EAACI guidelines on environmental science in allergic diseases and asthma - Leveraging artificial intelligence and machine learning to develop a causality model in exposomics.

Allergic diseases and asthma are intrinsically linked to the environment we live in and to patterns of exposure. The integrated approach to understanding the effects of exposures on the immune system includes the ongoing collection of large-scale and complex data. This requires sophisticated methods to take full advantage of what this data can offer. Here we discuss the progress and further promise of applying artificial intelligence and machine-learning approaches to help unlock the power of complex environmental data sets toward providing causality models of exposure and intervention. We discuss a range of relevant machine-learning paradigms and models including the way such models are trained and validated together with examples of machine learning applied to allergic disease in the context of specific environmental exposures as well as attempts to tie these environmental data streams to the full representative exposome. We also discuss the promise of artificial intelligence in personalized medicine and the methodological approaches to healthcare with the final AI to improve public health.

Multidimensional endotyping using nasal proteomics predicts molecular phenotypes in the asthmatic airways.

BACKGROUND: Unsupervised clustering of biomarkers derived from noninvasive samples such as nasal fluid is less evaluated as a tool for describing asthma endotypes. OBJECTIVE: We sought to evaluate whether protein expression in nasal fluid would identify distinct clusters of patients with asthma with specific lower airway molecular phenotypes. METHODS: Unsupervised clustering of 168 nasal inflammatory and immune proteins and Shapley values was used to stratify 43 patients with severe asthma (endotype of noneosinophilic asthma) using a 2 "modeling blocks" machine learning approach. This algorithm was also applied to nasal brushings transcriptomics from U-BIOPRED (Unbiased Biomarkers for the Prediction of Respiratory Diseases Outcomes). Feature reduction and functional gene analysis were used to compare proteomic and transcriptomic clusters. Gene set variation analysis provided enrichment scores of the endotype of noneosinophilic asthma protein signature within U-BIOPRED sputum and blood. RESULTS: The nasal protein machine learning model identified 2 severe asthma endotypes, which were replicated in U-BIOPRED nasal transcriptomics. Cluster 1 patients had significant airway obstruction, small airways disease, air trapping, decreased diffusing capacity, and increased oxidative stress, although only 4 of 18 were current smokers. Shapley identified 20 cluster-defining proteins. Forty-one proteins were significantly higher in cluster 1. Pathways associated with proteomic and transcriptomic clusters were linked to TH1, TH2, neutrophil, Janus kinase-signal transducer and activator of transcription, TLR, and infection activation. Gene set variation analysis of the nasal protein and gene signatures were enriched in subjects with sputum neutrophilic/mixed granulocytic asthma and in subjects with a molecular phenotype found in sputum neutrophil-high subjects. CONCLUSIONS: Protein or gene analysis may indicate molecular phenotypes within the asthmatic lower airway and provide a simple, noninvasive test for non-type 2 immune response asthma that is currently unavailable.

An overview of the National COVID-19 Chest Imaging Database: data quality and cohort analysis.

BACKGROUND: The National COVID-19 Chest Imaging Database (NCCID) is a centralized database containing mainly chest X-rays and computed tomography scans from patients across the UK. The objective of the initiative is to support a better understanding of the coronavirus SARS-CoV-2 disease (COVID-19) and the development of machine learning technologies that will improve care for patients hospitalized with a severe COVID-19 infection. This article introduces the training dataset, including a snapshot analysis covering the completeness of clinical data, and availability of image data for the various use-cases (diagnosis, prognosis, longitudinal risk). An additional cohort analysis measures how well the NCCID represents the wider COVID-19-affected UK population in terms of geographic, demographic, and temporal coverage. FINDINGS: The NCCID offers high-quality DICOM images acquired across a variety of imaging machinery; multiple time points including historical images are available for a subset of patients. This volume and variety make the database well suited to development of diagnostic/prognostic models for COVID-associated respiratory conditions. Historical images and clinical data may aid long-term risk stratification, particularly as availability of comorbidity data increases through linkage to other resources. The cohort analysis revealed good alignment to general UK COVID-19 statistics for some categories, e.g., sex, whilst identifying areas for improvements to data collection methods, particularly geographic coverage. CONCLUSION: The NCCID is a growing resource that provides researchers with a large, high-quality database that can be leveraged both to support the response to the COVID-19 pandemic and as a test bed for building clinically viable medical imaging models.

Towards nationally curated data archives for clinical radiology image analysis at scale: Learnings from national data collection in response to a pandemic.

