Disease Prediction Use Case: Renal Problems

More than 1 in 7 adults in the United States – approximately 37 million people – are estimated to have chronic kidney disease (CKD), according to the Centers for Disease Control and Prevention [i]

If left untreated, CKD can increase a person’s chances of developing other health problems like heart disease, stroke, kidney failure, early death, and more. [ii]

Renal Problems, AI, Lumiata

What if healthcare providers could predict patients who are likely to develop chronic kidney disease and other renal issues? By identifying these patients before disease onset, providers could improve care and help keep patients healthy. 

Lumiata makes this possible.

Our AI-powered data models predict the onset of diseases like CKD within the next 12 months, enabling providers to identify high-risk patients faster, intervene sooner, improve care, and help prevent patients from developing CKD and related health issues.

Here’s a closer look at how Lumiata’s predictive analytics can help in the case of chronic kidney disease and other renal problems.

How Lumiata works 

In order to generate high-quality predictions, you need to start with high-quality data. That’s why before you even begin running data models, we transfer your EHR data to our ecosystem and perform data quality checks. During this step, we look for basic B2B statistics and metrics around the data and provide you with a holistic data assessment document of our findings.

We then validate, clean, and normalize your data. This process includes validating and normalizing important objects like ICD or NDC codes using our reference coding systems data. We also leverage our data model cross coding normalization feature to transform healthcare-related codes from one coding system to another to ensure all data use the same coding system.

Our system also cleans your data by using smart cleaning algorithms to remove data that is incomplete, incorrect, duplicate, or improperly formatted. 

Following the cleaning process, your data goes through our disease tagging system, during which leading indicators for chronic conditions are identified. We also enrich your data with our data, which is derived from 35,000 physician curation hours, 130 million patient records, and 50 million published articles. 

Finally, we transform the dataset into the Lumiata Person360 object, at which point your data is optimized for machine learning consumption. This is where you can use our Lightning AI Model Builder to build and run data models or use our Spectrum AI Model Catalog to run ready-made financial and clinical data models, including our disease onset model.

AI, Renal Problems, Lumiata

Predict disease onset and identify high-risk patients 

Lumiata has a huge library of clinical and financial data models that provide actionable insights into your data. One such model is the disease onset model, which can be used to predict the onset of chronic kidney disease – and a number of other chronic and acute diseases – within the next 12 months.

High blood pressure and diabetes are common causes of CKD in adults, with other risk factors including heart disease, obesity, a family history CKD, old age, and more. [iii]

Improve care and achieve better outcomes 

When you know which patients are at risk of developing chronic kidney disease and other renal issues, you can provide more specialized care in a timely manner and achieve better outcomes for your patients. This may include suggesting treatment based on the patient’s specific risk factors and health status, providing the patient with guidance on steps they can take at home to improve their health, and more. 

For example, if Lumiata predicts the onset of CKD in a patient, and you see that the patient has diabetes, you can take a closer look at how their diabetes and any other risk factors are being managed and take steps to improve the patient’s health, helping them avoid CKD. 

If the patient eventually does develop chronic kidney disease, this may put them at risk of developing other health issues like heart disease, stroke, high blood pressure, and more. You can continue using Lumiata to predict the onset of these related health problems and stay aware of the patient’s risk factors. This empowers you to continue providing specialized care and help prevent your patients from experiencing further conditions.

Reduce costs for your healthcare organization 

By helping patients avoid chronic kidney disease and other renal issues and related complications, you help them avoid costly treatments, hospital admissions, procedures, and more. This helps your organization reduce costs as well as potentially avoid penalties from hospital readmissions. Patients benefit by avoiding the financial, physical, and mental stress of undergoing expensive procedures and treatments or staying in the hospital. Payer organizations also benefit by not having to cover the cost of these treatments and hospitalizations. 

About Lumiata 

AI, Renal Problems, Lumiata

Lumiata uses artificial intelligence and machine learning to make predictions and provide actionable insights for your healthcare organization. Whether you want to identify patients at risk of developing diseases, improve care management, enhance your underwriting strategy, improve data management, or something else, Lumiata can make that happen. 

Our suite of clinical models is up to 65% more accurate than the leading carrier, and our suite of financial models is up to 80% more accurate than the closest competitors’ cost prediction models. Even if you don’t have a data science team, our AI solutions can help you solve some of today’s biggest healthcare cost and care challenges. 

Our vision is to democratize AI in healthcare to reduce costs and improve the quality of care. Ready to see what Lumiata can do for your organization? Request a demo.