Search

The Breast Surgery theoretical and practical knowledge curriculum comprehensively describes the knowledge and skills expected of a fully trained surgeon practicing in the European Union and European Economic Area (EEA). It forms part of a range of factors that contribute to the delivery of high quality cancer care. It has been developed by a panel of experts from across Europe and has been validated by professional breast surgery societies in Europe. The curriculum maps closely to the syllabus of the Union of European Medical Specialists (UEMS) Breast Surgery Exam, the UK FRCS (breast specialist interest) curriculum and other professional standards across Europe and globally (USA Society of Surgical Oncology, SSO). It is envisioned that this will serve as the basis for breast surgery training, examination and accreditation across Europe to harmonise and raise standards as breast surgery develops as a separate discipline from its parent specialties (general surgery, gynaecology, surgical oncology and plastic surgery).\ud \ud \ud \ud The curriculum is not static but will be revised and updated by the curriculum development group of the European Breast Surgical Oncology Certification group (BRESO) every 2 years.

Background: Combining MRI techniques with machine learning methodology is rapidly gaining attention as a promising method for staging of brain gliomas. This study assesses the diagnostic value of such a framework applied to dynamic susceptibility contrast (DSC)-MRI in classifying treatment-naïve gliomas from a multi-center patients into WHO grades II-IV and across their isocitrate dehydrogenase (IDH) mutation status. Methods: Three hundred thirty-three patients from 6 tertiary centres, diagnosed histologically and molecularly with primary gliomas (IDH-mutant = 151 or IDH-wildtype = 182) were retrospectively identified. Raw DSC-MRI data was post-processed for normalised leakage-corrected relative cerebral blood volume (rCBV) maps. Shape, intensity distribution (histogram) and rotational invariant Haralick texture features over the tumour mask were extracted. Differences in extracted features across glioma grades and mutation status were tested using the Wilcoxon two-sample test. A random-forest algorithm was employed (2-fold cross-validation, 250 repeats) to predict grades or mutation status using the extracted features. Results: Shape, distribution and texture features showed significant differences across mutation status. WHO grade II-III differentiation was mostly driven by shape features while texture and intensity feature were more relevant for the III-IV separation. Increased number of features became significant when differentiating grades further apart from one another. Gliomas were correctly stratified by mutation status in 71% and by grade in 53% of the cases (87% of the gliomas grades predicted with distance less than 1). Conclusions: Despite large heterogeneity in the multi-center dataset, machine learning assisted DSC-MRI radiomics hold potential to address the inherent variability and presents a promising approach for non-invasive glioma molecular subtyping and grading. © 2020 The Author(s).

As the number of people diagnosed with Alzheimer's disease (AD) reaches epidemic proportions, there is an urgent need to develop effective treatment strategies to tackle the social and economic costs of this fatal condition. Dozens of candidate therapeutics are currently being tested in clinical trials, and compounds targeting the aberrant accumulation of tau proteins into neurofibrillary tangles (NFTs) are the focus of substantial current interest. Reliable, translatable biomarkers sensitive to both tau pathology and its modulation by treatment along with animal models that faithfully reflect aspects of the human disease are urgently required. Magnetic resonance imaging (MRI) is well established as a valuable tool for monitoring the structural brain changes that accompany AD progression. However the descent into dementia is not defined by macroscopic brain matter loss alone: non-invasive imaging measurements sensitive to protein accumulation, white matter integrity and cerebral haemodynamics probe distinct aspects of AD pathophysiology and may serve as superior biomarkers for assessing drug efficacy. Here we employ a multi-parametric array of five translatable MRI techniques to characterise the in vivo pathophysiological phenotype of the rTg4510 mouse model of tauopathy (structural imaging, diffusion tensor imaging (DTI), arterial spin labelling (ASL), chemical exchange saturation transfer (CEST) and glucose CEST). Tau-induced pathological changes included grey matter atrophy, increased radial diffusivity in the white matter, decreased amide proton transfer and hyperperfusion. We demonstrate that the above markers unambiguously discriminate between the transgenic group and age-matched controls and provide a comprehensive profile of the multifaceted neuropathological processes underlying the rTg4510 model. Furthermore, we show that ASL and DTI techniques offer heightened sensitivity to processes believed to precede detectable structural changes and, as such, provides a platform for the study of disease mechanisms and therapeutic intervention., Highlights • The rTg4510 mouse model exhibits specific tau pathology. • This enables dissection of the unique role of tau in Alzheimer's Disease. • We applied five quantitative, non-invasive MRI techniques to the rTg4510. • We demonstrate the sensitivity of each to discriminate regions of high tau burden. • This study provides a platform for the longitudinal assessment of tau therapies.

Identifying at what point atrophy rates first change in Alzheimer's disease is important for informing design of presymptomatic trials. Serial T1-weighed magnetic resonance imaging scans of 94 participants (28 noncarriers, 66 carriers) from the Dominantly Inherited Alzheimer Network were used to measure brain, ventricular, and hippocampal atrophy rates. For each structure, nonlinear mixed-effects models estimated the change-points when atrophy rates deviate from normal and the rates of change before and after this point. Atrophy increased after the change-point, which occurred 1-1.5 years (assuming a single step change in atrophy rate) or 3-8 years (assuming gradual acceleration of atrophy) before expected symptom onset. At expected symptom onset, estimated atrophy rates were at least 3.6 times than those before the change-point. Atrophy rates are pathologically increased up to seven years before "expected onset". During this period, atrophy rates may be useful for inclusion and tracking of disease progression.

International audience

Exudates in retinal images are one of the early signs of the vision-threatening diabetic retinopathy and diabetic macular edema. Early diagnosis is very helpful in preventing the progression of the disease. In this work, we propose a fully automatic exudate segmentation method based on the state-of-the-art residual learning framework. With our proposed end-to-end architecture the training is done on small patches, but at the test time, the full sized segmentation is obtained at once. The small number of exudates in the training set and the presence of other bright regions are the limiting factors, which are tackled by our proposed importance sampling approach. This technique selects the misleading normal patches with a higher priority, and at the same time avoids the network to overfit to those samples. Thus, no additional post-processing is needed. The method was evaluated on three public datasets for both detecting and segmenting the exudates and outperformed the state-of-the-art techniques.

Supervised learning is ubiquitous in medical image analysis. In this paper we consider the problem of meta-learning – predicting which methods will perform well in an unseen classification problem, given previous experience with other classification problems. We investigate the first step of such an approach: how to quantify the similarity of different classification problems. We characterize datasets sampled from six classification problems by performance ranks of simple classifiers, and define the similarity by the inverse of Euclidean distance in this meta-feature space. We visualize the similarities in a 2D space, where meaningful clusters start to emerge, and show that the proposed representation can be used to classify datasets according to their origin with 89.3% accuracy. These findings, together with the observations of recent trends in machine learning, suggest that meta-learning could be a valuable tool for the medical imaging community.