Diagnosis Rule Extraction from Patient Data for Chronic Kidney Disease Using Machine Learning

2016 ◽  
Vol 5 (2) ◽  
pp. 64-72 ◽  
Author(s):  
Alexander Arman Serpen
Keyword(s):  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Decision Making ◽  
Kidney Disease ◽  
Learning Algorithm ◽  
Primary Care Providers ◽  
Patient Data ◽  
Care Providers ◽  
Data Set ◽  
Actual Patient

This research study employed a machine learning algorithm on actual patient data to extract decision making rules that can be used to diagnose chronic kidney disease. The patient data set entails a number of health-related attributes or indicators and contains 250 patients positive for chronic kidney disease. The C4.5 decision tree algorithm was applied to the patient data to formulate a set of diagnosis rules for chronic kidney disease. The C4.5 algorithm utilizing 3-fold cross validation achieved 98.25% prediction accuracy and thus correctly classified 393 instances and incorrectly classified 7 instances for a total patient count of 400. The extracted rule set highlighted the need to monitor serum creatinine levels in patients as the primary indicator for the presence of disease. Secondary indicators were pedal edema, hemoglobin, diabetes mellitus and specific gravity. The set of rules provides a preliminary screening tool towards conclusive diagnosis of the chronic kidney disease by nephrologists following timely referral by the primary care providers or decision-making algorithms.

Diagnosis Rule Extraction from Patient Data for Chronic Kidney Disease Using Machine Learning

2020 ◽  
pp. 1165-1174
Author(s):  
Alexander Arman Serpen
Keyword(s):  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Decision Making ◽  
Kidney Disease ◽  
Learning Algorithm ◽  
Patient Data ◽  
Actual Patient ◽  
C4.5 Decision Tree ◽  
Health Related ◽  
Pedal Edema

This research study employed a machine learning algorithm on actual patient data to extract decision making rules that can be used to diagnose chronic kidney disease. The patient data set entails a number of health-related attributes or indicators and contains 250 patients positive for chronic kidney disease. The C4.5 decision tree algorithm was applied to the patient data to formulate a set of diagnosis rules for chronic kidney disease. The C4.5 algorithm utilizing 3-fold cross validation achieved 98.25% prediction accuracy and thus correctly classified 393 instances and incorrectly classified 7 instances for a total patient count of 400. The extracted rule set highlighted the need to monitor serum creatinine levels in patients as the primary indicator for the presence of disease. Secondary indicators were pedal edema, hemoglobin, diabetes mellitus and specific gravity. The set of rules provides a preliminary screening tool towards conclusive diagnosis of the chronic kidney disease by nephrologists following timely referral by the primary care providers or decision-making algorithms.

Predicting Chronic Kidney Disease Using KNN Algorithm

2021 ◽  
Vol 1 (2) ◽  
pp. 16-24
Author(s):  
V Mareeswari ◽  
Sunita Chalageri ◽  
Kavita K Patil
Keyword(s):  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Learning Algorithm ◽  
University Of California ◽  
Learning Models ◽  
Data Set ◽  
Deep Learning Algorithm ◽  
The University ◽  
Machine Learning Models

Chronic kidney disease (CKD) is a world heath issues, and that also includes damages and can’t filter blood the way it should be. since we cannot predict the early stages of CKD, patience will fail to recognise the disease. Pre detection of CKD will allow patience to get timely facility to ameliorate the progress of the disease. Machine learning models will effectively aid clinician’s progress this goal because of the early and accurate recognition performances. The CKD data set is collected from the University of California Irvine (UCI) Machine Learning Recognition. Multiple Machine and deep learning algorithm used to predict the chronic kidney disease.

