CNN-Based Neurodegenerative Disease Classification Using QR-Represented Gait Data.

PURPOSE: The primary aim of this study is to develop an effective and reliable diagnostic system for neurodegenerative diseases by utilizing gait data transformed into QR codes and classified using convolutional neural networks (CNNs). The objective of this method is to enhance the precision of diagnosing neurodegenerative diseases, including amyotrophic lateral sclerosis (ALS), Parkinson's disease (PD), and Huntington's disease (HD), through the introduction of a novel approach to analyze gait patterns. METHODS: The research evaluates the CNN-based classification approach using QR-represented gait data to address the diagnostic challenges associated with neurodegenerative diseases. The gait data of subjects were converted into QR codes, which were then classified using a CNN deep learning model. The dataset includes recordings from patients with Parkinson's disease (n = 15), Huntington's disease (n = 20), and amyotrophic lateral sclerosis (n = 13), and from 16 healthy controls. RESULTS: The accuracy rates obtained through 10-fold cross-validation were as follows: 94.86% for NDD versus control, 95.81% for PD versus control, 93.56% for HD versus control, 97.65% for ALS versus control, and 84.65% for PD versus HD versus ALS versus control. These results demonstrate the potential of the proposed system in distinguishing between different neurodegenerative diseases and control groups. CONCLUSION: The results indicate that the designed system may serve as a complementary tool for the diagnosis of neurodegenerative diseases, particularly in individuals who already present with varying degrees of motor impairment. Further validation and research are needed to establish its wider applicability.

Overview of nomenclature and diagnosis of amyotrophic lateral sclerosis.

The nomenclature of amyotrophic lateral sclerosis (ALS) currently is blurred, indistinct and no accurate and haven't been properly updated since the first description, which is far from being suitable for the current implementation of clinical practise and scientific research of ALS, and urgently need an solution. Furthermore, the current diagnostic criteria need also further been improved, because the current clinical diagnosis of ALS majorly depends on the clinical manifestations yet. Up to now, no any objective clinical auxiliary examination can be helpful to diagnose ALS besides the electromyogram identifying the lower motor neuron damage, which isn't conducive to early diagnosis and prolongs the time of ALS confirmed diagnosis. In this mini review, we discussed the current doubt about the nomenclature and diagnostic criteria of ALS, and prospected in order to further improve and normalize the nomenclature and diagnosis of ALS.

Simultaneous time-frequency analysis of gait signals of both legs in classifying neurodegenerative diseases.

BACKGROUND: Neurodegenerative diseases (NDDs) pose significant challenges due to their debilitating nature and limited therapeutic options. Accurate and timely diagnosis is crucial for optimizing patient care and treatment strategies. Gait analysis, utilizing wearable sensors, has shown promise in assessing motor abnormalities associated with NDDs. RESEARCH QUESTION: Research Question 1 To what extent can analyzing the interaction of both limbs in the time-frequency domain serve as a suitable methodology for accurately classifying NDDs? Research Question 2 How effective is the utilization of color-coded images, in conjunction with deep transfer learning models, for the classification of NDDs? METHODS: GaitNDD database was used, comprising recordings from patients with Huntington's disease, amyotrophic lateral sclerosis, Parkinson's disease, and healthy controls. The gait signals underwent signal preparation, wavelet coherence analysis, and principal component analysis for feature enhancement. Deep transfer learning models (AlexNet, GoogLeNet, SqueezeNet) were employed for classification. Performance metrics, including accuracy, sensitivity, specificity, precision, and F1 score, were evaluated using 5-fold cross-validation. RESULTS: The classification performance of the models varied depending on the time window used. For 5-second gait signal segments, AlexNet achieved an accuracy of 95.91 %, while GoogLeNet and SqueezeNet achieved accuracies of 96.49 % and 92.73 %, respectively. For 10-second segments, AlexNet outperformed other models with an accuracy of 99.20 %, while GoogLeNet and SqueezeNet achieved accuracies of 96.75 % and 95.00 %, respectively. Statistical tests confirmed the significance of the extracted features, indicating their discriminative power for classification. SIGNIFICANCE: The proposed method demonstrated superior performance compared to previous studies, offering a non-invasive and cost-effective approach for the automated diagnosis of NDDs. By analyzing the interaction between both legs during walking using wavelet coherence, and utilizing deep transfer learning models, accurate classification of NDDs was achieved.

