[An image of AI technology being used to analyze data about the body. Photo Credit: Pixabay]

A groundbreaking artificial intelligence (AI) model developed by researchers at Harvard Medical School is being used to accelerate the diagnosis of rare diseases.

The model, called popEVE, was published in the journal Nature Genetics, and represents a significant advancement in how genetic data can be interpreted to help clinicians identify rare conditions more accurately and quickly.

According to Debora Marks, co‑senior author of the study and a professor of systems biology at the Blavatnik Institute at Harvard Medical School, “Our goal was to develop a model that ranks variants by disease severity — providing a prioritized, clinically meaningful view of a person’s gemome.”

This technological advancement has the ability to reduce years of uncertainty for patients by giving them clearer answers about their conditions.

The model analyzes patient symptoms and genetic data to help doctors identify diseases that are often missed in the early stages.

A rare disease is defined as a disorder, illness, or condition that affects fewer than 200,000 individuals.

According to the National Organization for Rare Disorders (NORD), there are over 10,000 rare diseases that together affect more than 30 million Americans.

Despite the term “rare,” thousands of rare diseases exist worldwide. 

Many people live with these conditions, yet doctors often have limited knowledge about them. 

Consequently, rare diseases can be extremely difficult to diagnose and treat.

Symptoms of rare diseases often resemble those of common illnesses, leading to misdiagnosis.

Some patients must visit multiple doctors before receiving an accurate diagnosis, a process that can be both stressful and dangerous.

AI is especially beneficial for rare diseases because it can compare patient information with data from many other cases and identify patterns that may be difficult for physicians to recognize.

This allows doctors to consider conditions they might not normally suspect.

AI can also help researchers study rare diseases more quickly and efficiently to accelerate the search for potential treatments.

An example of a rare illness is Kawasaki Disease (KD), primarily affecting children under the age of five.

If left untreated early, KD can cause inflammation of the blood vessels and lead to serious heart complications.

Common symptoms of KD include a long-lasting fever, red eyes, skin rashes, and swollen hands and feet, which are similar to other childhood infections.

This similarity often leads to KD being overlooked, making it difficult to diagnose. 

AI has been used to help doctors recognize KD earlier by analyzing symptoms and laboratory results.

Early diagnosis is very important because proper treatment can reduce the risk of fatal outcomes and long-term heart damage.

Machine learning, a type of AI, has been developed to facilitate earlier diagnosis, particularly among young Asian patients.

One study found that machine learning models could differentiate KD from other febrile conditions with over 90% sensitivity and specificity, indicating that AI may help clinicians detect rare diseases earlier and more reliably.

AI is expected to become a valuable tool in improving KD management in clinical settings while increasing diagnostic accuracy and effectiveness.

However, AI has its limitations.

AI systems need large amounts of accurate data, and rare diseases often do not have enough of this data available.

There are also concerns about privacy and keeping patient medical information secure.

For these reasons, experts emphasize that AI should be used to support doctors rather than replace them.

In the end, rare diseases are often challenging to diagnose and treat, but artificial intelligence is helping improve healthcare outcomes.

AI enables faster diagnosis and supports research efforts, as demonstrated by its role in identifying Kawasaki disease.

While AI is not perfect, it has the potential to improve the lives of people with rare diseases in the future.