RESUMEN
Heart failure (HF) is a leading cause of morbidity, healthcare costs, and mortality. Guideline based segmentation of HF into distinct subtypes is coarse and unlikely to reflect the heterogeneity of etiologies and disease trajectories of patients. While analyses of electronic health records show promise in expanding our understanding of complex syndromes like HF in an evidence-driven way, limitations in data quality have presented challenges for large-scale EHR-based insight generation and decision-making. We present a hypothesis-free approach to generating real-world characteristics and progression patterns of HF. Patient disease state snapshots are extracted from the complaints mentioned in unstructured clinical notes. Typical disease states are generated by clustering and characterized in terms of their distinguishing features, temporal relationships, and risk of important clinical events. Our analysis generates a comprehensive "disease phenome" of real-world patients computed from large, noisy, secondary-use EHR datasets created in a routine clinical setting.
Asunto(s)
Registros Electrónicos de Salud , Insuficiencia Cardíaca , Humanos , SíndromeRESUMEN
As a leading cause of death and morbidity, heart failure (HF) is responsible for a large portion of healthcare and disability costs worldwide. Current approaches to define specific HF subpopulations may fail to account for the diversity of etiologies, comorbidities, and factors driving disease progression, and therefore have limited value for clinical decision making and development of novel therapies. Here we present a novel and data-driven approach to understand and characterize the real-world manifestation of HF by clustering disease and symptom-related clinical concepts (complaints) captured from unstructured electronic health record clinical notes. We used natural language processing to construct vectorized representations of patient complaints followed by clustering to group HF patients by similarity of complaint vectors. We then identified complaints that were significantly enriched within each cluster using statistical testing. Breaking the HF population into groups of similar patients revealed a clinically interpretable hierarchy of subgroups characterized by similar HF manifestation. Importantly, our methodology revealed well-known etiologies, risk factors, and comorbid conditions of HF (including ischemic heart disease, aortic valve disease, atrial fibrillation, congenital heart disease, various cardiomyopathies, obesity, hypertension, diabetes, and chronic kidney disease) and yielded additional insights into the details of each HF subgroup's clinical manifestation of HF. Our approach is entirely hypothesis free and can therefore be readily applied for discovery of novel insights in alternative diseases or patient populations.
Asunto(s)
Registros Electrónicos de Salud , Insuficiencia Cardíaca/patología , Anciano , Fibrilación Atrial/etiología , Fibrilación Atrial/patología , Fibrilación Atrial/fisiopatología , Análisis por Conglomerados , Femenino , Insuficiencia Cardíaca/etiología , Insuficiencia Cardíaca/fisiopatología , Humanos , Hipertensión/etiología , Hipertensión/patología , Hipertensión/fisiopatología , Masculino , Persona de Mediana Edad , Fenotipo , FilogeniaRESUMEN
Care for patients with chronic wounds can be complex, and the chances of poor outcomes are high if wound care is not optimized through evidence-based protocols. Tracking and managing every variable and comorbidity in patients with wounds is difficult despite the increasing use of wound-specific electronic medical records. Harnessing the power of big data analytics to help nurses and physicians provide optimized care based on the care provided to millions of patients can result in better outcomes. Numerous applications of machine learning toward workflow improvements, inpatient monitoring, outpatient communication, and hospital operations can improve overall efficiency and efficacy of care delivery in and out of the hospital, while reducing adverse events and complications. This article provides an overview of the application of big data analytics and machine learning in health care, highlights important recent advances, and discusses how these technologies may revolutionize advanced wound care.
Asunto(s)
Ciencia de los Datos/tendencias , Cicatrización de Heridas , Heridas y Lesiones/terapia , Humanos , Almacenamiento y Recuperación de la Información/métodos , Almacenamiento y Recuperación de la Información/normas , Aprendizaje Automático/tendenciasRESUMEN
In this paper, we introduce the Neural Acoustic Processing Library (NAPLib), a toolbox containing novel processing methods for real-time and offline analysis of neural activity in response to speech. Our method divides the speech signal and resultant neural activity into segmental units (e.g., phonemes), allowing for fast and efficient computations that can be implemented in real-time. NAPLib contains a suite of tools that characterize various properties of the neural representation of speech, which can be used for functionality such as characterizing electrode tuning properties, brain mapping and brain computer interfaces. The library is general and applicable to both invasive and non-invasive recordings, including electroencephalography (EEG), electrocorticography (ECoG) and magnetoecnephalography (MEG). In this work, we describe the structure of NAPLib, as well as demonstrating its use in both EEG and ECoG. We believe NAPLib provides a valuable tool to both clinicians and researchers who are interested in the representation of speech in the brain.
RESUMEN
Despite the recent success of deep learning, the nature of the transformations they apply to the input features remains poorly understood. This study provides an empirical framework to study the encoding properties of node activations in various layers of the network, and to construct the exact function applied to each data point in the form of a linear transform. These methods are used to discern and quantify properties of feedforward neural networks trained to map acoustic features to phoneme labels. We show a selective and nonlinear warping of the feature space, achieved by forming prototypical functions to account for the possible variation of each class. This study provides a joint framework where the properties of node activations and the functions implemented by the network can be linked together.
RESUMEN
Infants' speech perception adapts to the phonemic categories of their native language, a process assumed to be driven by the distributional properties of speech. This study investigates whether deep neural networks (DNNs), the current state-of-the-art in distributional feature learning, are capable of learning phoneme-like representations of speech in an unsupervised manner. We trained DNNs with unlabeled and labeled speech and analyzed the activations of each layer with respect to the phones in the input segments. The analyses reveal that the emergence of phonemic invariance in DNNs is dependent on the availability of phonemic labeling of the input during the training. No increased phonemic selectivity of the hidden layers was observed in the purely unsupervised networks despite successful learning of low-dimensional representations for speech. This suggests that additional learning constraints or more sophisticated models are needed to account for the emergence of phone-like categories in distributional learning operating on natural speech.