Decision trees to evaluate the risk of developing multiple sclerosis

Introduction

Multiple sclerosis (MS) is a persistent neurological condition impacting the central nervous system (CNS). The precise cause of multiple sclerosis is still uncertain; however, it is thought to arise from a blend of genetic and environmental factors. MS diagnosis includes assessing medical history, conducting neurological exams, performing magnetic resonance imaging (MRI) scans, and analyzing cerebrospinal fluid. While there is currently no cure for MS, numerous treatments exist to address symptoms, decelerate disease progression, and enhance the quality of life for individuals with MS.

Methods

This paper introduces a novel machine learning (ML) algorithm utilizing decision trees to address a key objective: creating a predictive tool for assessing the likelihood of MS development. It achieves this by combining prevalent demographic risk factors, specifically gender, with crucial immunogenetic risk markers, such as the alleles responsible for human leukocyte antigen (HLA) class I molecules and the killer immunoglobulin-like receptors (KIR) genes responsible for natural killer lymphocyte receptors.

Results

The study included 619 healthy controls and 299 patients affected by MS, all of whom originated from Sardinia. The gender feature has been disregarded due to its substantial bias in influencing the classification outcomes. By solely considering immunogenetic risk markers, the algorithm demonstrates an ability to accurately identify 73.24% of MS patients and 66.07% of individuals without the disease.

Discussion

Given its notable performance, this system has the potential to support clinicians in monitoring the relatives of MS patients and identifying individuals who are at an increased risk of developing the disease.

Decision trees for early prediction of inadequate immune response to coronavirus infections: a pilot study on COVID-19

Introduction

Few artificial intelligence models exist to predict severe forms of COVID-19. Most rely on post-infection laboratory data, hindering early treatment for high-risk individuals.

Methods

This study developed a machine learning model to predict inherent risk of severe symptoms after contracting SARS-CoV-2. Using a Decision Tree trained on 153 Alpha variant patients, demographic, clinical and immunogenetic markers were considered. Model performance was assessed on Alpha and Delta variant datasets. Key risk factors included age, gender, absence of KIR2DS2 gene (alone or with HLA-C C1 group alleles), presence of 14-bp polymorphism in HLA-G gene, presence of KIR2DS5 gene, and presence of KIR telomeric region A/A.

Results

The model achieved 83.01% accuracy for Alpha variant and 78.57% for Delta variant, with True Positive Rates of 80.82 and 77.78%, and True Negative Rates of 85.00% and 79.17%, respectively. The model showed high sensitivity in identifying individuals at risk.

Discussion

The present study demonstrates the potential of AI algorithms, combined with demographic, epidemiologic, and immunogenetic data, in identifying individuals at high risk of severe COVID-19 and facilitating early treatment. Further studies are required for routine clinical integration.

Towards a new image processing system at Wendelstein 7-X: From spatial calibration to characterization of thermal events

Wendelstein 7-X (W7-X) is the most advanced fusion experiment in the stellarator line and is aimed at proving that the stellarator concept is suitable for a fusion reactor. One of the most important issues for fusion reactors is the monitoring of plasma facing components when exposed to very high heat loads, through the use of visible and infrared (IR) cameras. In this paper, a new image processing system for the analysis of the strike lines on the inboard limiters from the first W7-X experimental campaign is presented. This system builds a model of the IR cameras through the use of spatial calibration techniques, helping to characterize the strike lines by using the information given by real spatial coordinates of each pixel. The characterization of the strike lines is made in terms of position, size, and shape, after projecting the camera image in a 2D grid which tries to preserve the curvilinear surface distances between points. The description of the strike-line shape is made by means of the Fourier Descriptors.

Infrared imaging systems for wall protection in the W7-X stellarator (invited)

Wendelstein 7-X aims at quasi-steady state operation with up to 10 MW of heating power for 30 min. Power exhaust will be handled predominantly via 10 actively water cooled CFC (carbon-fiber-reinforced carbon) based divertor units designed to withstand power loads of 10 MW/m² locally in steady state. If local loads exceed this value, a risk of local delamination of the CFC and failure of entire divertor modules arises. Infrared endoscopes to monitor all main plasma facing components are being prepared, and near real time software tools are under development to identify areas of excessive temperature rise, to distinguish them from non-critical events, and to trigger alarms. Tests with different cameras were made in the recent campaign. Long pulse operation enforces additional diagnostic design constraints: for example, the optics need to be thermally decoupled from the endoscope housing. In the upcoming experimental campaign, a graphite scraper element, in front of the island divertor throat, will be tested as a possible means to protect the divertor pumping gap edges during the transient discharge evolution.

Involvement of capsaicin-sensitive nerves in paraquat-induced mortality

Paraquat (PQ), a broad spectrum herbicide, produces severe lung inflammation and necrosis resulting in pulmonary fibrosis and respiratory failure. Tachykinins are peptides released by sensory C fibers and have the ability of influencing respiratory functions and cellular proliferation. To examine whether the damage caused by PQ involves tachykinins, rats were depleted in their content of tachykinins by systemic treatment with capsaicin prior to PQ exposure. The animal subjected to this treatment showed a 3-fold higher viability compared to those treated with PQ alone (75 vs 27%). Depletion of reduced glutathione (GSH) is associated with oxidative stress produced by reactive oxygen intermediates during PQ metabolism. This is considered to be critical in the pathogenesis of lung damage by PQ. PQ treatment induced a significant depletion of GSH during the first days and a similar effect was also observed in the group of capsaicin-pretreated rats. Four weeks after PQ treatment the levels of GSH were similar to controls in rat pretreated or not with capsaicin plus PQ. This may indicate that the reduced levels of GSH may be associated to the toxicity observed in the acute phase, but not of importance in the final PQ-induced mortality. Neutral endopeptidase (NEP) is an enzyme considered to be critical in controlling the levels of tachykinins. Exposure of crude membrane preparations of rat lung to PQ resulted in a dose-dependent inhibition of NEP activity. Since NEP inactivation may occur in lung following a PQ exposure in vivo, the results indicate that during PQ intoxication a more sustained activity of tachykinins may be present, producing effects such as cell proliferation, fluid extravasation and bronchoconstriction. In conclusion, this finding supports the hypothesis that neuropeptides released from capsaicin-sensitive nerves could be involved in the modulation of PQ-induced lung damage.