Electrodiagnostic subtyping in Guillain-Barré syndrome patients in the International Guillain-Barré Outcome Study.

BACKGROUND AND PURPOSE: Various electrodiagnostic criteria have been developed in Guillain-Barré syndrome (GBS). Their performance in a broad representation of GBS patients has not been evaluated. Motor conduction data from the International GBS Outcome Study (IGOS) cohort were used to compare two widely used criterion sets and relate these to diagnostic amyotrophic lateral sclerosis criteria. METHODS: From the first 1500 patients in IGOS, nerve conduction studies from 1137 (75.8%) were available for the current study. These patients were classified according to nerve conduction studies criteria proposed by Hadden and Rajabally. RESULTS: Of the 1137 studies, 68.3% (N = 777) were classified identically according to criteria by Hadden and Rajabally: 111 (9.8%) axonal, 366 (32.2%) demyelinating, 195 (17.2%) equivocal, 35 (3.1%) inexcitable and 70 (6.2%) normal. Thus, 360 studies (31.7%) were classified differently. The areas of differences were as follows: 155 studies (13.6%) classified as demyelinating by Hadden and axonal by Rajabally; 122 studies (10.7%) classified as demyelinating by Hadden and equivocal by Rajabally; and 75 studies (6.6%) classified as equivocal by Hadden and axonal by Rajabally. Due to more strictly defined cutoffs fewer patients fulfilled demyelinating criteria by Rajabally than by Hadden, making more patients eligible for axonal or equivocal classification by Rajabally. In 234 (68.6%) axonal studies by Rajabally the revised El Escorial (amyotrophic lateral sclerosis) criteria were fulfilled; in axonal cases by Hadden this was 1.8%. CONCLUSIONS AND DISCUSSION: This study shows that electrodiagnosis in GBS is dependent on the criterion set utilized, both of which are based on expert opinion. Reappraisal of electrodiagnostic subtyping in GBS is warranted.

A specific gene-splicing alteration in the SNRNP70 gene as a hallmark of an ALS subtype.

Amyotrophic lateral sclerosis (ALS) has been considered as one of the progressive neurodegenerative diseases. Numerous genetic factors in divergent molecular pathways have been identified as causative factors of ALS. However, the underlying molecular mechanism that causes this disease remains undetermined; as a result, this has driven the search to find consensus disease-specific hallmarks. In this study, we focused on the alteration of the ratio of two specific gene-splicing events in the SNRNP70 gene from RNA-seq data derived from patients with ALS and control subjects. The splicing profile was significantly and specifically changed in one previously identified ALS subtype. Conversely, the gene expression profile of other ALS cases containing a splicing alteration in the SNRNP70 gene was similar to that of the subtype, whereas ALS cases without this change have exhibited less similarity. These results indicate that this splicing event in the SNRNP70 gene could represent a novel and broadly applicable molecular hallmark of a subtype of ALS.

Classification of amyotrophic lateral sclerosis by brain volume, connectivity, and network dynamics.

Emerging studies corroborate the importance of neuroimaging biomarkers and machine learning to improve diagnostic classification of amyotrophic lateral sclerosis (ALS). While most studies focus on structural data, recent studies assessing functional connectivity between brain regions by linear methods highlight the role of brain function. These studies have yet to be combined with brain structure and nonlinear functional features. We investigate the role of linear and nonlinear functional brain features, and the benefit of combining brain structure and function for ALS classification. ALS patients (N = 97) and healthy controls (N = 59) underwent structural and functional resting state magnetic resonance imaging. Based on key hubs of resting state networks, we defined three feature sets comprising brain volume, resting state functional connectivity (rsFC), as well as (nonlinear) resting state dynamics assessed via recurrent neural networks. Unimodal and multimodal random forest classifiers were built to classify ALS. Out-of-sample prediction errors were assessed via five-fold cross-validation. Unimodal classifiers achieved a classification accuracy of 56.35-61.66%. Multimodal classifiers outperformed unimodal classifiers achieving accuracies of 62.85-66.82%. Evaluating the ranking of individual features' importance scores across all classifiers revealed that rsFC features were most dominant in classification. While univariate analyses revealed reduced rsFC in ALS patients, functional features more generally indicated deficits in information integration across resting state brain networks in ALS. The present work undermines that combining brain structure and function provides an additional benefit to diagnostic classification, as indicated by multimodal classifiers, while emphasizing the importance of capturing both linear and nonlinear functional brain properties to identify discriminative biomarkers of ALS.