The prevalence of the coronavirus SARS-CoV-2 disease has resulted in the unprecedented collection of health data to support research. Historically, coordinating the collation of such datasets on a national scale has been challenging to execute for several reasons, including issues with data privacy, the lack of data reporting standards, interoperable technologies, and distribution methods. The coronavirus SARS-CoV-2 disease pandemic has highlighted the importance of collaboration between government bodies, healthcare institutions, academic researchers and commercial companies in overcoming these issues during times of urgency. The National COVID-19 Chest Imaging Database, led by NHSX, British Society of Thoracic Imaging, Royal Surrey NHS Foundation Trust and Faculty, is an example of such a national initiative. Here, we summarise the experiences and challenges of setting up the National COVID-19 Chest Imaging Database, and the implications for future ambitions of national data curation in medical imaging to advance the safe adoption of artificial intelligence in healthcare.

Spatial coherence in 2D holography.

Holography is a long-established technique to encode an object's spatial information into a lower-dimensional representation. We investigate the role of the illumination's spatial coherence properties in the success of such an imaging system through point spread function and Fourier domain analysis. Incoherent illumination is shown to result in more robust imaging performance free of diffraction artifacts at the cost of incurring background noise and sacrificing phase retrieval. Numerical studies confirm that this background noise reduces image sensitivity as the image size increases, in agreement with other similar systems. Following this analysis, we demonstrate a 2D holographic imaging system realized with lensless, 1D measurements of microwave fields generated by dynamic metasurface apertures.

Erratum: Optimal Frames for Polarization State Reconstruction [Phys. Rev. Lett. 115, 263901 (2015)].

This corrects the article DOI: 10.1103/PhysRevLett.115.263901.

Automated Steerable Path Planning for Deep Brain Stimulation Safeguarding Fiber Tracts and Deep Gray Matter Nuclei.

Deep Brain Stimulation (DBS) is a neurosurgical procedure consisting in the stereotactic implantation of stimulation electrodes to specific brain targets, such as deep gray matter nuclei. Current solutions to place the electrodes rely on rectilinear stereotactic trajectories (RTs) manually defined by surgeons, based on pre-operative images. An automatic path planner that accurately targets subthalamic nuclei (STN) and safeguards critical surrounding structures is still lacking. Also, robotically-driven curvilinear trajectories (CTs) computed on the basis of state-of-the-art neuroimaging would decrease DBS invasiveness, circumventing patient-specific obstacles. This work presents a new algorithm able to estimate a pool of DBS curvilinear trajectories for reaching a given deep target in the brain, in the context of the EU's Horizon EDEN2020 project. The prospect of automatically computing trajectory plans relying on sophisticated newly engineered steerable devices represents a breakthrough in the field of microsurgical robotics. By tailoring the paths according to single-patient anatomical constraints, as defined by advanced preoperative neuroimaging including diffusion MR tractography, this planner ensures a higher level of safety than the standard rectilinear approach. Ten healthy controls underwent Magnetic Resonance Imaging (MRI) on 3T scanner, including 3DT1-weighted sequences, 3Dhigh-resolution time-of-flight MR angiography (TOF-MRA) and high angular resolution diffusion MR sequences. A probabilistic q-ball residual-bootstrap MR tractography algorithm was used to reconstruct motor fibers, while the other deep gray matter nuclei surrounding STN and vessels were segmented on T1 and TOF-MRA images, respectively. These structures were labeled as obstacles. The reliability of the automated planner was evaluated; CTs were compared to RTs in terms of efficacy and safety. Targeting the anterior STN, CTs performed significantly better in maximizing the minimal distance from critical structures, by finding a tuned balance between all obstacles. Moreover, CTs resulted superior in reaching the center of mass (COM) of STN, as well as in optimizing the entry angle in STN and in the skull surface.

Prior-predictive value from fast-growth simulations: error analysis and bias estimation.

Variants of fluctuation theorems recently discovered in the statistical mechanics of nonequilibrium processes may be used for the efficient determination of high-dimensional integrals as typically occurring in Bayesian data analysis. In particular for multimodal distributions, Monte Carlo procedures not relying on perfect equilibration are advantageous. We provide a comprehensive statistical error analysis for the determination of the prior-predictive value (the evidence) in a Bayes problem, building on a variant of the Jarzynski equation. Special care is devoted to the characterization of the bias intrinsic to the method and statistical errors arising from exponential averages. We also discuss the determination of averages over multimodal posterior distributions with the help of a consequence of the Crooks relation. All our findings are verified by extensive numerical simulations of two model systems with bimodal likelihoods.

Optimal Frames for Polarization State Reconstruction.

