Today's NHS faces severe time constraints and poses the risk of short consultations and concerns about the risk of misdiagnosis or delayed treatment. These challenges are further exacerbated by limited resources and overburdened staff, leading to long wait times for patients and generic treatment strategies.
Staff can work with a superficial view of patient data, relying on basic medical histories and current test results. This lack of comprehensive data impacts their ability to fully understand patients' needs and compromises the accuracy and individualization of diagnoses and treatments. Such an approach to health, characterized by these limitations and obligations, might aptly be called “superficial medicine.”
American cardiologist and scientist Eric Topol introduced the concept of “deep medicine” in his 2019 book “Deep Medicine: How Artificial Intelligence Can Make Healthcare Again Human Again.” He critiques the U.S.'s shallow model of medicine and offers insights from his clinical and personal experiences.
Deep medicine holds the potential to revolutionize medical diagnostics, treatment effectiveness and surgical considerations. Topol presents artificial intelligence (AI) as a transformative solution to these systemic, superficial problems. He outlines what he calls the Deep Medicine Framework as a comprehensive strategy for integrating AI into various aspects of healthcare.
The deep medicine framework is based on three core pillars: deep phenotyping, deep learning, and deep empathy. These pillars are all interconnected and the introduction of this framework could improve patient care, support healthcare workers and strengthen the entire NHS system.
In-depth phenotyping
Deep phenotyping is a comprehensive picture of an individual's health data across their entire life. A deep phenotype goes far beyond the limited data collected during a normal doctor's appointment or health episode. This includes things like a person's genetic code, all of a person's DNA, and information about the body's microbes or microbiome.
It includes the so-called “exposome”, the things in the environment that a person is exposed to over the course of their life, such as air pollution. It includes markers that reveal details about the metabolic processes in a person's body and the proteins their body expresses, as well as other biological measures and metrics. It includes a person's electronic health records, including their medical history, diagnoses, treatments and laboratory results.
Deep learning
The philosophy underlying deep phenotyping is to combine this diverse data to enable more accurate and rapid diagnoses, more precise and effective treatments, and advance predictive and preventive medical strategies. However, the sheer volume and complexity of the data collected presents significant analysis challenges. This is where deep learning – an area of AI designed to simulate the decision-making power of the human brain – is so valuable. Deep learning uses an algorithm called a neural network, which uses small mathematical computers called “neurons” that are connected to each other to share information and learn.
AI could potentially improve the use of diagnostic tools. Elif Bayraktar / Shutterstock
Advances in neural network algorithms, technology, and availability of digital data have enabled neural networks to demonstrate impressive performance. For example, they have enabled the rapid and accurate analysis of medical images such as X-rays and MRI images. You can generate reports and predict disease progression and patient outcomes.
AI is proving valuable in drug discovery and identifying chemical markers in the body, such as those that can signal the presence of cancer. They can control instruments used in robotic surgery. Additionally, AI technology like the one behind ChatGPT can process medical literature and patient records to help make complex diagnoses. You can automate writing tasks like note-taking and data entry.
Deep empathy
Integrating AI systems could help streamline operational tasks in healthcare services such as the NHS. This includes bed management and hospital processes. However, the development of AI technologies should not be haphazard, but must be focused on real clinical needs and designed to promote better relationships between patients and staff. This is the pillar of deep medicine known as deep empathy.
Healthcare has increasingly become a discipline in which the human touch, once the cornerstone, is overshadowed by a relentless pursuit of efficiency. Health workers are faced with an increased burden of administrative tasks. This can result in reducing the time they devote to each individual patient and negating the essence and potential benefits of compassionate care.
Staff need the sensitivity and time to respond to the emotional and psychological needs of patients and their families. This promotes a supportive and compassionate care environment and strengthens the human connection at the heart of healthcare.
AI solutions can be designed to reduce the administrative burden on staff and open up more opportunities for meaningful patient interaction. By removing these barriers, we can focus more on direct patient care, helping to improve the quality of services provided and hopefully patient satisfaction.
There is also a transformative opportunity to reimagine efficiency and focus on relationships between patients and staff. It envisions a future in which healthcare professionals have both technical skills and emotional intelligence, and are able to meet the psychological needs of patients with genuine understanding and compassion.
Will Jones, Research Director and Lecturer in Data Science, Artificial Intelligence and Modeling (DAIM), University of Hull
This article is republished from The Conversation under a Creative Commons license. Read the original article.