Molecular Classification and Interpretation of Amyotrophic Lateral Sclerosis Using Deep Convolution Neural Networks and Shapley Values.

Amyotrophic lateral sclerosis (ALS) is a prototypical neurodegenerative disease characterized by progressive degeneration of motor neurons to severely effect the functionality to control voluntary muscle movement. Most of the non-additive genetic aberrations responsible for ALS make its molecular classification very challenging along with limited sample size, curse of dimensionality, class imbalance and noise in the data. Deep learning methods have been successful in many other related areas but have low minority class accuracy and suffer from the lack of explainability when used directly with RNA expression features for ALS molecular classification. In this paper, we propose a deep-learning-based molecular ALS classification and interpretation framework. Our framework is based on training a convolution neural network (CNN) on images obtained from converting RNA expression values into pixels based on DeepInsight similarity technique. Then, we employed Shapley additive explanations (SHAP) to extract pixels with higher relevance to ALS classifications. These pixels were mapped back to the genes which made them up. This enabled us to classify ALS samples with high accuracy for a minority class along with identifying genes that might be playing an important role in ALS molecular classifications. Taken together with RNA expression images classified with CNN, our preliminary analysis of the genes identified by SHAP interpretation demonstrate the value of utilizing Machine Learning to perform molecular classification of ALS and uncover disease-associated genes.

Genome Wide Analysis Points towards Subtype-Specific Diseases in Different Genetic Forms of Amyotrophic Lateral Sclerosis.

Amyotropic lateral sclerosis (ALS) is a lethally progressive and irreversible neurodegenerative disease marked by apparent death of motor neurons present in the spinal cord, brain stem and motor cortex. While more and more gene mutants being established for genetic ALS, the vast majority suffer from sporadic ALS (>90%). It has been challenging, thus, to model sporadic ALS which is one reason why the underlying pathophysiology remains elusive and has stalled the development of therapeutic strategies of this progressive motor neuron disease. To further unravel these pathological signaling pathways, human induced pluripotent stem cell (hiPSCs)-derived motor neurons (MNs) from FUS- and SOD1 ALS patients and healthy controls were systematically compared to independent published datasets. Here through this study we created a gene profile of ALS by analyzing the DEGs, the Kyoto encyclopedia of Genes and Genomes (KEGG) pathways, the interactome and the transcription factor profiles (TF) that would identify altered molecular/functional signatures and their interactions at both transcriptional (mRNAs) and translational levels (hub proteins and TFs). Our findings suggest that FUS and SOD1 may develop from dysregulation in several unique pathways and herpes simplex virus (HSV) infection was among the topmost predominant cellular pathways connected to FUS and not to SOD1. In contrast, SOD1 is mainly characterized by alterations in the metabolic pathways and alterations in the neuroactive-ligand-receptor interactions. This suggests that different genetic ALS forms are singular diseases rather than part of a common spectrum. This is important for patient stratification clearly pointing towards the need for individualized medicine approaches in ALS.

High-Throughput Genetic Testing in ALS: The Challenging Path of Variant Classification Considering the ACMG Guidelines.