Complete determination of the polarization state of light requires at least four distinct projective measurements of the associated Stokes vector. Stability of state reconstruction, however, hinges on the condition number κ of the corresponding instrument matrix. Optimization of redundant measurement frames with an arbitrary number of analysis states, m, is considered in this Letter in the sense of minimization of κ. The minimum achievable κ is analytically found and shown to be independent of m, except for m=5 where this minimum is unachievable. Distribution of the optimal analysis states over the Poincaré sphere is found to be described by spherical 2 designs, including the Platonic solids as special cases. Higher order polarization properties also play a key role in nonlinear, stochastic, and quantum processes. Optimal measurement schemes for nonlinear measurands of degree t are hence also considered and found to correspond to spherical 2t designs, thereby constituting a generalization of the concept of mutually unbiased bases.

Comment on "Correct definition of the Poynting vector in electrically and magnetically polarizable medium reveals that negative refraction is impossible.".

Physically valid electromagnetic continuity equations can be generated from either the usual form of the Poynting vector E x H or the alternative E x B form. However, the continuity equations are not identical, which means that quantities following from E x H cannot always be compared directly to those from E x B. In particular, the work done on the bound current densities are attributed differently in the two representations.We also comment on the negative refraction condition used.

Pulmonary vein isolation predicts freedom from arrhythmia after circumferential antral ablation for paroxysmal atrial fibrillation.

OBJECTIVES: The aims of this observational study were to evaluate (i) the feasibility of obtaining bidirectional pulmonary vein (PV) isolation by means of circumferential radiofrequency ablation of the antral aspect of the PV ostium; (ii) whether the electrophysiological demonstration of bidirectional PV isolation predicts freedom from atrial tachyarrhythmia recurrence after ablation in patients with paroxysmal atrial fibrillation. METHODS: The study group comprised 28 patients affected by frequent recurrences of paroxysmal atrial fibrillation refractory to antiarrhythmic drugs, who underwent transcatheter ablation of the PVs by means of a non-fluoroscopic navigation system. Radiofrequency pulses were delivered in a point-by-point fashion at the antral aspect of the ostium of each vein presenting distal PV potentials. After ablation of each PV, bidirectional isolation was tested by means of a basket catheter. No antiarrhythmic drugs were prescribed on discharge. Outpatient visits, 24-h electrocardiographic Holter monitoring, and continuous 7-day digital electrocardiogram were scheduled at 3, 6, and 12 months. RESULTS: A distal potential was detected in 101/123 (82%) mapped PVs. Bidirectional isolation was obtained in 81/101 (80%) PVs; bidirectional isolation of all targeted PVs was obtained in 17 (61%) patients. After a mean follow-up of 12.2 +/- 4.2 months, clinical success was observed in 15 (53%) patients. On multivariate analysis, only bidirectional isolation of all targeted PVs predicted the clinical success of ablation (P < 0.003; hazard ratio 7.504; confidence interval 1.943-28.990). CONCLUSIONS: Circumferential antral ablation achieves bidirectional isolation in 80% of PVs. Bidirectional isolation of all PVs is essential to curing patients with paroxysmal atrial fibrillation.

Transesophageal echocardiography and intracardiac echocardiography differently predict potential technical challenges or failures of interatrial shunts catheter-based closure.

We sought to prospectively assess the role of transesophageal (TEE) and intracardiac echocardiography (ICE) in detecting potential technical difficulties or failures in patients submitted to interatrial shunts percutaneous closure. We prospectively enrolled 46 consecutive patients (mean age 35+/-28, 8 years, 30 female) referred to our center for catheter-based closure of interatrial shunts. All patients were screened with TEE before the intervention. Patients who met the inclusion criteria underwent ICE study before the closure attempt (40 patients). TEE detected potential technical difficulties in 22.5% (9/40) patients, whereas ICE detected technical difficulties in 32.5% (13/40 patients). In patients with positive TEE/ICE the procedural success (92.4% versus 100% and, P = ns) and follow-up failure rate (7.7% versus 0%, P = ns) were similar to patients with negative TEE/ICE, whereas the fluoroscopy time (7 +/- 1.2 versus 5 +/- 0.7 minutes, P < 0.03), the procedural time (41 +/- 4.1 versus 30 +/- 8.2 minutes, P +/- 0.03), and technical difficulties rate (23.1% versus 0%, P = 0.013) were higher. Differences between ICE and TEE in the evaluation of rims, measurement of ASD or fossa ovalis, and detection of venous valve and embryonic septal membrane remnants impacted on technical challenges and on procedural and fluoroscopy times but did not influence the success rate and follow-up failure rate.