Radiomics, a Promising New Discipline: Example of Hepatocellular Carcinoma.

Radiomics is a discipline that involves studying medical images through their digital data. Using "artificial intelligence" algorithms, radiomics utilizes quantitative and high-throughput analysis of an image's textural richness to obtain relevant information for clinicians, from diagnosis assistance to therapeutic guidance. Exploitation of these data could allow for a more detailed characterization of each phenotype, for each patient, making radiomics a new biomarker of interest, highly promising in the era of precision medicine. Moreover, radiomics is non-invasive, cost-effective, and easily reproducible in time. In the field of oncology, it performs an analysis of the entire tumor, which is impossible with a single biopsy but is essential for understanding the tumor's heterogeneity and is known to be closely related to prognosis. However, current results are sometimes less accurate than expected and often require the addition of non-radiomics data to create a performing model. To highlight the strengths and weaknesses of this new technology, we take the example of hepatocellular carcinoma and show how radiomics could facilitate its diagnosis in difficult cases, predict certain histological features, and estimate treatment response, whether medical or surgical.

Discrete Box-Constrained Minimax Classifier for Uncertain and Imbalanced Class Proportions.

This paper aims to build a supervised classifier for dealing with imbalanced datasets, uncertain class proportions, dependencies between features, the presence of both numeric and categorical features, and arbitrary loss functions. The Bayes classifier suffers when prior probability shifts occur between the training and testing sets. A solution is to look for an equalizer decision rule whose class-conditional risks are equal. Such a classifier corresponds to a minimax classifier when it maximizes the Bayes risk. We develop a novel box-constrained minimax classifier which takes into account some constraints on the priors to control the risk maximization. We analyze the empirical Bayes risk with respect to the box-constrained priors for discrete inputs. We show that this risk is a concave non-differentiable multivariate piecewise affine function. A projected subgradient algorithm is derived to maximize this empirical Bayes risk over the box-constrained simplex. Its convergence is established and its speed is bounded. The optimization algorithm is scalable when the number of classes is large. The robustness of our classifier is studied on diverse databases. Our classifier, jointly applied with a clustering algorithm to process mixed attributes, tends to equalize the class-conditional risks while being not too pessimistic.

Dynamic Image Quantization Using Leaky Integrate-and-Fire Neurons.

This paper introduces a novel coding/decoding mechanism that mimics one of the most important properties of the human visual system: its ability to enhance the visual perception quality in time. In other words, the brain takes advantage of time to process and clarify the details of the visual scene. This characteristic is yet to be considered by the state-of-the-art quantization mechanisms that process the visual information regardless the duration of time it appears in the visual scene. We propose a compression architecture built of neuroscience models; it first uses the leaky integrate-and-fire (LIF) model to transform the visual stimulus into a spike train and then it combines two different kinds of spike interpretation mechanisms (SIM), the time-SIM and the rate-SIM for the encoding of the spike train. The time-SIM allows a high quality interpretation of the neural code and the rate-SIM allows a simple decoding mechanism by counting the spikes. For that reason, the proposed mechanisms is called Dual-SIM quantizer (Dual-SIMQ). We show that (i) the time-dependency of Dual-SIMQ automatically controls the reconstruction accuracy of the visual stimulus, (ii) the numerical comparison of Dual-SIMQ to the state-of-the-art shows that the performance of the proposed algorithm is similar to the uniform quantization schema while it approximates the optimal behavior of the non-uniform quantization schema and (iii) from the perceptual point of view the reconstruction quality using the Dual-SIMQ is higher than the state-of-the-art.

Correction: Stratification and prediction of remission in first-episode psychosis patients: the OPTiMiSE cohort study.

The original Article did not feature the list of collaborators. This has now been corrected in the PDF and HTML versions of this Article.

Stratification and prediction of remission in first-episode psychosis patients: the OPTiMiSE cohort study.

Early response to first-line antipsychotic treatments is strongly associated with positive long-term symptomatic and functional outcome in psychosis. Unfortunately, attempts to identify reliable predictors of treatment response in first-episode psychosis (FEP) patients have not yet been successful. One reason for this could be that FEP patients are highly heterogeneous in terms of symptom expression and underlying disease biological mechanisms, thereby impeding the identification of one-size-fits-all predictors of treatment response. We have used a clustering approach to stratify 325 FEP patients into four clinical subtypes, termed C1A, C1B, C2A and C2B, based on their symptoms assessed using the Positive and Negative Syndrome Scale (PANSS) scale. Compared to C1B, C2A and C2B patients, those from the C1A subtype exhibited the most severe symptoms and were the most at risk of being non-remitters when treated with the second-generation antipsychotic drug amisulpride. Before treatment, C1A patients exhibited higher serum levels of several pro-inflammatory cytokines and inflammation-associated biomarkers therefore validating our stratification approach on external biological measures. Most importantly, in the C1A subtype, but not others, lower serum levels of interleukin (IL)-15, higher serum levels of C-X-C motif chemokine 12 (CXCL12), previous exposure to cytomegalovirus (CMV), use of recreational drugs and being younger were all associated with higher odds of being non-remitters 4 weeks after treatment. The predictive value of this model was good (mean area under the curve (AUC) = 0.73 ± 0.10), and its specificity and sensitivity were 45 ± 0.09% and 83 ± 0.03%, respectively. Further validation and replication of these results in clinical trials would pave the way for the development of a blood-based assisted clinical decision support system in psychosis.

Retina-Inspired Filter.

This paper introduces a novel filter, which is inspired by the human retina. The human retina consists of three different layers: the Outer Plexiform Layer (OPL), the inner plexiform layer, and the ganglionic layer. Our inspiration is the linear transform which takes place in the OPL and has been mathematically described by the neuroscientific model "virtual retina." This model is the cornerstone to derive the non-separable spatio-temporal OPL retina-inspired filter, briefly renamed retina-inspired filter, studied in this paper. This filter is connected to the dynamic behavior of the retina, which enables the retina to increase the sharpness of the visual stimulus during filtering before its transmission to the brain. We establish that this retina-inspired transform forms a group of spatio-temporal Weighted Difference of Gaussian (WDoG) filters when it is applied to a still image visible for a given time. We analyze the spatial frequency bandwidth of the retina-inspired filter with respect to time. It is shown that the WDoG spectrum varies from a lowpass filter to a bandpass filter. Therefore, while time increases, the retina-inspired filter enables to extract different kinds of information from the input image. Finally, we discuss the benefits of using the retina-inspired filter in image processing applications such as edge detection and compression.

Epsilon-optimal non-Bayesian anomaly detection for parametric tomography.

The non-Bayesian detection of an anomaly from a single or a few noisy tomographic projections is considered as a statistical hypotheses testing problem. It is supposed that a radiography is composed of an imaged nonanomalous background medium, considered as a deterministic nuisance parameter, with a possibly hidden anomaly. Because the full voxel-by-voxel reconstruction is impossible, an original tomographic method based on the parametric models of the nonanomalous background medium and radiographic process is proposed to fill up the gap in the missing data. Exploiting this "parametric tomography," a new detection scheme with a limited loss of optimality is proposed as an alternative to the nonlinear generalized likelihood ratio test, which is untractable in the context of nondestructive testing for the objects with uncertainties in their physical/geometrical properties. The theoretical results are illustrated by the processing of real radiographies for the nuclear fuel rod inspection.