BRinging Artificial INTelligencE home for a better cAre of amyotrophic lateral sclerosis and multiple SclERosis

Financiador

Organização

Publicações

Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis

Publication . Soares, Diogo F.; Henriques, Rui; Gromicho, Marta; Carvalho, Mamede; Madeira, Sara C.

Longitudinal cohort studies to study disease progression generally combine temporal features produced under periodic assessments (clinical follow-up) with static features associated with single-time assessments, genetic, psychophysiological, and demographic profiles. Subspace clustering, including biclustering and triclustering stances, enables the discovery of local and discriminative patterns from such multidimensional cohort data. These patterns, highly interpretable, are relevant to identifying groups of patients with similar traits or progression patterns. Despite their potential, their use for improving predictive tasks in clinical domains remains unexplored. In this work, we propose to learn predictive models from static and temporal data using discriminative patterns, obtained via biclustering and triclustering, as features within a state-of-the-art classifier, thus enhancing model interpretation. triCluster is extended to find time-contiguous triclusters in temporal data (temporal patterns) and a biclustering algorithm to discover coherent patterns in static data. The transformed data space, composed of bicluster and tricluster features, capture local and cross-variable associations with discriminative power, yielding unique statistical properties of interest. As a case study, we applied our methodology to follow-up data from Portuguese patients with Amyotrophic Lateral Sclerosis (ALS) to predict the need for non-invasive ventilation (NIV) since the last appointment. The results showed that, in general, our methodology outperformed baseline results using the original features. Furthermore, the bicluster/tricluster-based patterns used by the classifier can be used by clinicians to understand the models by highlighting relevant prognostic patterns.

2022Artigo científico

Acesso aberto

Ver mais

Motor neuron disease in three asymptomatic pVal50Met TTR gene carriers

Publication . Santos Silva, Cláudia; Oliveira Santos, Miguel; Gromicho, Marta; Pronto Laborinho, Ana Catarina; Conceição, isabel; Carvalho, Mamede

Purpose: This study aimed to estimate the impact of risk factors for peri-implant pathology, to identify potentially modifiable factors, and to evaluate the accuracy of the risk algorithm, risk scores and risk stratification. Methods: This retrospective case-control study with 1275 patients (255 cases; 1020 controls) retrieved a model according to the predictors: history of Periodontitis, bacterial plaque, bleeding, bone level, lack of passive fit or non-optimal screw joint, metal-ceramic restoration, proximity to other implants/teeth, and smoking habits. Outcome measures were the attributable fraction; the positive and negative likelihood ratios at different disease cut-off points illustrated by the area under the curve statistic. Results: Six predictors may be modified or controlled directly by either the patient or the clinician, accounting for a reduction in up to 95% of the peri-implant pathology cases. The positive and negative likelihood ratios were 9.69 and 0.13, respectively; the area under the curve was 0.96; a risk score was developed, making the complex statistical model useful to clinicians. Conclusions: Based on the results, six predictors for the incidence of peri-implant pathology can be modified to significantly improve the outcome. It was possible to stratify patients per risk category according to the risk score, providing a tool for clinicians to support their decision-making process.

2022Artigo científico

Acesso restrito

Ver mais

Novel compound heterozygous variants of SPG11 gene associated with young-adult amyotrophic lateral sclerosis

Publication . Santos Silva, Cláudia; Oliveira Santos, Miguel; Madureira, João; Reimão, Sofia; Carvalho, Mamede

SPG11 gene is localized on chromosome 15q21 and encodes spatacsin. Mutations of this gene are typically associated with autosomal recessive hereditary spastic paraplegia (HSP)-11, causing progressive spasticity of the lower limbs. This type of HSP is usually associated with other manifestations, such as thickness reduction of corpus callosum, cognitive impairment, peripheral neuropathy, pseudobulbar involvement. Nevertheless, SPG11 pathogenic variants are also associated with a spectrum of clinical manifestations, including juvenile amyotrophic lateral sclerosis (ALS), hereditary motor sensory neuropathy (Charcot–Marie–Tooth disease type 2X), and multiple sclerosis mimics.

2022Artigo científico

Acesso restrito

Ver mais

Entidade financiadora

European Commission

Programa de financiamento

H2020

Número da atribuição

101017598