RESUMEN
In machine learning, data often comes from different sources, but combining them can introduce extraneous variation that affects both generalization and interpretability. For example, we investigate the classification of neurodegenerative diseases using FDG-PET data collected from multiple neuroimaging centers. However, data collected at different centers introduces unwanted variation due to differences in scanners, scanning protocols, and processing methods. To address this issue, we propose a two-step approach to limit the influence of center-dependent variation on the classification of healthy controls and early vs. late-stage Parkinson's disease patients. First, we train a Generalized Matrix Learning Vector Quantization (GMLVQ) model on healthy control data to identify a "relevance space" that distinguishes between centers. Second, we use this space to construct a correction matrix that restricts a second GMLVQ system's training on the diagnostic problem. We evaluate the effectiveness of this approach on the real-world multi-center datasets and simulated artificial dataset. Our results demonstrate that the approach produces machine learning systems with reduced bias - being more specific due to eliminating information related to center differences during the training process - and more informative relevance profiles that can be interpreted by medical experts. This method can be adapted to similar problems outside the neuroimaging domain, as long as an appropriate "relevance space" can be identified to construct the correction matrix.
Asunto(s)
Neuroimagen , Enfermedad de Parkinson , Humanos , Tomografía de Emisión de Positrones , Aprendizaje Automático , Enfermedad de Parkinson/diagnóstico por imagenRESUMEN
INTRODUCTION: Non-alcoholic fatty liver disease (NAFLD) affects approximately one in four individuals and its prevalence continues to rise. The advanced stages of NAFLD with significant liver fibrosis are associated with adverse morbidity and mortality outcomes. Currently, liver biopsy remains the 'gold-standard' approach to stage NAFLD severity. Although generally well tolerated, liver biopsies are associated with significant complications, are resource intensive, costly, and sample only a very small area of the liver as well as requiring day case admission to a secondary care setting. As a result, there is a significant unmet need to develop non-invasive biomarkers that can accurately stage NAFLD and limit the need for liver biopsy. The aim of this study is to validate the use of the urine steroid metabolome as a strategy to stage NAFLD severity and to compare its performance against other non-invasive NAFLD biomarkers. METHODS AND ANALYSIS: The TrUSt-NAFLD study is a multicentre prospective test validation study aiming to recruit 310 patients with biopsy-proven and staged NAFLD across eight centres within the UK. 150 appropriately matched control patients without liver disease will be recruited through the Oxford Biobank. Blood and urine samples, alongside clinical data, will be collected from all participants. Urine samples will be analysed by liquid chromatography-tandem mass spectroscopy to quantify a panel of predefined steroid metabolites. A machine learning-based classifier, for example, Generalized Matrix Relevance Learning Vector Quantization that was trained on retrospective samples, will be applied to the prospective steroid metabolite data to determine its ability to identify those patients with advanced, as opposed to mild-moderate, liver fibrosis as a consequence of NAFLD. ETHICS AND DISSEMINATION: Research ethical approval was granted by West Midlands, Black Country Research Ethics Committee (REC reference: 21/WM/0177). A substantial amendment (TrUSt-NAFLD-SA1) was approved on 26 November 2021. TRIAL REGISTRATION NUMBER: ISRCTN19370855.
Asunto(s)
Enfermedad del Hígado Graso no Alcohólico , Humanos , Biomarcadores , Biopsia/efectos adversos , Hígado/patología , Cirrosis Hepática/diagnóstico , Metaboloma , Estudios Multicéntricos como Asunto , Enfermedad del Hígado Graso no Alcohólico/epidemiología , Estudios Retrospectivos , Esteroides , Estudios de Validación como AsuntoRESUMEN
Primary aldosteronism (PA) causes 5-10% of hypertension cases, but only a minority of patients are currently diagnosed and treated because of a complex, stepwise, and partly invasive workup. We tested the performance of urine steroid metabolomics, the computational analysis of 24-hour urine steroid metabolome data by machine learning, for the identification and subtyping of PA. Mass spectrometry-based multi-steroid profiling was used to quantify the excretion of 34 steroid metabolites in 24-hour urine samples from 158 adults with PA (88 with unilateral PA [UPA] due to aldosterone-producing adenomas [APAs]; 70 with bilateral PA [BPA]) and 65 sex- and age-matched healthy controls. All APAs were resected and underwent targeted gene sequencing to detect somatic mutations associated with UPA. Patients with PA had increased urinary metabolite excretion of mineralocorticoids, glucocorticoids, and glucocorticoid precursors. Urine steroid metabolomics identified patients with PA with high accuracy, both when applied to all 34 or only the three most discriminative steroid metabolites (average areas under the receiver-operating characteristics curve [AUCs-ROC] 0.95-0.97). Whilst machine learning was suboptimal in differentiating UPA from BPA (average AUCs-ROC 0.65-0.73), it readily identified APA cases harbouring somatic KCNJ5 mutations (average AUCs-ROC 0.79-85). These patients showed a distinctly increased urine excretion of the hybrid steroid 18-hydroxycortisol and its metabolite 18-oxo-tetrahydrocortisol, the latter identified by machine learning as by far the most discriminative steroid. In conclusion, urine steroid metabolomics is a non-invasive candidate test for the accurate identification of PA cases and KCNJ5-mutated APAs.