The development of high-throughput sequencing technologies and screening of big patient cohorts with familial and sporadic amyotrophic lateral sclerosis (ALS) led to the identification of a significant number of genetic variants, which are sometimes difficult to interpret. The American College of Medical Genetics and Genomics (ACMG) provided guidelines to help molecular geneticists and pathologists to interpret variants found in laboratory testing. We assessed the application of the ACMG criteria to ALS-related variants, combining data from literature with our experience. We analyzed a cohort of 498 ALS patients using massive parallel sequencing of ALS-associated genes and identified 280 variants with a minor allele frequency < 1%. Examining all variants using the ACMG criteria, thus considering the type of variant, inheritance, familial segregation, and possible functional studies, we classified 20 variants as "pathogenic". In conclusion, ALS's genetic complexity, such as oligogenic inheritance, presence of genes acting as risk factors, and reduced penetrance, needs to be considered when interpreting variants. The goal of this work is to provide helpful suggestions to geneticists and clinicians dealing with ALS.

Is it accurate to classify ALS as a neuromuscular disorder?

INTRODUCTION: Amyotrophic lateral sclerosis (ALS) is a fatal disorder characterized by the progressive loss of upper and lower motor neurons. ALS has traditionally been classified within the domain of neuromuscular diseases, which are a unique spectrum of disorders that predominantly affect the peripheral nervous system. However, over the past decades compounding evidence has emerged that there is extensive involvement of the central nervous system. Therefore, one can question whether it remains accurate to classify ALS as a neuromuscular disorder. AREAS COVERED: In this review, the authors sought to discuss current approaches toward disease classification and how we should classify ALS based on novel insights from clinical, imaging, pathophysiological, neuropathological and genetic studies. EXPERT OPINION: ALS exhibits the cardinal features of a neurodegenerative disease. Therefore, classifying ALS as a neuromuscular disease in the strict sense has become untenable. Diagnosing ALS however does require significant neuromuscular expertise and therefore neuromuscular specialists remain best equipped to evaluate this category of patients. Designating motor neuron diseases as a separate category in the ICD-11 is justified and adequately deals with this issue. However, to drive effective therapy development the fields of motor neuron disease and neurodegenerative disorders must come together.

Gait Rhythm Dynamics for Neuro-Degenerative Disease Classification via Persistence Landscape- Based Topological Representation.

Neuro-degenerative disease is a common progressive nervous system disorder that leads to serious clinical consequences. Gait rhythm dynamics analysis is essential for evaluating clinical states and improving quality of life for neuro-degenerative patients. The magnitude of stride-to-stride fluctuations and corresponding changes over time-gait dynamics-reflects the physiology of gait, in quantifying the pathologic alterations in the locomotor control system of health subjects and patients with neuro-degenerative diseases. Motivated by algebra topology theory, a topological data analysis-inspired nonlinear framework was adopted in the study of the gait dynamics. Meanwhile, the topological representation-persistence landscapes were used as input of classifiers in order to distinguish different neuro-degenerative disease type from healthy. In this work, stride-to-stride time series from healthy control (HC) subjects are compared with the gait dynamics from patients with amyotrophic lateral sclerosis (ALS), Huntington's disease (HD), and Parkinson's disease (PD). The obtained results show that the proposed methodology discriminates healthy subjects from subjects with other neuro-degenerative diseases with relatively high accuracy. In summary, our study is the first attempt to provide a topological representation-based method into the disease classification with gait rhythms measured from the stride intervals to visualize gait dynamics and classify neuro-degenerative diseases. The proposed method could be potentially used in earlier interventions and state monitoring.

Inflammatory profiles relate to survival in subtypes of amyotrophic lateral sclerosis.