Medications Used Routinely in Primary Care to be Dose-Adjusted or Avoided in People With Chronic Kidney Disease: Results of a Modified Delphi Study

2020 ◽  
Vol 54 (7) ◽  
pp. 625-632 ◽  
Author(s):  
Leena Taji ◽  
Marisa Battistella ◽  
Allan K. Grill ◽  
Jessie Cunningham ◽  
Brenda L. Hemmelgarn ◽  
...  
Keyword(s):  
Primary Care ◽  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Delphi Study ◽  
Dose Adjustment ◽  
Sglt2 Inhibitor ◽  
Primary Care Providers ◽  
Delphi Panel ◽  
Care Providers ◽  
Modified Delphi

Background: Chronic kidney disease (CKD) affects up to 18% of those over the age of 65 years. Potentially inappropriate medication prescribing in people with CKD is common. Objectives: Develop a pragmatic list of medications used in primary care that required dose adjustment or avoidance in people with CKD, using a modified Delphi panel approach, followed by a consensus workshop. Methods: We conducted a comprehensive literature search to identify potential medications. A group of 17 experts participated in a 3-round modified Delphi panel to identify medications for inclusion. A subsequent consensus workshop of 8 experts reviewed this list to prioritize medications for the development of point-of-care knowledge translation materials for primary care. Results: After a comprehensive literature review, 59 medications were included for consideration by the Delphi panel, with a further 10 medications added after the initial round. On completion of the 3 Delphi rounds, 66 unique medications remained, 63 requiring dose adjustment and 16 medications requiring avoidance in one or more estimated glomerular filtration rate categories. The consensus workshop prioritized this list further to 24 medications that must be dose-adjusted or avoided, including baclofen, metformin, and digoxin, as well as the newer SGLT2 inhibitor agents. Conclusion and Relevance: We have developed a concise list of 24 medications commonly used in primary care that should be dose-adjusted or avoided in people with CKD to reduce harm. This list incorporates new and frequently prescribed medications and will inform an updated, easy to access source for primary care providers.

scholarly journals Patient and Provider Characteristics Associated With Sodium–Glucose Cotransporter 2 Inhibitor Prescription in Patients With Diabetes and Proteinuric Chronic Kidney Disease

2020 ◽  
Author(s):  
Ian E. McCoy ◽  
Jialin Han ◽  
Maria E. Montez-Rath ◽  
Glenn M. Chertow ◽  
Jinnie J. Rhee
Keyword(s):  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Sglt2 Inhibitor ◽  
Primary Care Providers ◽  
Cross Sectional Study ◽  
Care Providers ◽  
Cross Sectional ◽  
Provider Characteristics ◽  
Sodium Glucose Cotransporter ◽  
Patients With Diabetes

Despite accumulating evidence of cardiorenal benefits from sodium–glucose cotransporter 2 (SGLT2) inhibitors, prescription of agents in this drug class may be limited by concerns regarding adverse effects and interdisciplinary care coordination. To investigate these potential barriers, we performed a cross-sectional study of SGLT2 inhibitor prescriptions in 2017 in 3,779 adults with type 2 diabetes and proteinuric chronic kidney disease from a nationwide database. Only 173 (5%) of these patients received an SGLT2 inhibitor in 2017. Younger age, renin-angiotensin-aldosterone system inhibitor prescription, and higher estimated glomerular filtration rate were associated with SGLT2 inhibitor prescription. Primary care providers were responsible for the majority of the prescriptions. Continued efforts should be made to track and improve SGLT2 inhibitor use in indicated populations.

scholarly journals Potentially Inappropriately Prescribed Medications Among Medicare Medication Therapy Management Eligible Patients with Chronic Kidney Disease: an Observational Analysis

2021 ◽  
Author(s):  
Armando Silva-Almodóvar ◽  
Edward Hackim ◽  
Hailey Wolk ◽  
Milap C. Nahata
Keyword(s):  
Chronic Kidney Disease ◽  
Logistic Regression ◽  
Kidney Disease ◽  
Female Gender ◽  
Primary Care Providers ◽  
Medication Therapy Management ◽  
Care Providers ◽  
Therapy Management ◽  
Medication Therapy ◽  
Stage 4

