Feature selection from magnetic resonance imaging data in ALS: a systematic review

Cortical thickness correlated with peripheral inflammatory cytokines in amyotrophic lateral sclerosis

Introduction

Amyotrophic lateral sclerosis (ALS) is a rare, devastating neurodegenerative disease that affects upper and lower motor neurons, resulting in muscle atrophy, spasticity, hyperreflexia, and paralysis. Inflammation plays an important role in the development of ALS, and associated with rapid disease progression. Current observational studies indicate the thinning of cortical thickness in patients with ALS is associated with rapid disease progression and cognitive changes. However, the effects of inflammatory cytokines on cortical thickness in patients with ALS are unclear. Here, we investigated the relationship between inflammatory cytokines and cortical thickness in patients with ALS.

Methods

We evaluated 51 patients with ALS for inflammatory cytokines including interleukin (IL)-4, interferon (IFN)-α, IL-1β, IL-2, IL-5, IL-12, tumor necrosis factor (TNF)-α, IL-6, IL-10, IL-8, IL-17, and IFN-γ and analyzed the correlation between these indicators and the ALS functional rating scale-revised (ALSFRS-R) score or disease progression rate (ΔFS score). Twenty-six patients with ALS and 26 controls were studied using whole-cortex analysis, and post-hoc analyses were performed to examine the correlation between brain cortical thickness and ALSFRS-R or ΔFS scores.

Results

IL-4, IFN-α, IL-1β, and IL-2 levels were significantly correlated with ALSFRS-R scores, and the IL-2 level was significantly correlated with ΔFS scores. After controlling for age and sex, the ALS group had thinner cortexes in multiple clusters across the brain than the control group. Further analyses revealed that cortical thickness in the right superior temporal and lingual gyrus regions was inversely correlated with ΔFS scores. There was a significant positive correlation between the clusters in the right lingual cortex and IL-2 level.

Conclusion

These results suggest cortical thickness was reduced in patients with ALS in motor and non-motor cortical areas. Inflammatory factors (especially IL-2) were correlated with cortical thickness, and both were related to the disease progression rate, suggesting IL-2 plays an important role in ALS.

Surface-based Analyses of Diffusional Kurtosis Imaging in Amyotrophic Lateral Sclerosis: Relationship with Onset Subtypes

PURPOSE

Here, we aimed to characterize the cortical and subcortical microstructural alterations in the brains of patients with amyotrophic lateral sclerosis (ALS). In particular, we compared these features between bulbar-onset ALS (b-ALS) and limb-onset ALS (l-ALS).

METHODS

Diffusion MRI data (b = 0, 700, 2000 ms/mm2, 1.7-mm isotropic voxel) from 28 patients with ALS (9 b-ALS and 19 l-ALS) and 17 healthy control subjects (HCs) were analyzed. Diffusional kurtosis imaging (DKI) metrics were sampled at the mid-cortical and subcortical surfaces. We used permutation testing with a nonparametric combination of mean diffusivity (MD), fractional anisotropy (FA), and mean kurtosis (MK) to assess intergroup differences over the cerebrum. We also carried out an atlas-based analysis focusing on Brodmann Area 4 and 6 (primary motor and premotor areas) and investigated the correlation between MRI metrics and clinical parameters.

RESULTS

At both the mid-cortical and subcortical surfaces, b-ALS was associated with significantly greater MD, smaller FA, and smaller MK in the motor and premotor areas than HC. In contrast, the patients with l-ALS showed relatively moderate differences relative to HCs. The ALS Functional Rating Scale-Revised bulbar subscore was significantly correlated with the diffusion metrics in Brodmann Area 4.

CONCLUSION

The distribution of abnormalities over the cerebral hemispheres and the more severe microstructural alteration in b-ALS compared to l-ALS were in good agreement with findings from postmortem histology. Our results suggest the feasibility of surface-based DKI analyses for exploring brain microstructural pathologies in ALS. The observed differences between b-ALS and l-ALS and their correlations with functional bulbar impairment support the clinical relevance of DKI measurement in the cortical and juxtacortical regions of patients with ALS.

