Blood tests are currently one-size-fits-all – machine learning can figure out what exactly is ‘normal’ for each patient. Health News

But despite the test’s ubiquity, the way clinicians interpret and use it in the clinic is less than ideal. Currently, blood test readings are based on one-size-fits-all reference intervals that do not account for individual differences.

I’m a mathematician at the University of Washington School of Medicine, and my team studies ways to use computational tools to improve. Clinical blood tests. To develop better ways to capture individual patient definitions of “normal” lab values, my colleagues at Harvard Medical School’s Higgins Lab and I examined 20 years of blood count tests from thousands of patients on both the East and West coasts.

In our recently published research, we used machine learning to identify healthy blood count ranges for individual patients and predict their future risk of disease.

Clinical examination and complete blood count

Many people usually think of clinical trials as purely diagnostic. For example, a COVID-19 or pregnancy test comes back as positive or negative, telling you whether you have a particular condition. However, most tests do not work this way. Instead, they measure a biological characteristic that your body constantly regulates up and down to stay within certain limits.

Your complete blood count is also a continuum. The CBC test Creates a detailed profile of your blood cells – such as how many red blood cells, platelets and white blood cells are in your blood. These markers are used every day in almost all areas of medicine. For example, hemoglobin is an iron-containing protein that allows your red blood cells to carry oxygen. If your hemoglobin level is low, it means you are iron deficient.

Platelets are cells that help blood clot and stop bleeding. If your platelet count is low, it means you may have had some internal bleeding and your body is using platelets to help form a blood clot to plug the wound. White blood cells are part of your immune system. If your white blood cell count is high, it may mean you have an infection and your body is producing more of these cells to fight it.

Normal ranges and reference intervals

But all of this begs the question: Is a blood test really considered too high or too low? Traditionally, physicians determine what are called reference intervals by measuring blood tests in a range of healthy individuals. They usually take the middle 95% of these healthy values ​​and call it “normal,” being too low or too high above or below. These general ranges are used almost everywhere in medicine.

But reference intervals face a bigger challenge: what’s normal for you may not be normal for someone else.
Almost all blood count markers are heritable, meaning that your genetics and environment determine what a healthy value for each marker will be for you.

At the population level, for example, a normal platelet count is between about 150 and 400 billion cells per liter of blood. But your body wants to maintain a platelet count of 200 – this value is called your set point. This means your normal range may be only 150 to 250.

Differences between a patient’s actual normal range and a population-based reference interval can create problems for doctors. They may be less likely to diagnose the disease if your set point is far from the cutoff. On the contrary, they can run Unnecessary tests If your set point is too close to the cutoff.

Learn all about CBCs (Source: Getty Images/Thinkstock)

Defining what is normal for you

Fortunately, most patients get blood counts every year as part of a routine checkup. Using machine learning models, my team and I were able to estimate blood count set points based on over 50,000 patients’ clinic visit histories. This allowed us to study how the body regulates these set points and test whether we could develop better ways to personalize lab test readings.

Over several decades, we found that individual normal ranges were about three times smaller than at the population level. For example, while the “normal” range for a white blood cell count is around 4.0 to 11.0 billion cells per liter of blood, we found that many people’s individual range was much narrower, such as 4.5 to 7, or 7.5 to 10.

When we used these set points to interpret new test results, they helped improve the diagnosis of diseases such as iron deficiency, chronic kidney disease and hypothyroidism. We were able to note when someone’s result was outside their small personal range, potentially indicating a problem, even if the result was within the normal range for the population as a whole.

The set points themselves were strong indicators of future risk of developing the disease. For example, patients with high white blood cell set points are more likely to develop type 2 diabetes in the future. They were also nearly twice as likely to die from any cause as similar patients with low white cell counts. Other blood count markers were also strong predictors of future morbidity and mortality risk.

In the future, doctors could potentially use set points to improve disease screening and how they interpret new test results. This is an exciting opportunity for personalized medicine: to use your own medical history to define what healthy means for you.