Abstract Background Potentially inappropriately prescribed medications (PIPMs) among patients with chronic kidney disease (CKD) may vary among clinical settings. Rates of PIPM are unknown among Medicare-enrolled Medication Therapy Management (MTM) eligible patients. Objectives Determine prevalence of PIPM among patients with CKD and evaluate characteristics of patients and providers associated with PIPM. Design An observational cross-sectional investigation of a Medicare insurance plan for the year 2018. Patients Medicare-enrolled MTM eligible patients with stage 3–5 CKD. Main Measures PIPM was identified utilizing a tertiary database. Logistic regression assessed relationship between patient characteristics and PIPM. Key Results Investigation included 3624 CKD patients: 2856 (79%), 548 (15%), and 220 (6%) patients with stage 3, 4, and 5 CKD, respectively. Among patients with stage 3, stage 4, and stage 5 CKD, 618, 430, and 151 were with at least one PIPM, respectively. Logistic regression revealed patients with stage 4 or 5 CKD had 7–14 times the odds of having a PIPM in comparison to patients with stage 3 disease (p < 0.001). Regression also found PIPM was associated with increasing number of years qualified for MTM (odds ratio (OR) 1.46–1.74, p ≤ 0.005), female gender (OR 1.25, p = 0.008), and increasing polypharmacy (OR 1.30–1.57, p ≤ 0.01). Approximately 14% of all medications (2879/21093) were considered PIPM. Majority of PIPMs (62%) were prescribed by physician primary care providers (PCPs). Medications with the greatest percentage of PIPM were spironolactone, canagliflozin, sitagliptin, levetiracetam, alendronate, pregabalin, pravastatin, fenofibrate, metformin, gabapentin, famotidine, celecoxib, naproxen, meloxicam, rosuvastatin, diclofenac, and ibuprofen. Conclusion Over one-third of Medicare MTM eligible patients with CKD presented with at least one PIPM. Worsening renal function, length of MTM eligibility, female gender, and polypharmacy were associated with having PIPM. Majority of PIPMs were prescribed by PCPs. Clinical decision support tools may be considered to potentially reduce PIPM among Medicare MTM–enrolled patients with CKD.

scholarly journals Excess Clinical Comorbidity Among HIV-Infected Patients Accessing Primary Care in US Community Health Centers

2017 ◽  
Vol 133 (1) ◽  
pp. 109-118 ◽  
Author(s):  
Kenneth H. Mayer ◽  
Stephanie Loo ◽  
Phillip M. Crawford ◽  
Heidi M. Crane ◽  
Michael Leo ◽  
...  
Keyword(s):  
Primary Care ◽  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Community Health ◽  
Community Health Centers ◽  
Primary Care Providers ◽  
Health Centers ◽  
Research Network ◽  
People Living With Hiv ◽  
Care Providers

Objectives: As the life expectancy of people infected with human immunodeficiency virus (HIV) infection has increased, the spectrum of illness has evolved. We evaluated whether people living with HIV accessing primary care in US community health centers had higher morbidity compared with HIV-uninfected patients receiving care at the same sites. Methods: We compared data from electronic health records for 12 837 HIV-infected and 227 012 HIV-uninfected patients to evaluate the relative prevalence of diabetes mellitus, hypertension, chronic kidney disease, dyslipidemia, and malignancies by HIV serostatus. We used multivariable logistic regression to evaluate differences. Participants were patients aged ≥18 who were followed for ≥3 years (from January 2006 to December 2016) in 1 of 17 community health centers belonging to the Community Health Applied Research Network. Results: Nearly two-thirds of HIV-infected and HIV-uninfected patients lived in poverty. Compared with HIV-uninfected patients, HIV-infected patients were significantly more likely to be diagnosed and/or treated for diabetes (odds ratio [OR] = 1.18; 95% confidence interval [CI], 1.22-1.41), hypertension (OR = 1.38; 95% CI, 1.31-1.46), dyslipidemia (OR = 2.30; 95% CI, 2.17-2.43), chronic kidney disease (OR = 4.75; 95% CI, 4.23-5.34), lymphomas (OR = 4.02; 95% CI, 2.86-5.67), cancers related to human papillomavirus (OR = 5.05; 95% CI, 3.77-6.78), or other cancers (OR = 1.25; 95% CI, 1.10-1.42). The prevalence of stroke was higher among HIV-infected patients (OR = 1.32; 95% CI, 1.06-1.63) than among HIV-uninfected patients, but the prevalence of myocardial infarction or coronary artery disease did not differ between the 2 groups. Conclusions: As HIV-infected patients live longer, the increasing burden of noncommunicable diseases may complicate their clinical management, requiring primary care providers to be trained in chronic disease management for this population.

