Manifold learning for amyotrophic lateral sclerosis functional loss assessment

Abstract

Amyotrophic lateral sclerosis (ALS) is an inexorably progressive neurodegenerative condition with no effective diseasemodifying therapy at present. Given the striking clinical heterogeneity of the condition, the development and validation of reliable prognostic models is a recognised research priority. We present a prognostic model for functional decline in ALS where outcome uncertainty is taken into account. Patient data were reduced and projected onto a 2D space using Uniform Manifold Approximation and Projection (UMAP), a novel non-linear dimension reduction technique. Information from 3756 patients was included. Development data were sourced from past clinical trials. Real-world population data were used as validation data. Predictors included age, gender, region of onset, symptom duration, weight at baseline, functional impairment, and estimated rate of functional loss. UMAP projection of patients showed an informative 2D data distribution. As limited data availability precluded complex model designs, the projection was divided into three zones defined by a functional impairment range probability. Zone membership allowed individual patient prediction. Patients belonging to the first zone had a probability of 83% (± 3% ) to have an ALSFRS score over 20 at 1-year follow-up. Patients within the second zone had a probability of 89% (± 4% ) to have an ALSFRS score between 10 and 30 at 1 year follow-up. Finally, patients within the third zone had a probability of 88% (± 7% ) to have an ALSFRS score lower than 20 at 1 year follow-up. This approach requires a limited set of features, is easily updated, improves with additional patient data, and accounts for results uncertainty. This method could therefore be used in a clinical setting for patient stratification and outcome projection.

Clusters of anatomical disease-burden patterns in ALS: a data-driven approach confirms radiological subtypes

“…Certain anatomical areas such as the parietal lobes and occipital lobe may not be characteristic regions of degeneration, yet, as illustrated, may have a role in segregating specific ALS subtypes. This observation is consistent with the emerging machine-learning literature of ALS [ 52 , 53 ] which suggests that feature importance analyses, especially in multi-class classification schemes, may identify brain regions which are not classically associated with ALS [ 54 , 55 ].…”

Mapping cortical disease-burden at individual-level in frontotemporal dementia: implications for clinical care and pharmacological trials

“…Moreover, only grey matter analyses were conducted, despite the contribution of white matter pathology to the clinical manifestations of these phenotypes (Bede et al, 2016 , 2018a , b ; Qin et al, 2021 ; Schuster et al, 2016a , b ; Zhou et al, 2010 ). Finally, while our approach provides individualised atrophy maps, supervised and unsupervised machine learning approaches offer direct individual patient categorisation into diagnostic and prognostic groups (Bede et al, 2017 ; Grollemund et al, 2020a , b ; Grollemund et al, 2020a , b ; Querin et al, 2018 ; Schuster et al, 2016a , b ).…”

Pathological neural networks and artificial neural networks in ALS: diagnostic classification based on pathognomonic neuroimaging features

“…As opposed to the scholarly pursuit of group-level descriptions, the priority of clinical neurology is the precision classification of a specific, single patient into diagnostic, phenotypic and prognostic categories through the quantitative interpretation of their biomarker profile. Relatively few studies have focussed on the classification of individual patient imaging data in ALS [17,18]. A variety of innovative approaches have been explored [19] spanning from z-score based approaches, through support vector machine frameworks, discriminant function analyses, to regression models, with varying degree of classification accuracy [16,[20][21][22][23][24].…”