Improving predictive models for intensive care unit (ICU) inpatients requires a new strategy that periodically includes the latest clinical data and can be updated to reflect local characteristics. We extracted data from all adult patients admitted to the ICUs of two university hospitals with different characteristics from 2006 to 2020, and a total of 85,146 patients were included in this study. Machine learning algorithms were trained to predict in-hospital mortality. The predictive performance of conventional scoring models and machine learning algorithms was assessed by the area under the receiver operating characteristic curve (AUROC). The conventional scoring models had various predictive powers, with the SAPS III (AUROC 0.773 [0.766–0.779] for hospital S) and APACHE III (AUROC 0.803 [0.795–0.810] for hospital G) showing the highest AUROC among them. The best performing machine learning models achieved an AUROC of 0.977 (0.973–0.980) in hospital S and 0.955 (0.950–0.961) in hospital G. The use of ML models in conjunction with conventional scoring systems can provide more useful information for predicting the prognosis of critically ill patients. In this study, we suggest that the predictive model can be made more robust by training with the individual data of each hospital.
To investigate whether postoperative filtering bleb size affects the surgical outcome after trabeculectomy.Methods: In this study, we retrospectively reviewed 145 medically uncontrolled glaucoma patients with intraocular pressure (IOP) values >21 mmHg before surgery and data from ≥2 years of follow-up.Postoperative IOP, filtering bleb size including extent and height, and other clinical factors were measured after trabeculectomy. We divided bleb extent into quadrants and bleb height by 0.5 intervals of corneal thickness (CT). The main outcome measure was surgical success. We confirmed complete success when the IOP was ≤21 mmHg and decreased by >20% from baseline without medication or additional procedures. Qualified success used the same criteria but allowed for medication or additional procedures. Cases with reoperation or 2 consecutive IOP measurements < 6 mmHg were considered failures.Results: A total of 145 eyes of 145 patients was included. The average observation period was 30.8 ± 10.9 months. During multivariate Cox regression analysis, a larger extent of filtering bleb revealed significantly low hazard ratios (HRs) in both complete and surgical success (0.509 and 0.494, respectively); however, there was no significant relationship between bleb height and surgical outcome. Conclusion:The extent of the filtering bleb was associated with surgical outcomes of trabeculectomy in glaucoma patients.
The etiology of open-angle glaucoma (OAG) is yet unclear. This study investigated possible risk factors, such as the morphology of the border tissue that affect the pattern of visual field (VF) progression in eyes with OAG. 166 eyes of 166 OAG patients with an externally oblique border tissue (EOBT) at least in one direction were included. EOBT was obtained by analyzing enhanced depth imaging spectral-domain optical coherence tomography images. A pointwise linear regression was used to determine VF progression by measuring the deterioration rate of each point in the VF. The odds ratio of VF progression for each risk factor was estimated using logistic regression analysis. Seventy (42.2%) eyes showed VF deterioration. In multivariate analysis, longer follow-up period, higher baseline intraocular pressure (IOP), lower mean ocular perfusion pressure (MOPP), and smaller angular location of the longest EOBT were associated with VF progression (all p values were below 0.05). In the multivariate analysis, the location of the longest EOBT was significantly associated with inferior (p = 0.002) and central (p = 0.017) VF progression. In conclusion, VF progression pattern in OAG eyes is associated with the location of the longest EOBT as well as other known risk factors.
e16116 Background: In Korea, liver cancer is the fifth most frequently diagnosed cancer and has the second highest cancer mortality rate. Despite recent advances in systemic therapy, patients with Hepatocellular carcinoma (HCC) continue to have poor prognoses as many are diagnosed at an advanced disease stage. To better understand the current treatment landscape and contribute to improved patient outcomes, the LINK (Liver Cancer in Korea) research network was established as a longitudinal real-world HCC database with annual data refresh. The aim was to provide current and up-to-date real-world insights into HCC patient characteristics, treatment patterns, and outcomes. Methods: LINK’s data sources are the electronic medical records of three leading academic hospitals (Asan Medical Center, Samsung Medical Center, Severance Hospital). Extracted data are standardized using a study-specific common data model to ensure comparability across hospitals. Adult patients with primary diagnosis of HCC (ICD-10: C22.0) in 2015-2020 were included and followed until death, loss to follow-up, or June 2021, whichever occurred first. Patients without valid treatment records or morphology codes or with other primary cancer diagnosis were excluded. Data on key patient and clinical characteristics, treatments, such as systemic anti-cancer therapy (SACT) regimens, and real-world outcomes were collected. Rule-based algorithms were developed to define initial treatments and lines of therapy in palliative SACT. Results: Of 25,480 HCC patients identified, 10,298 (40%) met eligibility criteria and were included in the analysis. The median age was 60 years and 19% were female. There were 74% of patients classified as Child-Pugh (CP) class A, 16% as B, 1% as C, and 9% with unknown CP at diagnosis. For albumin-bilirubin (ALBI) grade at diagnosis, 49% of patients were classified as grade 1, 45% as 2, 5% as 3, and 1% unknown. Loco-regional therapy was the most common initial treatment, with 54% and 13% of patients receiving transarterial therapy and local ablation therapy, respectively. A total of 628 patients (6%) undergone liver transplantation regardless of preceding treatments and 15% of patients were initially treated with hepatectomy. Only 2% received radiation therapy (RT) or RT in combination with cytotoxic chemotherapy as initial treatment. Among 3,026 patients who received SACT either as an initial (10%) or subsequent treatment, 67%, 20%, and 13% received 1 line, 2 lines, and 3 lines of therapy. The most common first-line SACT was sorafenib, followed by lenvatinib. Conclusions: The LINK database can serve as a multi-center research platform to facilitate fast and reliable real-world evidence generation for HCC patients. Future updates and subgroup analyses of interest can further evaluate the impact of new therapies and changes to HCC treatment guidelines in Korea.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.