Behind the scenes of EQA - characteristics, capabilities, benefits and assets of external quality assessment (EQA).

This is the first in a series of five papers that detail the role and substantial impact that external quality assessment (EQA) and their providers' services play in ensuring diagnostic (IVD) performance quality. The aim is to give readers and users of EQA services an insight into the processes in EQA, explain to them what happens before EQA samples are delivered and after examination results are submitted to the provider, how they are assessed, what benefits participants can expect, but also who are stakeholders other than participants and what significance do EQA data and assessment results have for them. This first paper presents the history of EQA, insights into legal, financing and ethical matters, information technology used in EQA, structure and lifecycle of EQA programs, frequency and intensity of challenges, and unique requirements of extra-examination and educational EQA programs.

European intersocietal recommendations for the biomarker-based diagnosis of neurocognitive disorders.

The recent commercialisation of the first disease-modifying drugs for Alzheimer's disease emphasises the need for consensus recommendations on the rational use of biomarkers to diagnose people with suspected neurocognitive disorders in memory clinics. Most available recommendations and guidelines are either disease-centred or biomarker-centred. A European multidisciplinary taskforce consisting of 22 experts from 11 European scientific societies set out to define the first patient-centred diagnostic workflow that aims to prioritise testing for available biomarkers in individuals attending memory clinics.

Supervised Learning for Nonsequential Data: A Canonical Polyadic Decomposition Approach.

Efficient modeling of feature interactions underpins supervised learning for nonsequential tasks, characterized by a lack of inherent ordering of features (variables). The brute force approach of learning a parameter for each interaction of every order comes at an exponential computational and memory cost (curse of dimensionality). To alleviate this issue, it has been proposed to implicitly represent the model parameters as a tensor, the order of which is equal to the number of features; for efficiency, it can be further factorized into a compact tensor train (TT) format.