Search

The aim of this study was to properly define the information regarding patient exposure to Ionizing Radiations in the radiological report, according to the European Directive 2013/59/EURATOM (EU 2013/59 art.58(b)). For this purpose, we evaluated the results from other Member States EU 2013/59 transpositions and from Guidelines recommendation published by International Organizations involved in diagnostic radiology. A practical way for implementing art.58 is also traced. Dosimetric quantities, such as exposure, absorbed dose and effective dose which may be included in radiological report, were first analyzed; then, in order to define international state of art of Member States EU 2013/59 transposition, a Web research using French, English, Spanish and German key words was performed. EU 2013/59 transposition for 5 Member States was reported. Especially regarding art.58, a European project reports that few European countries (11 of 28) have identified the dose metrics to be used in radiological report. Scientific organizations supporting clinical radiologists and medical physicists have published Guidelines reporting parameters useful to quantify the radiation output and to assess patient dose. Our research revealed that there is not a shared interpretation of patient exposure information to be included in radiological report. Nevertheless, according to scientific community, authors believe that the exposure is the most appropriate information that could be included in radiological report. Alternatively, but with more expensiveness, a risk index based on effective dose could be used. Moreover, the systematic exposure information recorded could be useful for dose estimates of population from medical exposure.

The aim of this article was to explore the implications of the new Euratom dose limit for occupational radiation protection in the context of medical occupational radiation exposures. The European Directive 2013/59/Euratom takes into account the new recommendations on reduction in the dose limit for the lens of the eye for planned occupational exposures released in 2012 by the International Commission on Radiological Protection (ICRP 118). Different dose-monitoring procedures and devices were considered. Occupational eye lens doses reported by previous studies were analyzed, mainly considering workers involved in interventional procedures with X-rays. The current status of eye lens radiation protection and the main methods for dose reduction were investigated. The analysis showed that the workers, potentially exceeding the new limit, are clinical staff performing interventional procedures with a relatively high X-ray dose. Regarding radiological protection issues, the considered literature reports that the proper use of personal protective equipment may reduce the eye lens absorbed dose. The evaluation of the occupational eye lens dose is essential to establish which method of personal dose monitoring should be preferred. Furthermore, education and training about the right use of personal protective equipment are important for medical staff working with ionizing radiation.

The ICRP Publication 118 introduced new threshold doses for tissue reactions and other non cancer effects on lens:the new threshold doses resulted significantly lower than previous scientific evidences. Lenses are often involved in brain radiotherapy treatment volumes and the risk of cataratta becomes relevant. Our study compares the results obtained from treatment plans calculated with both Rapid Arc technique and a new SSW IMRT template, for each patient recruited in the study, in terms of lens maximum absorbed doses. 10 patients with brain tumours diagnosis (PTV medium size af about 90 cc) localized in different areas were treated with the new SSW IMRT template, that consists in five fields spreading over a The dose prescription for all patients was 60 Gy in 2 Gy daily fractions. On the same patients, Rapid Arc plans were conformed with equal optimization objectives. The target coverage was at least 97% of prescription dose to 95 % of volume in both calculated plans. All the patiens treated with the SSW IMRT technique received a lower maximum omolateral lens dose (range from 0.4 to 2.4 Gy; mean value 1.7 Gy) compared to doses obtained with Rapid Arc plans (range from 0.5 to 4.4 Gy; mean value 2.4 Gy). Relatively to contralateral dose lens, we obtained a mean value of max dose of 1.5 Gy (range 0.4–2.2 Gy) for SSW IMRT plans and 2.2 Gy (range 0.5–4.6 Gy) for Rapid Arc plans. Patients treated with the new SSW IMRT template showed a lens dose sparing of about the 28% compared to doses they would have received with Rapid Arc plans.The other OAR involved as chiasm, optical nerve and brainstem are widely lower than the corresponding constraints. The new SSW IMRT template could be applied to treat target with different size, shape and anatomical position. Furthermore, the lens maximum dose reduction could potentially reduce the risk of cataract.