Artificial intelligence for screening and diagnosis of amyotrophic lateral sclerosis: a systematic review and meta-analysis

INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a rare and fatal neurological disease that leads to progressive motor function degeneration. Diagnosing ALS is challenging due to the absence of a specific detection test. The use of artificial intelligence (AI) can assist in the investigation and treatment of ALS.

METHODS

We searched seven databases for literature on the application of AI in the early diagnosis and screening of ALS in humans. The findings were summarized using random-effects summary receiver operating characteristic curve. The risk of bias (RoB) analysis was carried out using QUADAS-2 or QUADAS-C tools.

RESULTS

In the 34 analyzed studies, a meta-prevalence of 47% for ALS was noted. For ALS detection, the pooled sensitivity of AI models was 94.3% (95% CI - 63.2% to 99.4%) with a pooled specificity of 98.9% (95% CI - 92.4% to 99.9%). For ALS classification, the pooled sensitivity of AI models was 90.9% (95% CI - 86.5% to 93.9%) with a pooled specificity of 92.3% (95% CI - 84.8% to 96.3%). Based on type of input for classification, the pooled sensitivity of AI models for gait, electromyography, and magnetic resonance signals was 91.2%, 92.6%, and 82.2%, respectively. The pooled specificity for gait, electromyography, and magnetic resonance signals was 94.1%, 96.5%, and 77.3%, respectively.

CONCLUSIONS

Although AI can play a significant role in the screening and diagnosis of ALS due to its high sensitivities and specificities, concerns remain regarding quality of evidence reported in the literature.

Potential of neuroimaging as a biomarker in amyotrophic lateral sclerosis: from structure to metabolism

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease characterized by motor neuron degeneration. The development of ALS involves metabolite alterations leading to tissue lesions in the nervous system. Recent advances in neuroimaging have significantly improved our understanding of the underlying pathophysiology of ALS, with findings supporting the corticoefferent axonal disease progression theory. Current studies on neuroimaging in ALS have demonstrated inconsistencies, which may be due to small sample sizes, insufficient statistical power, overinterpretation of findings, and the inherent heterogeneity of ALS. Deriving meaningful conclusions solely from individual imaging metrics in ALS studies remains challenging, and integrating multimodal imaging techniques shows promise for detecting valuable ALS biomarkers. In addition to giving an overview of the principles and techniques of different neuroimaging modalities, this review describes the potential of neuroimaging biomarkers in the diagnosis and prognostication of ALS. We provide an insight into the underlying pathology, highlighting the need for standardized protocols and multicenter collaborations to advance ALS research.

Cerebrospinal fluid and blood exosomes as biomarkers for amyotrophic lateral sclerosis; a systematic review

BACKGROUND

Amyotrophic lateral sclerosis (ALS) is a progressive and fatal motor neuron disease. Due to the limited knowledge about potential biomarkers that help in early diagnosis and monitoring disease progression, today's diagnoses are based on ruling out other diseases, neurography, and electromyography examination, which takes a time-consuming procedure.

METHODS

PubMed, ScienceDirect, and Web of Science were explored to extract articles published from January 2015 to June 2023. In the searching strategy following keywords were included; amyotrophic lateral sclerosis, biomarkers, cerebrospinal fluid, serum, and plama.

RESULTS

A total number of 6 studies describing fluid-based exosomal biomarkers were included in this study. Aggregated proteins including SOD1, TDP-43, pTDP-43, and FUS could be detected in the microvesicles (MVs). Moreover, TDP-43 and NFL extracted from plasma exosomes could be used as prognostic biomarkers. Also, downregulated miR-27a-3p detected through exoEasy Maxi and exoQuick Kit in the plasma could be measured as a diagnostic biomarker. Eventually, the upregulated level of CORO1A could be used to monitor disease progression.

CONCLUSION

Based on the results, each biomarker alone is insufficient to evaluate ALS. CNS-derived exosomes contain multiple ALS-related biomarkers (SOD1, TDP-43, pTDP-43, FUS, and miRNAs) that are detectable in cerebrospinal fluid and blood is a proper alternation. Exosome detecting kits listed as exoEasy, ExoQuick, Exo-spin, ME kit, ExoQuick Plus, and Exo-Flow, are helpful to reach this purpose.