scholarly journals Survey for the Prediction of Chronic Kidney Disease using Machine Learning

2019 ◽  
pp. 154-158
Author(s):  
Pooja Sharma ◽  
Saket J Swarndeep
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Kidney Diseases ◽  
Detection System ◽  
Machine Learning Techniques ◽  
Network System ◽  
Data Set ◽  
Neural Network System

According the 2010 global burden of disease study, Chronic Kidney Diseases (CKD) was ranked 18th in the list of causes of total no. of deaths worldwide. 10% of the population worldwide is affected by CKD. The prediction of CKD can become a boon for the population to predict the health. Various method and techniques are undergoing the research phase for developing the most accurate CKD prediction system. Using Machine Learning techniques is the most promising one in this area due to its computing function and Machine Learning rules. Existing Systems are working well in predicting the accurate result but still more attributes of data and complicity of health parameter make the root layer for the innovation of new approaches. This study focuses on a novel approach for improving the prediction of CKD. In recent time Neural network system has discovered its use in disease diagnoses, which is depended upon prediction from symptoms data set. Chronic kidney disease detection system using neural network is shown here. This system of neural network accepts disease-symptoms as input and it is trained according to various training algorithms. After neural network is trained using back propagation algorithms, this trained neural network system is used for detection of kidney disease in the human body.

scholarly journals Enhancing primary care capacity in chronic kidney disease management: a quality improvement educational initiative

BMJ Open ◽  
2021 ◽  
Vol 11 (11) ◽  
pp. e046068
Author(s):  
Michelle D Smekal ◽  
Aminu K Bello ◽  
Maoliosa Donald ◽  
Deenaz Zaidi ◽  
Kerry McBrien ◽  
...  
Keyword(s):  
Primary Care ◽  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Educational Intervention ◽  
Primary Care Providers ◽  
System Level ◽  
Future Research ◽  
Care Providers ◽  
Post Intervention ◽  
Interactive Workshop

BackgroundGaps in identification, medical management and appropriate referral for patients with chronic kidney disease (CKD) are evident.ObjectiveWe designed and implemented an interactive educational intervention (accredited workshop) to improve primary care providers’ awareness of tools to support guideline-concordant CKD management.DesignWe used the Kern method to design the educational intervention and targeted the accredited workshops to primary care team members (physicians, nurses and allied health) in Alberta, Canada. We conducted anonymous pre-workshop and post-workshop surveys to identify practice-specific barriers to care, identify potential solutions, and evaluate provider confidence pre-intervention and post-intervention. We used non-parametric statistics to analyse Likert-type survey data and descriptive content analysis to categorise responses to open-ended survey questions.ResultsWe delivered 12 workshops to 114 providers from September 2017 through March 2019. Significant improvements (p<0.001) in confidence to appropriately identify, manage and refer patients with CKD were observed. Participants identified several patient-level, provider-level, and system-level barriers and potential solutions to care for patients with CKD; the majority of these barriers were addressed in the interactive workshop.ConclusionsThe Kern model was an effective methodology to design and implement an educational intervention to improve providers’ confidence in managing patients with CKD in primary care. Future research is needed to determine if these perceived knowledge and confidence improvements affect patient outcomes and whether improvements are sustained long term.

Sign in / Sign up

Export Citation Format

Share Document