Medical image computing raises new challenges due to the scale and the complexity of the required analyses. Medical image databases are currently available to supply clinical diagnosis. For instance, it is possible to provide diagnostic information based on an imaging biomarker comparing a single case to the reference group (controls or patients with disease). At the same time many sophisticated and computationally intensive algorithms have been implemented to extract useful information from medical images. Many applications would take great advantage by using scientific workflow technology due to its design, rapid implementation and reuse. However this technology requires a distributed computing infrastructure (such as Grid or Cloud) to be executed efficiently. One of the most used workflow manager for medical image processing is the LONI pipeline (LP), a graphical workbench developed by the Laboratory of Neuro Imaging (http://pipeline.loni.usc.edu). In this article we present a general approach to submit and monitor workflows on distributed infrastructures using LONI Pipeline, including European Grid Infrastructure (EGI) and Torque-based batch farm. In this paper we implemented a complete segmentation pipeline in brain magnetic resonance imaging (MRI). It requires time-consuming and data-intensive processing and for which reducing the computing time is crucial to meet clinical practice constraints. The developed approach is based on web services and can be used for any medical imaging application.

In this study we present a novel fully automated Hippocampal Unified Multi-Atlas-Networks (HUMAN) algorithm for the segmentation of the hippocampus in structural magnetic resonance imaging. In multi-atlas approaches atlas selection is of crucial importance for the accuracy of the segmentation. Here we present an optimized method based on the definition of a small peri-hippocampal region to target the atlas learning with linear and non-linear embedded manifolds. All atlases were co-registered to a data driven template resulting in a computationally efficient method that requires only one test registration. The optimal atlases identified were used to train dedicated artificial neural networks whose labels were then propagated and fused to obtain the final segmentation. To quantify data heterogeneity and protocol inherent effects, HUMAN was tested on two independent data sets provided by the Alzheimer's Disease Neuroimaging Initiative and the Open Access Series of Imaging Studies. HUMAN is accurate and achieves state-of-the-art performance (Dice[Formula: see text] and Dice[Formula: see text]). It is also a robust method that remains stable when applied to the whole hippocampus or to sub-regions (patches). HUMAN also compares favorably with a basic multi-atlas approach and a benchmark segmentation tool such as FreeSurfer.

Neurodegenerative diseases are frequently associated with structural changes in the brain. Magnetic Resonance Imaging (MRI) scans can show these variations and therefore be used as a supportive feature for a number of neurodegenerative diseases. The hippocampus has been known to be a biomarker for Alzheimer disease and other neurological and psychiatric diseases. However, it requires accurate, robust and reproducible delineation of hippocampal structures. Fully automatic methods are usually the voxel based approach, for each voxel a number of local features were calculated. In this paper we compared four different techniques for feature selection from a set of 315 features extracted for each voxel: (i) filter method based on the Kolmogorov-Smirnov test; two wrapper methods, respectively, (ii) Sequential Forward Selection and (iii) Sequential Backward Elimination; and (iv) embedded method based on the Random Forest Classifier on a set of 10 T1-weighted brain MRIs and tested on an independent set of 25 subjects. The resulting segmentations were compared with manual reference labelling. By using only 23 features for each voxel (sequential backward elimination) we obtained comparable state of-the-art performances with respect to the standard tool FreeSurfer., Comment: To appear on "Computational and Mathematical Methods in Medicine", Hindawi Publishing Corporation. 19 pages, 7 figures

This study presents a fully automated algorithm for the segmentation of the hippocampus in structural Magnetic Resonance Imaging (MRI) and its deployment as a service on an open cloud infrastructure. Optimal atlases strategies for multi-atlas learning are combined with a voxel-wise classification approach. The method efficiency is optimized as training atlases are previously registered to a data driven template, accordingly for each test MRI scan only a registration is needed. The selected optimal atlases are used to train dedicated random forest classifiers whose labels are fused by majority voting. The method performances were tested on a set of 100 MRI scans provided by the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Leave-one-out results (Dice = \(0.910\,\pm \,0.004\)) show the presented method compares well with other state-of-the-art techniques and a benchmark segmentation tool as FreeSurfer. The proposed strategy significantly improves a standard multi-atlas approach (\(p < .001\)).