predictive model
Recently Published Documents





2022 ◽  
Vol 277 ◽  
pp. 108388
Chunfeng Gu ◽  
Wopke van der Werf ◽  
Lammert Bastiaans

2023 ◽  
Vol 19 (1) ◽  
pp. 1
Luis Carlos Vanegas Granados ◽  
Sandra Paola Espitia Layton ◽  
Jhonny Erick Valle Mendoza ◽  
Leyner Cardenas Mercado ◽  
Sergio Aldair Castiblanco Ballesteros ◽  

Salma Firdose ◽  
Surendran Swapna Kumar ◽  
Ravinda Gayan Narendra Meegama

Social distancing is one of the simple and effective shields for every individual to control spreading of virus in present scenario of pandemic coronavirus disease (COVID-19). However, existing application of social distancing is a basic model and it is also characterized by various pitfalls in case of dynamic monitoring of infected individual accurately. Review of existing literature shows that there has been various dedicated research attempt towards social distancing using available technologies, however, there are further scope of improvement too. This paper has introduced a novel framework which is capable of computing the level of threat with much higher degree of accuracy using distance and duration of stay as elementary parameters. Finally, the model can successfully classify the level of threats using deep learning. The study outcome shows that proposed system offers better predictive performance in contrast to other approaches.

2022 ◽  
Vol 22 (1) ◽  
Tenghui Han ◽  
Jun Zhu ◽  
Xiaoping Chen ◽  
Rujie Chen ◽  
Yu Jiang ◽  

Abstract Background Liver is the most common metastatic site of colorectal cancer (CRC) and liver metastasis (LM) determines subsequent treatment as well as prognosis of patients, especially in T1 patients. T1 CRC patients with LM are recommended to adopt surgery and systematic treatments rather than endoscopic therapy alone. Nevertheless, there is still no effective model to predict the risk of LM in T1 CRC patients. Hence, we aim to construct an accurate predictive model and an easy-to-use tool clinically. Methods We integrated two independent CRC cohorts from Surveillance Epidemiology and End Results database (SEER, training dataset) and Xijing hospital (testing dataset). Artificial intelligence (AI) and machine learning (ML) methods were adopted to establish the predictive model. Results A total of 16,785 and 326 T1 CRC patients from SEER database and Xijing hospital were incorporated respectively into the study. Every single ML model demonstrated great predictive capability, with an area under the curve (AUC) close to 0.95 and a stacking bagging model displaying the best performance (AUC = 0.9631). Expectedly, the stacking model exhibited a favorable discriminative ability and precisely screened out all eight LM cases from 326 T1 patients in the outer validation cohort. In the subgroup analysis, the stacking model also demonstrated a splendid predictive ability for patients with tumor size ranging from one to50mm (AUC = 0.956). Conclusion We successfully established an innovative and convenient AI model for predicting LM in T1 CRC patients, which was further verified in the external dataset. Ultimately, we designed a novel and easy-to-use decision tree, which only incorporated four fundamental parameters and could be successfully applied in clinical practice.

Lindsey W. Vilca ◽  
Blanca V. Chávez ◽  
Yoselin Shara Fernández ◽  
Tomás Caycho-Rodríguez ◽  
Michael White

2022 ◽  
Vol 12 ◽  
Daniel Kevin Llanera ◽  
Rebekah Wilmington ◽  
Haika Shoo ◽  
Paulo Lisboa ◽  
Ian Jarman ◽  

ObjectiveTo identify clinical and biochemical characteristics associated with 7- & 30-day mortality and intensive care admission amongst diabetes patients admitted with COVID-19.Research Design and MethodsWe conducted a cohort study collecting data from medical notes of hospitalised people with diabetes and COVID-19 in 7 hospitals within the Mersey-Cheshire region from 1 January to 30 June 2020. We also explored the impact on inpatient diabetes team resources. Univariate and multivariate logistic regression analyses were performed and optimised by splitting the dataset into a training, test, and validation sets, developing a robust predictive model for the primary outcome.ResultsWe analyzed data from 1004 diabetes patients (mean age 74.1 (± 12.6) years, predominantly men 60.7%). 45% belonged to the most deprived population quintile in the UK. Median BMI was 27.6 (IQR 23.9-32.4) kg/m2. The primary outcome (7-day mortality) occurred in 24%, increasing to 33% by day 30. Approximately one in ten patients required insulin infusion (9.8%). In univariate analyses, patients with type 2 diabetes had a higher risk of 7-day mortality [p < 0.05, OR 2.52 (1.06, 5.98)]. Patients requiring insulin infusion had a lower risk of death [p = 0.02, OR 0.5 (0.28, 0.9)]. CKD in younger patients (<70 years) had a greater risk of death [OR 2.74 (1.31-5.76)]. BMI, microvascular and macrovascular complications, HbA1c, and random non-fasting blood glucose on admission were not associated with mortality. On multivariate analysis, CRP and age remained associated with the primary outcome [OR 3.44 (2.17, 5.44)] allowing for a validated predictive model for death by day 7.ConclusionsHigher CRP and advanced age were associated with and predictive of death by day 7. However, BMI, presence of diabetes complications, and glycaemic control were not. A high proportion of these patients required insulin infusion warranting increased input from the inpatient diabetes teams.

Sign in / Sign up

Export Citation Format

Share Document