OBJECTIVE: To investigate inflammatory cytokines in patients with motor neuron disease (MND) evaluating the putative contribution of amyotrophic lateral sclerosis (ALS)-causing gene variants. METHODS: This study is a retrospective case series with prospective follow-up (1994-2016) of 248 patients with MND, of whom 164 had ALS who were screened for mutations in the genes for SOD1 and C9orf72. Paired CSF and plasma were collected at the diagnostic evaluation before treatment. A panel of cytokines were measured blindly via digital ELISA on the Simoa platform. RESULTS: Time from disease onset to death was longer for patients with ALS-causing SOD1 mutations (mSOD1, n = 24) than those with C9orf72 hexanucleotide repeat expansion (C9orf72HRE) ALS (n = 19; q = 0.001) and other ALS (OALS) (n = 119; q = 0.0008). Patients with OALS had higher CSF tumor necrosis factor alpha (TNF-α) compared with those with C9orf72HRE ALS (q = 0.014). Patients with C9orf72HRE ALS had higher CSF interferon alpha compared with those with OALS and mSOD1 ALS (q = 0.042 and q = 0.042). In patients with ALS, the survival was negatively correlated with plasma interleukin (IL) 10 (hazard ratio [HR] 1.17, 95% CI 1.05-1.30). Plasma TNF-α, IL-10, and TNF-related apoptosis-inducing ligand (TRAIL) (HR 1.01 [1.00-1.02], 1.15 [1.02-1.30], and 1.01 [1.00-1.01], respectively) of patients with OALS, plasma IL-1ß (HR 5.90 [1.27-27.5]) of patients with C9orf72HRE ALS, and CSF TRAIL (10.5 [1.12-98.6]) of patients with mSOD1 ALS all correlated negatively with survival. CONCLUSIONS: Differences in survival times in ALS subtypes were correlated with cytokine levels, suggesting specific immune responses related to ALS genetic variants.

ALS phenotype is influenced by age, sex, and genetics: A population-based study.

OBJECTIVE: To assess the determinants of amyotrophic lateral sclerosis (ALS) phenotypes in a population-based cohort. METHODS: The study population included 2,839 patients with ALS diagnosed in Piemonte, Italy (1995-2015). Patients were classified according to motor (classic, bulbar, flail arm, flail leg, predominantly upper motor neuron [PUMN], respiratory) and cognitive phenotypes (normal, ALS with cognitive impairment [ALSci], ALS with behavioral impairment [ALSbi], ALSci and ALSbi combined [ALScbi], ALS-frontotemporal dementia [FTD]). Binary logistic regression analysis was adjusted for sex, age, and genetics. RESULTS: Bulbar phenotype correlated with older age (p < 0.0001), women were more affected than men at increasing age (p < 0.0001), classic with younger age (p = 0.029), men were more affected than women at increasing age (p < 0.0001), PUMN with younger age (p < 0.0001), flail arm with male sex (p < 0.0001) and younger age (p = 0.04), flail leg with male sex with increasing age (p = 0.008), and respiratory with male sex (p < 0.0001). C9orf72 expansions correlated with bulbar phenotype (p < 0.0001), and were less frequent in PUMN (p = 0.041); SOD1 mutations correlated with flail leg phenotype (p < 0.0001), and were less frequent in bulbar (p < 0.0001). ALS-FTD correlated with C9orf72 (p < 0.0001) and bulbar phenotype (p = 0.008), ALScbi with PUMN (p = 0.014), and ALSci with older age (p = 0.008). CONCLUSIONS: Our data suggest that the spatial-temporal combination of motor and cognitive events leading to the onset and progression of ALS is characterized by a differential susceptibility to the pathologic process of motor and prefrontal cortices and lower motor neurons, and is influenced by age, sex, and gene variants. The identification of those factors that regulate ALS phenotype will allow us to reclassify patients into pathologically homogenous subgroups, responsive to targeted personalized therapies.

Texture classification of MR images of the brain in ALS using M-CoHOG: A multi-center study.

Gradient-based texture analysis methods have become popular in computer vision and image processing and has many applications including medical image analysis. This motivates us to develop a texture feature extraction method to discriminate Amyotrophic Lateral Sclerosis (ALS) patients from controls. But, the lack of data in ALS research is a major constraint and can be mitigated by using data from multiple centers. However, multi-center data gives some other challenges such as differing scanner parameters and variation in intensity of the medical images, which motivate the development of the proposed method. To investigate these challenges, we propose a gradient-based texture feature extraction method called Modified Co-occurrence Histograms of Oriented Gradients (M-CoHOG) to extract texture features from 2D Magnetic Resonance Images (MRI). We also propose a new feature-normalization technique before feeding the normalized M-CoHOG features into an ensemble of classifiers, which can accommodate for variation of data from different centers. ALS datasets from four different centers are used in the experiments. We analyze the classification accuracy of single center data as well as that arising from multiple centers. It is observed that the extracted texture features from downsampled images are more significant in distinguishing between patients and controls. Moreover, using an ensemble of classifiers shows improvement in classification accuracy over a single classifier in multi-center data. The proposed method outperforms the state-of-the-art methods by a significant margin.

