University of Tübingen

About: University of Tübingen is a education organization based out in Tübingen, Germany. It is known for research contribution in the topics: Population & Immune system. The organization has 40555 authors who have published 84108 publications receiving 3015320 citations. The organization is also known as: Eberhard Karls University & Eberhard-Karls-Universität Tübingen.

Simultaneous PET-MRI: a new approach for functional and morphological imaging

Abstract: Noninvasive imaging at the molecular level is an emerging field in biomedical research. This paper introduces a new technology synergizing two leading imaging methodologies: positron emission tomography (PET) and magnetic resonance imaging (MRI). Although the value of PET lies in its high-sensitivity tracking of biomarkers in vivo, it lacks resolving morphology. MRI has lower sensitivity, but produces high soft-tissue contrast and provides spectroscopic information and functional MRI (fMRI). We have developed a three-dimensional animal PET scanner that is built into a 7-T MRI. Our evaluations show that both modalities preserve their functionality, even when operated isochronously. With this combined imaging system, we simultaneously acquired functional and morphological PET-MRI data from living mice. PET-MRI provides a powerful tool for studying biology and pathology in preclinical research and has great potential for clinical applications. Combining fMRI and spectroscopy with PET paves the way for a new perspective in molecular imaging.

Adversarial machine learning

Abstract: In this paper (expanded from an invited talk at AISEC 2010), we discuss an emerging field of study: adversarial machine learning---the study of effective machine learning techniques against an adversarial opponent. In this paper, we: give a taxonomy for classifying attacks against online machine learning algorithms; discuss application-specific factors that limit an adversary's capabilities; introduce two models for modeling an adversary's capabilities; explore the limits of an adversary's knowledge about the algorithm, feature space, training, and input data; explore vulnerabilities in machine learning algorithms; discuss countermeasures against attacks; introduce the evasion challenge; and discuss privacy-preserving learning techniques.

Temozolomide chemotherapy alone versus radiotherapy alone for malignant astrocytoma in the elderly: the NOA-08 randomised, phase 3 trial

Abstract: Summary Background Radiotherapy is the standard care in elderly patients with malignant astrocytoma and the role of primary chemotherapy is poorly defined. We did a randomised trial to compare the efficacy and safety of dose-dense temozolomide alone versus radiotherapy alone in elderly patients with anaplastic astrocytoma or glioblastoma. Methods Between May 15, 2005, and Nov 2, 2009, we enrolled patients with confirmed anaplastic astrocytoma or glioblastoma, age older than 65 years, and a Karnofsky performance score of 60 or higher. Patients were randomly assigned 100 mg/m 2 temozolomide, given on days 1–7 of 1 week on, 1 week off cycles, or radiotherapy of 60·0 Gy, administered over 6–7 weeks in 30 fractions of 1·8–2·0 Gy. The primary endpoint was overall survival. We assessed non-inferiority with a 25% margin, analysed for all patients who received at least one dose of assigned treatment. This trial is registered with ClinicalTrials.gov, number NCT01502241. Findings Of 584 patients screened, we enrolled 412. 373 patients (195 randomly allocated to the temozolomide group and 178 to the radiotherapy group) received at least one dose of treatment and were included in efficacy analyses. Median overall survival was 8·6 months (95% CI 7·3–10·2) in the temozolomide group versus 9·6 months (8·2–10·8) in the radiotherapy group (hazard ratio [HR] 1·09, 95% CI 0·84–1·42, p non-inferiority =0·033). Median event-free survival (EFS) did not differ significantly between the temozolomide and radiotherapy groups (3·3 months [95% CI 3·2–4·1] vs 4·7 [4·2–5·2]; HR 1·15, 95% CI 0·92–1·43, p non-inferiority =0·043). Tumour MGMT promoter methylation was seen in 73 (35%) of 209 patients tested. MGMT promoter methylation was associated with longer overall survival than was unmethylated status (11·9 months [95% CI 9·0 to not reached] vs 8·2 months [7·0–10·0]; HR 0·62, 95% CI 0·42–0·91, p=0·014). EFS was longer in patients with MGMT promoter methylation who received temozolomide than in those who underwent radiotherapy (8·4 months [95e% CI 5·5–11·7] vs 4·6 [4·2–5·0]), whereas the opposite was true for patients with no methylation of the MGMT promoter (3·3 months [3·0–3·5] vs 4·6 months [3·7–6·3]). The most frequent grade 3–4 intervention-related adverse events were neutropenia (16 patients in the temozolomide group vs two in the radiotherapy group), lymphocytopenia (46 vs one), thrombocytopenia (14 vs four), raised liver-enzyme concentrations (30 vs 16), infections (35 vs 23), and thromboembolic events (24 vs eight). Interpretation Temozolomide alone is non-inferior to radiotherapy alone in the treatment of elderly patients with malignant astrocytoma. MGMT promoter methylation seems to be a useful biomarker for outcomes by treatment and could aid decision-making. Funding Merck Sharp & Dohme.

The normal distributions transform: a new approach to laser scan matching

Abstract: Matching 2D range scans is a basic component of many localization and mapping algorithms. Most scan match algorithms require finding correspondences between the used features, i.e. points or lines. We propose an alternative representation for a range scan, the normal distributions transform. Similar to an occupancy grid, we subdivide the 2D plane into cells. To each cell, we assign a normal distribution, which locally models the probability of measuring a point. The result of the transform is a piecewise continuous and differentiable probability density, that can be used to match another scan using Newton's algorithm. Thereby, no explicit correspondences have to be established. We present the algorithm in detail and show the application to relative position tracking and simultaneous localization and map building (SLAM). First results on real data demonstrate, that the algorithm is capable to map unmodified indoor environments reliable and in real time, even without using odometry data.

Autosomal dominant cerebellar ataxias: clinical features, genetics, and pathogenesis.

Abstract: Summary Autosomal dominant cerebellar ataxias are hereditary neurodegenerative disorders that are known as spinocerebellar ataxias (SCA) in genetic nomenclature. In the pregenomic era, ataxias were some of the most poorly understood neurological disorders; the unravelling of their molecular basis enabled precise diagnosis in vivo and explained many clinical phenomena such as anticipation and variable phenotypes even within one family. However, the discovery of many ataxia genes and loci in the past decade threatens to cause more confusion than optimism among clinicians. Therefore, the provision of guidance for genetic testing according to clinical findings and frequencies of SCA subtypes in different ethnic groups is a major challenge. The identification of ataxia genes raises hope that essential pathogenetic mechanisms causing SCA will become more and more apparent. Elucidation of the pathogenesis of SCA hopefully will enable the development of rational therapies for this group of disorders, which currently can only be treated symptomatically.