Cell and gene therapy for amyotrophic lateral sclerosis

Amyotrophic lateral sclerosis (ALS) is a fatal and incurable neurodegenerative disorder with rapidly progressive skeletal muscle weakness, which can also cause a variable cognitive deficit. Genetic causes are only identified in approximately 10% of all cases, with complex genotype-phenotype associations, making it challenging to identify treatment targets. What further hampers therapeutic development is a broad heterogeneity in mechanisms, possible targets, and disturbances across various cell types, aside from the cortical and spinal motor neurons that lie at the heart of the pathology of ALS. Over the last decade, significant progress in biotechnologic techniques, cell and ribonucleic acid (RNA) engineering, animal models, and patient-specific human stem cell and organoid models have accelerated both mechanistic and therapeutic discoveries. The growing number of clinical trials mirrors this. This chapter reviews the current state of human preclinical models supporting trial strategies as well as recent clinical cell and gene therapy approaches.

Amyotrophic lateral sclerosis with upper motor neuron predominance: diagnostic accuracy of qualitative and quantitative susceptibility metrics in the precentral gyrus

OBJECTIVE

The study aims at comparing the diagnostic accuracy of qualitative and quantitative assessment of the susceptibility in the precentral gyrus in detecting amyotrophic lateral sclerosis (ALS) with predominance of upper motor neuron (UMN) impairment.

METHODS

We retrospectively collected clinical and 3T MRI data of 47 ALS patients, of whom 12 with UMN predominance (UMN-ALS). We further enrolled 23 healthy controls (HC) and 15 ALS Mimics (ALS-Mim). The Motor Cortex Susceptibility (MCS) score was qualitatively assessed on the susceptibility-weighted images (SWI) and automatic metrics were extracted from the quantitative susceptibility mapping (QSM) in the precentral gyrus. MCS scores and QSM-based metrics were tested for correlation, and ROC analyses.

RESULTS

The correlation of MCS score and susceptibility skewness was significant (Rho = 0.55, p < 0.001). The susceptibility SD showed an AUC of 0.809 with a specificity and positive predictive value of 100% in differentiating ALS and ALS Mim versus HC, significantly higher than MCS (Z = -3.384, p-value = 0.00071). The susceptibility skewness value of -0.017 showed specificity of 92.3% and predictive positive value of 91.7% in differentiating UMN-ALS versus ALS mimics, even if the performance was not significantly better than MCS (Z = 0.81, p = 0.21).

CONCLUSION

The MCS and susceptibility skewness of the precentral gyrus show high diagnostic accuracy in differentiating UMN-ALS from ALS-mimics subjects. The quantitative assessment might be preferred being an automatic measure unbiased by the reader.

CLINICAL RELEVANCE STATEMENT

The clinical diagnostic evaluation of ALS patients might benefit from the qualitative and/or quantitative assessment of the susceptibility in the precentral gyrus as imaging marker of upper motor neuron predominance.

KEY POINTS

• Amyotrophic lateral sclerosis diagnostic work-up lacks biomarkers able to identify upper motor neuron involvement. • Susceptibility-weighted imaging/quantitative susceptibility mapping-based measures showed good diagnostic accuracy in discriminating amyotrophic lateral sclerosis with predominant upper motor neuron impairment from patients with suspected motor neuron disorder. • Susceptibility-weighted imaging/quantitative susceptibility mapping-based assessment of the magnetic susceptibility provides a diagnostic marker for amyotrophic lateral sclerosis with upper motor neuron predominance.

Digital Alternative Communication for Individuals with Amyotrophic Lateral Sclerosis: What We Have