Amyotrophic Lateral Sclerosis with Pallidonigroluysian Degeneration: A Clinicopathological Study.

OBJECTIVE: The pallidonigroluysian (PNL) system, the primary component of corticosubcortical circuits, is generally spared in amyotrophic lateral sclerosis (ALS). We evaluated the clinicopathological features of an unusual form of ALS with PNL degeneration (PNLD) and assessed whether ALS with PNLD represents a distinct ALS subtype. METHODS: From a cohort of 97 autopsied cases of sporadic ALS with phosphorylated 43kDa TAR DNA-binding protein (TDP-43) inclusions, we selected those with PNLD and analyzed their clinicopathological features. RESULTS: Eleven cases (11%) that showed PNLD were divided into 2 subtypes depending on the lesion distribution: (1) extensive type (n = 6), showing widespread TDP-43 pathology and multisystem degeneration, both involving the PNL system; and (2) limited type (n = 5), showing selective PNL and motor system involvement, thus being unclassifiable in terms of Brettschneider's staging or Nishihira's typing of ALS. The limited type showed a younger age at onset and predominant PNLD that accounted for the early development of extrapyramidal signs. The limited type exhibited the heaviest pathology in the subthalamus and external globus pallidus, suggesting that TDP-43 inclusions propagated via indirect or hyperdirect pathways, unlike ALS without PNLD, where the direct pathway is considered to convey TDP-43 aggregates from the cerebral cortex to the substantia nigra. INTERPRETATION: The PNL system can be involved in the disease process of ALS, either nonselectively as part of multisystem degeneration, or selectively. ALS with selective involvement of the PNL and motor systems exhibits unique clinicopathological features and TDP-43 propagation routes, thus representing a distinct subtype of ALS. ANN NEUROL 2020;87:302-312.

Postmortem Cortex Samples Identify Distinct Molecular Subtypes of ALS: Retrotransposon Activation, Oxidative Stress, and Activated Glia.

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterized by the progressive loss of motor neurons. While several pathogenic mutations have been identified, the vast majority of ALS cases have no family history of disease. Thus, for most ALS cases, the disease may be a product of multiple pathways contributing to varying degrees in each patient. Using machine learning algorithms, we stratify the transcriptomes of 148 ALS postmortem cortex samples into three distinct molecular subtypes. The largest cluster, identified in 61% of patient samples, displays hallmarks of oxidative and proteotoxic stress. Another 19% of the samples shows predominant signatures of glial activation. Finally, a third group (20%) exhibits high levels of retrotransposon expression and signatures of TARDBP/TDP-43 dysfunction. We further demonstrate that TDP-43 (1) directly binds a subset of retrotransposon transcripts and contributes to their silencing in vitro, and (2) pathological TDP-43 aggregation correlates with retrotransposon de-silencing in vivo.

Validation of the revised classification of cognitive and behavioural impairment in ALS.

OBJECTIVE: In 2017, the diagnostic criteria for cognitive and behavioural impairment in amyotrophic lateral sclerosis (ALS) with frontotemporal dementia (ALSFTD-1) have been modified (ALSFTD-2) with the inclusion of a novel category (ALS with combined cognitive and behavioural impairment, ALScbi) and with changes of operational criteria of the other categories (ALS with cognitive impairment (ALSci), ALS with behavioural impairment (ALSbi) and ALS with frontotemporal dementia (ALS-FTD)). We compared the two sets of criteria to assess the effect of the revised criteria on the cognitive classification of patients with ALS. METHODS: Two cohorts of patients with ALS were included in this study: a population-based cohort including patients identified through the Piemonte/Valle d'Aosta register for ALS in the 2014-2017 period (n=321), and a referral cohort recruited at the Turin ALS centre and at the ALS centre of the Maugeri Institute in Milan in the same period (n=205). Cognitive function was classified in blind by two neuropsychologists expert in ALS. RESULTS: ALSFTD-2 criteria determined a shift of about 15% of patients from their original category to a new one. In both cohorts, about 9% of patients were reclassified to the novel category ALScbi. Among patients previously classified as cognitively normal, 14 (4.3%, population-based cohort) and 19 (9.3%, referral cohort) were reclassified as ALSbi or ALSci. The median survival of the different categories was significantly different with both with sets of criteria. CONCLUSIONS: The new ALSFTD-2 criteria, compared with the old ones, have positive effects on the clinical practice being more sensitive to the early cognitive impairment and having a better prognostic yield.

