Purpose/Significance: This study on the social support behavior of online smoking cessation community users can provide support for community management personnel to guide user behavior, enrich and ...deepen the functions of online smoking cessation communities and improve user stickiness, and also provides effective basis for public health workers to develop smoking cessation strategy in online communities. Method/Process: All the posts published in Baidu JieYanBa from August 1, 2018 to October 31, 2018 were gained. 2758 posts from core users were selected. Theme coding was adopted to divide the stage of smoking cessation of the poster. Keywords extraction and co-keywords network analysis were carried out to identify the types of social support and analyze its similarities and differences in different smoking cessation stages. Results: With the development of the smoking cessation stage, the proportion of emotional support and information support is on the rise. Emotional support is the main theme of social support in the preparatory stage, the action stage and the maintenance stage. The types and proportions of social support change regularly at different stages of smoking cessation.
Human readers or radiologists routinely perform full-body multi-organ
multi-disease detection and diagnosis in clinical practice, while most medical
AI systems are built to focus on single organs ...with a narrow list of a few
diseases. This might severely limit AI's clinical adoption. A certain number of
AI models need to be assembled non-trivially to match the diagnostic process of
a human reading a CT scan. In this paper, we construct a Unified Tumor
Transformer (CancerUniT) model to jointly detect tumor existence & location and
diagnose tumor characteristics for eight major cancers in CT scans. CancerUniT
is a query-based Mask Transformer model with the output of multi-tumor
prediction. We decouple the object queries into organ queries, tumor detection
queries and tumor diagnosis queries, and further establish hierarchical
relationships among the three groups. This clinically-inspired architecture
effectively assists inter- and intra-organ representation learning of tumors
and facilitates the resolution of these complex, anatomically related
multi-organ cancer image reading tasks. CancerUniT is trained end-to-end using
a curated large-scale CT images of 10,042 patients including eight major types
of cancers and occurring non-cancer tumors (all are pathology-confirmed with 3D
tumor masks annotated by radiologists). On the test set of 631 patients,
CancerUniT has demonstrated strong performance under a set of clinically
relevant evaluation metrics, substantially outperforming both multi-disease
methods and an assembly of eight single-organ expert models in tumor detection,
segmentation, and diagnosis. This moves one step closer towards a universal
high performance cancer screening tool.
In order to develop a gold corpus for Biomedical Natural Language Processing community for the sake of knowledge discovery in drug repurposing, an active gene annotation corpus (AGAC) was developed ...in this research. Five semantic trigger labels and three root regulatory trigger labels were designed as molecular- and cell- level biological entity annotations, which focused on the information of function change in biological processes resulted from mutated genes. In addition, predicates 'ThemeOf' and 'CauseOf' were as well annotated manually for the semantic knowledge extraction. Eventually, roles of gene mutation including gain of function (GOF) and loss of function (LOF) were curated through the AGAC annotation guideline. The information from AGAC annotation effectively bridge the association between mutation, gene, drug and disease, and make it possible to predict new drug direction in a large scale. AGAC corpus availability: The corpus is available in PubAn-notation platform 1 1 http://pubannotation.org/projects/HZAU_Active_Gene_Corpus.
We explore different ways of detecting the glottal feature changes when the subject is under workload. Glottal characteristics from speech production representing the vocal cord behavior will be ...discussed. We believe that workload of the human subject has specific impact on the behavior of the vocal folds, which may result in the variations in glottal flow. This paper investigates the variation of the glottal flow in workload state. Glottal source is discussed and the parameters from glottal flow representing the variations in workload are proposed. Through a study on a database containing over 700 voice signals from 11 speakers (four male and seven female), we prove that two glottal features (NAQ and CPR) are significantly modified due to the increased workload pressure, which will provide a standard in classification of two different states. Experimental results show that NAQ and CPR values of vowel segments are more effective than traditional features.
In human-robots interfaces studies have shown that stress caused by these factors results in variations in speaker's pronunciation, making highly reliable speech recognition systems of interface ...difficult to achieve. In this paper, we focus on the glottal source of stress speech under workload condition. The glottal flow resulted from vibration of the vocal folds is mainly discussed and features derived from glottal source are investigated to separate stressed speech from neutral speech. Experimental results show the proposed features representing glottal source could be able to lead to improvements in classification for the stressed speech.
Human readers or radiologists routinely perform full-body multi-organ multi-disease detection and diagnosis in clinical practice, while most medical AI systems are built to focus on single organs ...with a narrow list of a few diseases. This might severely limit AI's clinical adoption. A certain number of AI models need to be assembled non-trivially to match the diagnostic process of a human reading a CT scan. In this paper, we construct a Unified Tumor Transformer (UniT) model to detect (tumor existence and location) and diagnose (tumor characteristics) eight major cancer-prevalent organs in CT scans. UniT is a query-based Mask Transformer model with the output of multi-organ and multi-tumor semantic segmentation. We decouple the object queries into organ queries, detection queries and diagnosis queries, and further establish hierarchical relationships among the three groups. This clinically-inspired architecture effectively assists inter- and intra-organ representation learning of tumors and facilitates the resolution of these complex, anatomically related multi-organ cancer image reading tasks. UniT is trained end-to-end using a curated large-scale CT images of 10,042 patients including eight major types of cancers and occurring non-cancer tumors (all are pathology-confirmed with 3D tumor masks annotated by radiologists). On the test set of 631 patients, UniT has demonstrated strong performance under a set of clinically relevant evaluation metrics, substantially outperforming both multi-organ segmentation methods and an assembly of eight single-organ expert models in tumor detection, segmentation, and diagnosis. Such a unified multi-cancer image reading model (UniT) can significantly reduce the number of false positives produced by combined multi-system models. This moves one step closer towards a universal high-performance cancer screening tool.