of microcytic red cells, hypochromic red cells, and levels of hemoglobin content in both reticulocytes and mature red cells were determined. Receiver operating characteristic analysis was used to evaluate the accuracy of the parameters in differentiating between the different types of anemia. There was no significant difference between the iron deficient group and anemia of chronic disease associated with absolute iron deficiency in respect to any parameter. The percentage of hypochromic red cells was the best parameter to discriminate anemia of chronic disease with and without absolute iron deficiency (area under curve=0.785; 95% confidence interval: 0.661-0.909, with sensitivity of 72.7%, and specificity of 70.4%; cut-off value 1.8%). The formula microcytic red cells minus hypochromic red cells was very accurate in differentiating iron deficiency anemia and heterozygous β thalassemia (area under curve=0.977; 95% confidence interval: 0.950-1.005; with sensitivity of 96.2%, and specificity of 92.7%; cut-off value 13.8). The indices related to red cells and reticulocytes have a moderate performance in identifying absolute iron deficiency in patients with anemia of chronic disease.