AI-powered ‘deep medicine’ could transform healthcare in the NHS


Today’s NHS faces severe time constraints, with the risk of short consultations and concerns about the risk of misdiagnosis or delayed care. These challenges are compounded by limited resources and overstretched staff that results in protracted patient wait times and generic treatment strategies.

Staff can operate with a surface level view of patient data, relying on basic medical histories and recent test results. This lack of comprehensive data interferes with their ability to fully understand patient needs and compromises the accuracy and individualisation of diagnoses and treatments. Such a healthcare approach, characterised by these limitations and engagements, could aptly be termed “shallow medicine.”

The American cardiologist and scientist Eric Topol introduced the concept of “deep medicine” in his 2019 book Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again. He critiques the US’s shallow medicine model, offering insights from his clinical and personal experiences.

Deep medicine holds the potential to revolutionise medical diagnostics, the effectiveness of treatments, and operational considerations. Topol presents artificial intelligence (AI) as the transformative solution to these systemic shallow issues. He outlines what he calls the deep medicine framework as a comprehensive strategy for the incorporation of AI into different aspects of healthcare.

The framework of deep medicine is built upon three core pillars: deep phenotyping, deep learning, and deep empathy. These pillars are all interconnected and adopting this framework could enhance patient care, support healthcare staff, and strengthen the entire NHS system.

Deep phenotyping

Deep phenotyping refers to a comprehensive picture of an individual’s health data across a full lifetime. A deep phenotype goes far beyond the limited data collected during a standard medical appointment or health episode. It includes things such as a person’s genetic code, the entirety of an individual’s DNA, and information about the body’s microbes or microbiome.

It encompasses what’s known as the “exposome,” the things in the environment that a person is exposed to during life, such as air pollution. It includes markers that reveal details of the metabolic processes going on in a person’s body and the proteins their body is expressing, as well as other biological measures and metrics. It comprises a person’s electronic health records, including their medical history, diagnoses, treatments, and lab results.

Deep learning

The philosophy underpinning deep phenotyping is to combine this diverse data to enable more accurate and speedy diagnoses, precise and effective treatments, and to advance predictive and preventative medicine strategies. However, the sheer volume and complexity of the collected data pose significant challenges for analysing it all. This is where deep learning — an area of AI that seeks to simulate the decision-making power of the human brain — is so valuable. Deep learning uses an algorithm called a neural network that uses little mathematical computers, called “neurons,” that are connected to one another to share and learn information.

CT scans of the brain.