Latest News

Cricket Health Announces Its New Machine Learning Model For Chronic Kidney Disease Detection And Risk Evaluation

Cricket Health, a wide-ranging kidney care provider company has published a white paper today on “Machine Learning for “Chronic Kidney Disease Detection & Risk Stratification.” This is an announcement for a new machine learning model to help patients at high risk of having or living with chronic kidney disease (CKD).

The new model is designed by Cricket Health’s interdisciplinary data team. It uses innovative technology to detect predictive model marks which accurately tell the risk of chronic kidney disease in a patient. This new model doesn’t even need the patient’s electronic health record (EHR) or their clinical and laboratory test reports.

There are so many people that are living with chronic kidney disease (CKD) without even realizing it. They won’t be able to detect it unless there they get into a severe problem of kidneys. Unfortunately, most of these problems cause irreversible damage and there is nothing that could treat it at that time.

But it is now probable to avoid this situation because Cricket Heath’s model will identify people at the highest risk of CKD. This way, it is possible to control disease progression at early stages.

Diagnosis of CKD and end-stage renal disease (ESRD) is incredibly expensive in the USA. It has become highly prevalent in the last few years and now more than 35 million US adults are suffering from CKD. There are millions of others who are at risk of developing it soon.

More than half a million people are living with ESRD, simply meaning that they have suffered from kidney failure at least once. The high prevalence of both these conditions is growing larger and is expected to grow more in the future.

Medicare alone is spending over $110 billion to care for CKD and ESRD patients. Yet nine out of ten persons at stage 1-3 of chronic kidney disease are not aware of having it. An even worrisome fact is that half of the patients suffering from severe loss of kidney function but not on dialysis are still unaware of it.

This little to no awareness of kidney disease is because CKD shows almost no symptoms until it completely damages the kidneys. Also, it is not diagnosed in early stages which speed up the disease progression and eventually lead to chronic stage infection.
Cricket Health is a San Francisco based company that follows a personalized and evidence-based approach to detect chronic kidney disease (CKD) and end-stage renal disease (ESRD). It identifies patients that are at a high risk of these kidney diseases and provides them patient-centered, medical care. It educates them on the disease, treatment options and even offers support i.e. transplant support, home-based dialysis, conservative care, or even in-center dialysis facilities.

This machine-learning model is helpful to predict the estimated glomerular filtration rate (eGFR) in patients. The eGFR is a common proxy to check kidney health accurately. There is no need of getting a lab test or patient involvement. Cricket Health experts can run this model frequently on a nominal or no cost for the entire affected community.

The predictive model uses a variety of predicting factors, such as demographics, comorbidity data, and utilization patterns of a patient. The whitepaper by the company explains the development and training of this machine learning algorithm and now it could be applied to get a predictive value. The model is developed in 2019 by experts from clinical/medicinal, epidemiology, statistics, and data science field.

To know more about Cricket Health’s proprietary machine learning model, click here to read the full white paper.

Areeba Hussain

The author is a fulltime medical and healthcare writer. She graduated in Medical Microbiology and Immunology with distinction. Her areas of prime interest are medicine, medical technology, disease awareness, and research analysis. Twitter @Areeba94789300

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *


Adblock Detected

Please consider supporting us by disabling your ad blocker