Amyotrophic Lateral Sclerosis is a disease that compromises the motor system and the functional abilities of the person in an irreversible way, causing the progressive loss of the ability to communicate. Tools based on Augmentative and Alternative Communication are essential for promoting autonomy and improving communication, life quality, and survival. This Systematic Literature Review aimed to provide evidence on eye-image-based Human-Computer Interaction approaches for the Augmentative and Alternative Communication of people with Amyotrophic Lateral Sclerosis. The Systematic Literature Review was conducted and guided following a protocol consisting of search questions, inclusion and exclusion criteria, and quality assessment, to select primary studies published between 2010 and 2021 in six repositories: Science Direct, Web of Science, Springer, IEEE Xplore, ACM Digital Library, and PubMed. After the screening, 25 primary studies were evaluated. These studies showcased four low-cost, non-invasive Human-Computer Interaction strategies employed for Augmentative and Alternative Communication in people with Amyotrophic Lateral Sclerosis. The strategies included Eye-Gaze, which featured in 36% of the studies; Eye-Blink and Eye-Tracking, each accounting for 28% of the approaches; and the Hybrid strategy, employed in 8% of the studies. For these approaches, several computational techniques were identified. For a better understanding, a workflow containing the development phases and the respective methods used by each strategy was generated. The results indicate the possibility and feasibility of developing Human-Computer Interaction resources based on eye images for Augmentative and Alternative Communication in a control group. The absence of experimental testing in people with Amyotrophic Lateral Sclerosis reiterates the challenges related to the scalability, efficiency, and usability of these technologies for people with the disease. Although challenges still exist, the findings represent important advances in the fields of health sciences and technology, promoting a promising future with possibilities for better life quality.

Sequential alterations in diffusion metrics as correlates of disease severity in amyotrophic lateral sclerosis

BACKGROUND AND OBJECTIVE

The neuropathology of amyotrophic lateral sclerosis (ALS) follows a regional distribution pattern in the brain with four stages. Using diffusion tensor imaging (DTI), this pattern can be translated into a tract-based staging scheme to assess cerebral progression in vivo. This study investigates the association between the sequential alteration pattern and disease severity in patients with ALS.

METHODS

DTI data of 325 patients with ALS and 130 healthy controls were analyzed in a tract of interest (TOI)-based approach. Patients were categorized according to their ALS-FRS-R scores into groups with declining functionality. The fractional anisotropy (FA) values in the tracts associated with neuropathological stages were group-wise compared with healthy controls.

RESULTS

The FA in the tracts associated with ALS stages showed a decrease which could be related to the disease severity stratification, i.e., at the group level, the lower the ALS-FRS-R of the categorized patient group, the higher was the effect size of the stage-related tract. In the patient group with the highest ALS-FRS-R, Cohen's d showed a medium effect size in the corticospinal tract and small effect sizes in the other stage-related tracts. Overall, the lower the ALS-FRS-R of the categorized patient group the higher was the effect size of the comparison with healthy controls.

CONCLUSION

The progression of white matter alterations across tracts according to the model of sequential tract involvement is associated with clinical disease severity in patients with ALS, suggesting the use of staging-based DTI as a technical marker for disease progression.

Diffusion Tensor Imaging in Amyotrophic Lateral Sclerosis: Machine Learning for Biomarker Development

Diffusion tensor imaging (DTI) allows the in vivo imaging of pathological white matter alterations, either with unbiased voxel-wise or hypothesis-guided tract-based analysis. Alterations of diffusion metrics are indicative of the cerebral status of patients with amyotrophic lateral sclerosis (ALS) at the individual level. Using machine learning (ML) models to analyze complex and high-dimensional neuroimaging data sets, new opportunities for DTI-based biomarkers in ALS arise. This review aims to summarize how different ML models based on DTI parameters can be used for supervised diagnostic classifications and to provide individualized patient stratification with unsupervised approaches in ALS. To capture the whole spectrum of neuropathological signatures, DTI might be combined with additional modalities, such as structural T1w 3-D MRI in ML models. To further improve the power of ML in ALS and enable the application of deep learning models, standardized DTI protocols and multi-center collaborations are needed to validate multimodal DTI biomarkers. The application of ML models to multiparametric MRI/multimodal DTI-based data sets will enable a detailed assessment of neuropathological signatures in patients with ALS and the development of novel neuroimaging biomarkers that could be used in the clinical workup.

