Dataset Information

A multidimensional and multi-feature framework for cardiac interoception.

ABSTRACT: Interoception (the sensing of inner-body signals) is a multi-faceted construct with major relevance for basic and clinical neuroscience research. However, the neurocognitive signatures of this domain (cutting across behavioral, electrophysiological, and fMRI connectivity levels) are rarely reported in convergent or systematic fashion. Additionally, various controversies in the field might reflect the caveats of standard interoceptive accuracy (IA) indexes, mainly based on heartbeat detection (HBD) tasks. Here we profit from a novel IA index (md) to provide a convergent multidimensional and multi-feature approach to cardiac interoception. We found that outcomes from our IA-md index are associated with -and predicted by- canonical markers of interoception, including the hd-EEG-derived heart-evoked potential (HEP), fMRI functional connectivity within interoceptive hubs (insular, somatosensory, and frontal networks), and socio-emotional skills. Importantly, these associations proved more robust than those involving current IA indexes. Furthermore, this pattern of results persisted when taking into consideration confounding variables (gender, age, years of education, and executive functioning). This work has relevant theoretical and clinical implications concerning the characterization of cardiac interoception and its assessment in heterogeneous samples, such as those composed of neuropsychiatric patients.

SUBMITTER: Fittipaldi S

PROVIDER: S-EPMC7165068 | biostudies-literature | 2020 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A multidimensional and multi-feature framework for cardiac interoception.

Fittipaldi Sol S Abrevaya Sofía S Fuente Alethia de la A Pascariello Guido Orlando GO Hesse Eugenia E Birba Agustina A Salamone Paula P Hildebrandt Malin M Martí Sofía Alarco SA Pautassi Ricardo Marcos RM Huepe David D Martorell Miquel Martorell MM Yoris Adrián A Roca María M García Adolfo M AM Sedeño Lucas L Ibáñez Agustín A

NeuroImage 20200223

Interoception (the sensing of inner-body signals) is a multi-faceted construct with major relevance for basic and clinical neuroscience research. However, the neurocognitive signatures of this domain (cutting across behavioral, electrophysiological, and fMRI connectivity levels) are rarely reported in convergent or systematic fashion. Additionally, various controversies in the field might reflect the caveats of standard interoceptive accuracy (IA) indexes, mainly based on heartbeat detection (HB ...[more]

PMID: 32101777

Similar Datasets

Project description:BackgroundIdentifying key variables is essential for developing clinical outcome prediction models based on high-dimensional electronic medical records (EMR). However, despite the abundance of feature selection (FS) methods available, challenges remain in choosing the most appropriate method, deciding how many top-ranked variables to include, and ensuring these selections are meaningful from a medical perspective.MethodsWe developed a practical multi-step feature selection (FS) framework that integrates data-driven statistical inference with a knowledge verification strategy. This framework was validated using two distinct EMR datasets targeting different clinical outcomes. The first cohort, sourced from the Medical Information Mart for Intensive Care III (MIMIC-III), focused on predicting acute kidney injury (AKI) in ICU patients. The second cohort, drawn from the MIMIC-IV Emergency Department (MIMIC-IV-ED), aimed to estimate in-hospital mortality (IHM) for patients transferred from the ED to the ICU. We employed various machine learning (ML) methods and conducted a comparative analysis considering accuracy, stability, similarity, and interpretability. The effectiveness of our FS framework was evaluated using discrimination and calibration metrics, with SHAP applied to enhance the interpretability of model decisions.ResultsCohort 1 comprised 48,780 ICU encounters, of which 8,883 (18.21%) developed AKI. Cohort 2 included 29,197 transfers from the ED to the ICU, with 3,219 (11.03%) resulting in IHM. Among the ten ML methods evaluated, the tree-based ensemble method achieved the highest accuracy. As the number of top-ranking features increased, the models' accuracy began to stabilize, while feature subset stability (considering sample variations) and inter-method feature similarity reached optimal levels, confirming the validity of the FS framework. The integration of interpretative methods and expert knowledge in the final step further improved feature interpretability. The FS framework effectively reduced the number of features (e.g., from 380 to 35 for Cohort 1, and from 273 to 54 for Cohort 2) without significantly affecting prediction performance (Delong test, p > 0.05).ConclusionThe multi-step FS method developed in this study successfully reduces the dimensionality of features in EMR while preserving the accuracy of clinical outcome prediction. Furthermore, it improves the interpretability of risk factors by incorporating expert knowledge validation.

Dataset Information

A multidimensional and multi-feature framework for cardiac interoception.

Publications

A multidimensional and multi-feature framework for cardiac interoception.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets