Embedded binary PSO integrating classical methods for multilevel improved feature selection in liver and kidney disease diagnosis

Author(s):  
Gunasundari Selvaraj ◽  
Janakiraman Subbiah ◽  
Meenambal Selvaraj
2021 ◽  
Vol 317 ◽  
pp. 05030
Author(s):  
Siti Noor Chotimah ◽  
Budi Warsito ◽  
Bayu Surarso

The number of factors that can be categorized into the diagnosis of Chronic Kidney Disease (CKD) at an early stage makes information about the diagnosis of the disease divided into information that has many influences and has little influence. This study aims to select diagnoses in medical records with the most influential information on chronic kidney disease. The first step is to select a diagnosis with much influence by implementing the Sequential Backward Feature Selection (SBFS). This algorithm eliminates features that are considered to have little influence when compared to other features. In the second step, the features of the best diagnoses are used as input to the Artificial Neural Network (ANN) classification algorithm. The results obtained from this study are information in the form of the best diagnoses that have much influence on chronic kidney disease and the accuracy results based on the selected diagnoses. Based on the study results, 15 features are considered the best of the 18 features used to achieve 88% accuracy results. Compared with conventional methods, this method still requires consideration from the medical staff because it is not a final diagnosis for patients.


2020 ◽  
Vol 47 (1) ◽  
pp. 67
Author(s):  
Areti Stavropoulou ◽  
Michael Rovithis ◽  
Maria G. Grammatikopoulou ◽  
Konstantina Kyriakidi ◽  
Andriani Pylarinou ◽  
...  

Author(s):  
Pramila Arulanthu ◽  
Eswaran Perumal

: The medical data has an enormous quantity of information. This data set requires effective classification for accurate prediction. Predicting medical issues is an extremely difficult task in which Chronic Kidney Disease (CKD) is one of the major unpredictable diseases in medical field. Perhaps certain medical experts do not have identical awareness and skill to solve the issues of their patients. Most of the medical experts may have underprivileged results on disease diagnosis of their patients. Sometimes patients may lose their life in nature. As per the Global Burden of Disease (GBD-2015) study, death by CKD was ranked 17th place and GBD-2010 report 27th among the causes of death globally. Death by CKD is constituted 2·9% of all death between the year 2010 and 2013 among people from 15 to 69 age. As per World Health Organization (WHO-2005) report, 58 million people expired by CKD. Hence, this article presents the state of art review on Chronic Kidney Disease (CKD) classification and prediction. Normally, advanced data mining techniques, fuzzy and machine learning algorithms are used to classify medical data and disease diagnosis. This study reviews and summarizes many classification techniques and disease diagnosis methods presented earlier. The main intention of this review is to point out and address some of the issues and complications of the existing methods. It is also attempts to discuss the limitations and accuracy level of the existing CKD classification and disease diagnosis methods.


2021 ◽  
Vol 10 (10) ◽  
pp. 2046
Author(s):  
Goren Saenz-Pipaon ◽  
Saioa Echeverria ◽  
Josune Orbe ◽  
Carmen Roncal

Diabetic kidney disease (DKD) is the leading cause of end stage renal disease (ESRD) in developed countries, affecting more than 40% of diabetes mellitus (DM) patients. DKD pathogenesis is multifactorial leading to a clinical presentation characterized by proteinuria, hypertension, and a gradual reduction in kidney function, accompanied by a high incidence of cardiovascular (CV) events and mortality. Unlike other diabetes-related complications, DKD prevalence has failed to decline over the past 30 years, becoming a growing socioeconomic burden. Treatments controlling glucose levels, albuminuria and blood pressure may slow down DKD evolution and reduce CV events, but are not able to completely halt its progression. Moreover, one in five patients with diabetes develop DKD in the absence of albuminuria, and in others nephropathy goes unrecognized at the time of diagnosis, urging to find novel noninvasive and more precise early diagnosis and prognosis biomarkers and therapeutic targets for these patient subgroups. Extracellular vesicles (EVs), especially urinary (u)EVs, have emerged as an alternative for this purpose, as changes in their numbers and composition have been reported in clinical conditions involving DM and renal diseases. In this review, we will summarize the current knowledge on the role of (u)EVs in DKD.


2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Peter Bramlage ◽  
Stefanie Lanzinger ◽  
Sascha R. Tittel ◽  
Eva Hess ◽  
Simon Fahrner ◽  
...  

Abstract Background Recent European Society of Cardiology (ESC)/European Association for the Study of Diabetes (EASD) guidelines provide recommendations for detecting and treating chronic kidney disease (CKD) in diabetic patients. We compared clinical practice with guidelines to determine areas for improvement. Methods German database analysis of 675,628 patients with type 1 or type 2 diabetes, with 134,395 included in this analysis. Data were compared with ESC/EASD recommendations. Results This analysis included 17,649 and 116,747 patients with type 1 and type 2 diabetes, respectively. The analysis showed that 44.1 and 49.1 % patients with type 1 and type 2 diabetes, respectively, were annually screened for CKD. Despite anti-diabetic treatment, only 27.2 % patients with type 1 and 43.5 % patients with type 2 achieved a target HbA1c of < 7.0 %. Use of sodium-glucose transport protein 2 inhibitors (1.5 % type 1/8.7 % type 2 diabetes) and glucagon-like peptide-1 receptor agonists (0.6 % type 1/5.2 % type 2 diabetes) was limited. Hypertension was controlled according to guidelines in 41.1 and 67.7 % patients aged 18–65 years with type 1 and 2 diabetes, respectively, (62.4 vs. 68.4 % in patients > 65 years). Renin angiotensin aldosterone inhibitors were used in 24.0 and 40.9 % patients with type 1 diabetes (micro- vs. macroalbuminuria) and 39.9 and 47.7 %, respectively, in type 2 diabetes. Conclusions Data indicate there is room for improvement in caring for diabetic patients with respect to renal disease diagnosis and treatment. While specific and potentially clinically justified reasons for non-compliance exist, the data may serve well for a critical appraisal of clinical practice decisions.


Sign in / Sign up

Export Citation Format

Share Document