A standard operating procedure for King's ALS clinical staging.

Objective: Clinical stages in amyotrophic lateral sclerosis (ALS) can be measured using a simple system based on the number of CNS regions involved and requirement for gastrostomy or noninvasive ventilation (NIV). We aimed to design a standard operating procedure (SOP) to define the standardized use and application of the King's staging system. Methods: We designed a SOP for the King's staging system. We wrote case vignettes representative of ALS patients at different disease stages. During two workshops, we taught health care professionals how to use the SOP, then asked them to stage the vignettes using the SOP. We measured the extent to which SOP staging corresponded with correct clinical stage. Results: The reliability of staging using the SOP was excellent, with a Spearman's Rank coefficient of 0.95 (p < 0.001), and was high for different groups of health care professionals, and for those with different levels of experience in ALS. The limits of agreement between SOP staging and actual clinical stage lie within a single stage, confirming that there is a clinically acceptable level of agreement between staging using the SOP and actual King's clinical stage. There were also no systematic biases of the SOP over the range of stages, either for over-staging or under-staging. Conclusions: We have demonstrated that the staging SOP provides a reliable method of calculating clinical stages in ALS patients and can be used prospectively by a range of health care professionals with different levels of experience, as for example may be the case in multicentre clinical trials.

Stratification of amyotrophic lateral sclerosis patients: a crowdsourcing approach.

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease where substantial heterogeneity in clinical presentation urgently requires a better stratification of patients for the development of drug trials and clinical care. In this study we explored stratification through a crowdsourcing approach, the DREAM Prize4Life ALS Stratification Challenge. Using data from >10,000 patients from ALS clinical trials and 1479 patients from community-based patient registers, more than 30 teams developed new approaches for machine learning and clustering, outperforming the best current predictions of disease outcome. We propose a new method to integrate and analyze patient clusters across methods, showing a clear pattern of consistent and clinically relevant sub-groups of patients that also enabled the reliable classification of new patients. Our analyses reveal novel insights in ALS and describe for the first time the potential of a crowdsourcing to uncover hidden patient sub-populations, and to accelerate disease understanding and therapeutic development.

Primary lateral sclerosis: a distinct entity or part of the ALS spectrum?

Primary lateral sclerosis (PLS) has been traditionally viewed as a distinct upper motor neuron condition (UMN) but is increasingly regarded as a sub-phenotype within the amyotrophic lateral sclerosis (ALS) spectrum. Despite established diagnostic criteria, formal diagnosis can be challenging and the protracted diagnostic journey and uncertainty about longer-term prognosis cause considerable distress to patients and caregivers. PLS patients are invariably excluded from ALS clinical trials, while PLS pharmacological trials are lacking. There remains an unmet need for diagnostic biomarkers for upper motor neuron predominant conditions and prognostic indicators regarding prognosis, survival, and risk of conversion to ALS. Validated biomarkers will not only have implications for individualized patient care but also serve as outcome measures in pharmaceutical trials. Given the paucity of post-mortem studies in PLS, novel pathological insights are generally inferred from state-of-the-art imaging studies. Computational neuroimaging has already contributed significantly to the characterization of PLS-associated pathology in vivo and has underscored the role of neuro-inflammation, the presence of extra-motor changes, and confirmed pathological patterns similar to ALS. This systematic review assesses the current state of PLS research across clinical, neuroimaging and neuropathological domains from a combined clinical and academic perspective. We discuss patterns of pathological overlap with other ALS phenotypes, examine if the biological processes of PLS warrant therapeutic strategies distinct from ALS, and evaluate the evidence that classes PLS as a distinct clinico-pathological entity.