We propose a new type of generative model of high-dimensional data that learns a manifold geometry of the data, rather than density, and can generate points evenly along this manifold. This is in contrast to existing ...
详细信息
The recent adoption of Electronic Health Records (EHRs) by healthcare providers has introduced an important source of data that provides detailed and highly specific insights into patient phenotypes over large cohorts...
详细信息
The recent adoption of Electronic Health Records (EHRs) by healthcare providers has introduced an important source of data that provides detailed and highly specific insights into patient phenotypes over large cohorts. These datasets, in combination with machine learning and statistical approaches, generate new opportunities for research and clinical care. However, many methods require the patient representations to be in structured formats, while the information in the EHR is often locked in unstructured text designed for human readability. In this work, we develop the methodology to automatically extract clinical features from clinical narratives from large EHR corpora without the need for prior knowledge. We consider medical terms and sentences appearing in clinical narratives as atomic information units. We propose an efficient clustering strategy suitable for the analysis of large text corpora and utilize the clusters to represent information about the patient compactly. Additionally, we define the sentences on ontologic and natural language vocabularies to automatically detect pertinent combinations of concepts present in the corpus, even when an ontology is not available. To demonstrate the utility of our approach, we perform an association study of clinical features with somatic mutation profiles from 4,007 cancer patients and their tumors. We apply the proposed algorithm to a dataset consisting of .65 thousand documents with a total of .3.2 million sentences. After correcting for cancer type and other confounding factors, we identify a total of 340 significant statistical associations between the presence of somatic mutations and clinical features. We annotated these associations according to their novelty and we report several known associations. We also propose 37 plausible, testable hypothesis for associations where the underlying biological mechanism does not appear to be known. These results illustrate that the automated discovery of clinical features
Latent variable models with hidden binary units appear in various applications. Learning such models, in particular in the presence of noise, is a challenging computational problem. In this paper we propose a novel sp...
详细信息
Background & Aims Magnetic resonance-based steatosis measurement of the liver has become the non-invasive gold standard; however very few diagnostic thresholds have been independently validated and none in childre...
Background & Aims Magnetic resonance-based steatosis measurement of the liver has become the non-invasive gold standard; however very few diagnostic thresholds have been independently validated and none in children. This study was designed to validate diagnostic thresholds, assess repeatability and the relationship to histology of a proprietary MRI volumetric liver fat fraction (VLFF) method (HepaFat-Scan ® ) and magnetic resonance spectroscopy (MRS) proton density fat fraction (PDFF) method. Methods This cross-sectional study assesses the diagnostic accuracy of MRI-VLFF and MRS-PDFF of 50 children who had a clinically indicated liver biopsy. Twenty-five children were randomly selected for intra-examination repeatability. The positive pixel count (PPC) algorithm was used to quantify the area of fat vesicles in the histology. The area under the receiver operating characteristic curve (AUROC) was used to assess the diagnostic accuracy. Bland-Altman plots were used to compare the methods and examine the repeatability. Results The MRI-VLFF ranged from 0.3% to 35.4% and the MRS-PDFF ranged from 0.5% to 36.9%. Based on histopathological analysis, the diagnostic accuracy of MRI-VLFF for steatosis detection (grade 0 vs 1-3) showed an AUROC of 0.95 (95% confidence interval [CI]: 0.89-1.00), whereas MRI-PDFF classified grade 0 vs 1-3 showed an AUROC of 0.96 (95% CI: 0.91-1.00). The repeatability coefficients were 1.8 (±0.5) % for MRI-VLFF and 2.1 (±0.6) % for MRS-PDFF. There was no bias between MRI-VLFF and PPC steatosis. To allow direct comparison, MRS-PDFF was converted to MRS-VLFF. Small, but significant biases were measured between PPC steatosis and MRS-VLFF (−1.38%) and between MRI-VLFF and MRS-VLFF (−1.39%). Conclusions Both MRI-VLFF and MRS-PDFF accurately estimated hepatic steatosis in children with excellent agreement with the histological assessment.
The Open Databases Integration for Materials Design (OPTIMADE) consortium has designed a universal application programming interface (API) to make materials databases accessible and interoperable. We outline the first...
详细信息
Pediatric tumors of the central nervous system are the most common cause of cancer-related death in children. The five-year survival rate for high-grade gliomas in children is less than 20%. Due to their rarity, the d...
详细信息
The 2018 arrival of African swine fever (ASF) in China was followed by reports of wild pig deaths across most countries in Southeast Asia. However, the magnitude and duration of population-level impacts of ASF on wild...
详细信息
The 2018 arrival of African swine fever (ASF) in China was followed by reports of wild pig deaths across most countries in Southeast Asia. However, the magnitude and duration of population-level impacts of ASF on wild pig species remain unclear. To elucidate the spatiotemporal spread of ASF in the region for native pig species, we gathered qualitative information on wild pig population dynamics in Southeast Asia between 2018 and 2024 from 88 expert elicitation questionnaires representing sites in 11 countries. Peak reported population declines occurred in 2021 and 2022, with more than half of respondents reporting declining wild pig populations, far higher than in earlier years. The reported declines waned to 44.23% in 2024, whereas simultaneously, the number of populations reported to be “increasing” increased from 11.3%–13.2% in 2019–2022 to 28.9% in 2024. These reports suggest that the ASF outbreak may have peaked for wild boars and bearded pigs in mainland Southeast Asia, Borneo, and Sumatra, with some subsequent recovery. However, the disease is still expanding into the ranges of island endemic species, such as new reports for the Sulawesi warty pig ( Sus celebensis ) in September of 2024. Island endemics remain particularly vulnerable to extinction from ASF and require urgent monitoring and conservation action.
Objective: To identify and quantify risk factors for incident knee osteoarthritis (KOA) across the lifespan. Methods: This systematic review and meta-analysis identified eligible studies from seven electronic database...
详细信息
Using the full four-year SPTpol 500 deg2 dataset in both the 95 GHz and 150 GHz frequency bands, we present measurements of the temperature and E-mode polarization of the cosmic microwave background (CMB), as well as ...
详细信息
暂无评论