No abstract

BackgroundThere are few data on the prevalence of low T3 (triiodothyronine) syndrome in patients with non-dialysis chronic kidney disease (CKD) and it is unclear whether low T3 can be used to predict the progression of CKD.Material/MethodsWe retrospectively studied 279 patients who had been definitively diagnosed with CKD, without needing maintenance dialysis. Thyroid function was analyzed in all enrolled subjects and the incidence of thyroid dysfunction (low T3 syndrome, low T4 syndrome, and subclinical hypothyroidism) in patients at different stages of CKD was determined.ResultsGlomerular filtration rate (GFR) of CKD patients was estimated as follows: 145 subjects (52%) had GFR <60 ml/min per 1.73 m2; 47 subjects (16.8%) had GFR between 30 and 59 ml/min per 1.73 m2, and 98 subjects (35.1%) had GFR <30 ml/min per 1.73 m2. Among all enrolled subjects, 4.7% (n=13) had subclinical hypothyroidism, 5.4% (n=15) had low T4 syndrome, and 47% (n=131) had low T3 syndrome. In 114 CKD patients in stages 3–5, serum T3 was positively related to protein metabolism (STP, PA, and ALB) and anemia indicators (Hb and RBC), and negatively related to inflammatory status (CRP and IL-6).ConclusionsA high prevalence of low T3 syndrome was observed in CKD patients without dialysis, even in early stages (1 and 2). The increasing prevalence of low T3 as CKD progresses indicates its value as a predictor of worsening CKD. Furthermore, low T3 syndrome is closely associated with both malnutrition-inflammation complex syndrome (MICS) and anemia.

Web-scale search systems typically tackle the scalability challenge with a two-step paradigm: retrieval and ranking. The retrieval step, also known as candidate selection, often involves extracting standardized entities, creating an inverted index, and performing term matching for retrieval. Such traditional methods require manual and time-consuming development of query models. In this paper, we discuss applying learning-to-retrieve technology to enhance LinkedIn's job search and recommendation systems. In the realm of promoted jobs, the key objective is to improve the quality of applicants, thereby delivering value to recruiter customers. To achieve this, we leverage confirmed hire data to construct a graph that evaluates a seeker's qualification for a job, and utilize learned links for retrieval. Our learned model is easy to explain, debug and adjust. On the other hand, the focus for organic jobs is to optimize seeker engagement. We accomplished this by training embeddings for personalized retrieval, fortified by a set of rules derived from the categorization of member feedbacks. In addition to a solution based on a conventional inverted index, we developed an on-GPU solution capable of supporting both KNN and term matching efficiently.

No abstract