Artificial neural networks for non-linear age correction of diffusion metrics in the brain

Human aging is characterized by progressive loss of physiological functions. To assess changes in the brain that occur with increasing age, the concept of brain aging has gained momentum in neuroimaging with recent advancements in statistical regression and machine learning (ML). A common technique to assess the brain age of a person is, first, fitting a regression model to neuroimaging data from a group of healthy subjects, and then, using the resulting model for age prediction. Although multiparametric MRI-based models generally perform best, models solely based on diffusion tensor imaging have achieved similar results, with the benefits of faster data acquisition and better replicability across scanners and field strengths. In the present study, we developed an artificial neural network (ANN) for brain age prediction based upon tract-based fractional anisotropy (FA). Consequently, we investigated if this age-prediction model could also be used for non-linear age correction of white matter diffusion metrics in healthy adults. The brain age prediction accuracy of the ANN (R 2 = 0.47) was similar to established multimodal models. The comparison of the ANN-based age-corrected FA with the tract-wise linear age-corrected FA resulted in an R 2 value of 0.90 [0.82; 0.93] and a mean difference of 0.00 [-0.04; 0.05] for all tract systems combined. In conclusion, this study demonstrated the applicability of complex ANN models to non-linear age correction of tract-based diffusion metrics as a proof of concept.

Longitudinal monitoring of amyotrophic lateral sclerosis by diffusion tensor imaging: Power calculations for group studies

Introduction

Diffusion tensor imaging (DTI) can be used to map disease progression in amyotrophic lateral sclerosis (ALS) and therefore is a promising candidate for a biomarker in ALS. To this end, longitudinal study protocols need to be optimized and validated regarding group sizes and time intervals between visits. The objective of this study was to assess the influences of sample size, the schedule of follow-up measurements, and measurement uncertainties on the statistical power to optimize longitudinal DTI study protocols in ALS.

Patients and methods

To estimate the measurement uncertainty of a tract-of–interest-based DTI approach, longitudinal test-retest measurements were applied first to a normal data set. Then, DTI data sets of 80 patients with ALS and 50 healthy participants were analyzed in the simulation of longitudinal trajectories, that is, longitudinal fractional anisotropy (FA) values for follow-up sessions were simulated for synthetic patient and control groups with different rates of FA decrease in the corticospinal tract. Monte Carlo simulations of synthetic longitudinal study groups were used to estimate the statistical power and thus the potentially needed sample sizes for a various number of scans at one visit, different time intervals between baseline and follow-up measurements, and measurement uncertainties.

Results

From the simulation for different longitudinal FA decrease rates, it was found that two scans per session increased the statistical power in the investigated settings unless sample sizes were sufficiently large and time intervals were appropriately long. The positive effect of a second scan per session on the statistical power was particularly pronounced for FA values with high measurement uncertainty, for which the third scan per session increased the statistical power even further.

Conclusion

With more than one scan per session, the statistical power of longitudinal DTI studies can be increased in patients with ALS. Consequently, sufficient statistical power can be achieved even with limited sample sizes. An improved longitudinal DTI study protocol contributes to the detection of small changes in diffusion metrics and thereby supports DTI as an applicable and reliable non-invasive biomarker in ALS.

Structural magnetic resonance imaging findings and histopathological correlations in motor neuron diseases—A systematic review and meta-analysis

Objectives

The lack of systematic evidence on neuroimaging findings in motor neuron diseases (MND) hampers the diagnostic utility of magnetic resonance imaging (MRI). Thus, we aimed at performing a systematic review and meta-analysis of MRI features in MND including their histopathological correlation.

Methods

In a comprehensive literature search, out of 5941 unique publications, 223 records assessing brain and spinal cord MRI findings in MND were eligible for a qualitative synthesis. 21 records were included in a random effect model meta-analysis.

Results

Our meta-analysis shows that both T2-hyperintensities along the corticospinal tracts (CST) and motor cortex T2*-hypointensitites, also called “motor band sign”, are more prevalent in ALS patients compared to controls [OR 2.21 (95%-CI: 1.40–3.49) and 10.85 (95%-CI: 3.74–31.44), respectively]. These two imaging findings correlate to focal axonal degeneration/myelin pallor or glial iron deposition on histopathology, respectively. Additionally, certain clinical MND phenotypes such as amyotrophic lateral sclerosis (ALS) seem to present with distinct CNS atrophy patterns.

Conclusions

Although CST T2-hyperintensities and the “motor band sign” are non-specific imaging features, they can be leveraged for diagnostic workup of suspected MND cases, together with certain brain atrophy patterns. Collectively, this study provides high-grade evidence for the usefulness of MRI in the diagnostic workup of suspected MND cases.

Systematic review registration

https://www.crd.york.ac.uk/PROSPERO/, identifier: CRD42020182682.

Simultaneous PET/MRI: The future gold standard for characterizing motor neuron disease-A clinico-radiological and neuroscientific perspective

Neuroimaging assessment of motor neuron disease has turned into a cornerstone of its clinical workup. Amyotrophic lateral sclerosis (ALS), as a paradigmatic motor neuron disease, has been extensively studied by advanced neuroimaging methods, including molecular imaging by MRI and PET, furthering finer and more specific details of the cascade of ALS neurodegeneration and symptoms, facilitated by multicentric studies implementing novel methodologies. With an increase in multimodal neuroimaging data on ALS and an exponential improvement in neuroimaging technology, the need for harmonization of protocols and integration of their respective findings into a consistent model becomes mandatory. Integration of multimodal data into a model of a continuing cascade of functional loss also calls for the best attempt to correlate the different molecular imaging measurements as performed at the shortest inter-modality time intervals possible. As outlined in this perspective article, simultaneous PET/MRI, nowadays available at many neuroimaging research sites, offers the perspective of a one-stop shop for reproducible imaging biomarkers on neuronal damage and has the potential to become the new gold standard for characterizing motor neuron disease from the clinico-radiological and neuroscientific perspectives.

Relaxation-weighted 23Na magnetic resonance imaging maps regional patterns of abnormal sodium concentrations in amyotrophic lateral sclerosis

Objectives:

Multiparametric magnetic resonance imaging (MRI) is established as a technical instrument for the characterisation of patients with amyotrophic lateral sclerosis (ALS). The contribution of relaxation-weighted sodium (²³NaR) MRI remains to be defined. The aim of this study is to apply ²³NaR MRI to investigate brain sodium homeostasis and map potential alterations in patients with ALS as compared with healthy controls.

Materials and Methods:

Seventeen patients with ALS (mean age 61.1 ± 11.4 years, m/f = 9/8) and 10 healthy control subjects (mean age 60.3 ± 15.3 years, m/f = 6/4) were examined by ²³NaR MRI at 3 T. Regional sodium maps were obtained by the calculation of the weighted difference from two image data sets with different echo times (TE₁ = 0.3 ms, TE₂ = 25 ms). Voxel-based analysis of the relaxation-weighted maps, together with ²³Na concentration maps for comparison, was performed.

Results:

ROI-based analyses of relaxation-weighted brain sodium concentration maps demonstrated increased sodium concentrations in the upper corticospinal tracts and in the frontal lobes in patients with ALS; no differences between ALS patients and controls were found in reference ROIs, where no involvement in ALS-associated neurodegeneration could be anticipated.

Conclusion:

²³NaR MRI mapped regional alterations within disease-relevant areas in ALS which correspond to the stages of the central nervous system (CNS) pathology, providing evidence that the technique is a potential biological marker of the cerebral neurodegenerative process in ALS.

Multiparametric Microstructural MRI and Machine Learning Classification Yields High Diagnostic Accuracy in Amyotrophic Lateral Sclerosis: Proof of Concept

The potential of multiparametric quantitative neuroimaging has been extensively discussed as a diagnostic tool in amyotrophic lateral sclerosis (ALS). In the past, the integration of multimodal, quantitative data into a useful diagnostic classifier was a major challenge. With recent advances in the field, machine learning in a data driven approach is a potential solution: neuroimaging biomarkers in ALS are mainly observed in the cerebral microstructure, with diffusion tensor imaging (DTI) and texture analysis as promising approaches. We set out to combine these neuroimaging markers as age-corrected features in a machine learning model with a cohort of 502 subjects, divided into 404 patients with ALS and 98 healthy controls. We calculated a linear support vector classifier (SVC) which is a very robust model and then verified the results with a multilayer perceptron (MLP)/neural network. Both classifiers were able to separate ALS patients from controls with receiver operating characteristic (ROC) curves showing an area under the curve (AUC) of 0.87-0.88 ("good") for the SVC and 0.88-0.91 ("good" to "excellent") for the MLP. Among the coefficients of the SVC, texture data contributed the most to a correct classification. We consider these results as a proof of concept that demonstrated the power of machine learning in the application of multiparametric quantitative neuroimaging data to ALS.