Artificial intelligence in healthcare is moving beyond treatment to prevention

Artificial intelligence (AI) and machine learning have gained momentum in recent years, driving significant changes in traditional lifestyles. Advancements in AI have greatly influenced social change and innovation in public health. The technology played a key role during the COVID-19 outbreak, helping to speed up virus diagnosis, detect clusters of cases, monitor people’s movement, predict and track potential future patients, and more.

Due to the unprecedented global health emergency, public healthcare is being discussed today more than ever. Technology is seen as the driving force behind improvements and has successfully transformed various aspects of human life.

It is anticipated that technologies such as AI, machine learning, natural language processing, deep learning and generative AI can be utilised to transcribe patient interactions, analyse vast amounts of unstructured data, and prepare medical reports. They can provide opportunities for the public health system to enhance existing healthcare services and offer highly accessible options at reduced costs.

Artificial intelligence can dramatically influence healthcare. It can reduce operating costs and help existing systems scale by improving the operations of healthcare facilities. One estimate from McKinsey predicts that big data could save medicine and pharma up to $100 billion annually across the United States healthcare system, by optimising innovation, improving the efficiency of research and clinical trials, and developing more individualised approaches through tools for physicians, consumers, insurers, and regulators.

The true impact will be in patient care — giving providers better and more personalised access to information across medical facilities to help inform patient care will be life-changing.

Deep learning, a subset of machine learning based on artificial neural networks, helps identify likely causes quickly by processing large amounts of information, drastically reducing the diagnosis-to-treatment cycle. Early diagnosis through these techniques increases the chances of successful treatment.

The power of AI has also been employed to assist surgeons during operations, allowing them to work with greater precision and flexibility, resulting in fewer mistakes and smaller scars. This helps patients recover more quickly when operated on through AI-assisted robotic surgery.

Virtual nursing assistants, available 24/7, can answer questions and monitor patients. Advanced implementations of this technique can even provide wellness checks through speech recognition. The use of AI in drug discovery is helping pharmaceutical companies streamline both discovery and repurposing. 

Many pharma giants have partnered with AI service providers such as IBM Watson and Exscientia’s artificial intelligence to drive their oncology drug discovery programmes. An automatic speech recognition service named Amazon Transcribe Medical makes it easy to quickly create accurate transcriptions from medical consultations between patients and physicians in real time, improving the efficiency of clinical documentation.

While there are many remarkable benefits offered by AI, there are also a few risks. The most obvious risk is that AI systems may sometimes go wrong, leading to patient injury or other healthcare problems. At the core of AI systems are the datasets through which the AI is trained, and training with particular types of datasets can introduce bias.

For instance, if the data available for AI is primarily gathered in medical centres located in urban areas, the resulting systems will be less effective in treating populations from rural regions.

Training AI systems requires large amounts of data from different sources such as health records, pharmacy records, and insurance claims, which are typically fragmented across many different systems. Another set of risks arises around privacy. Some patients may be concerned that such data collection could violate their privacy, as AI can predict private information about patients even if the algorithm never directly received that information.

AI has enormous potential to improve the preventative component of public health in addition to its uses in diagnosis and treatment. Predictive models, for example, can assist local and governmental entities in anticipating disease outbreaks by using behavioural, meteorological, and environmental data, allowing for prompt interventions.

Similarly, AI can support large-scale health communication efforts by analysing data at the community level to identify awareness gaps and tailor messaging to specific groups. These innovative applications of AI will be essential in shifting the focus from merely treating patients to building stronger, healthier communities.

AI is emerging as an important tool that can assist public health authorities in reducing health inequality across populations. It is set to revolutionise the public healthcare system, as it holds enormous promise for reshaping the provision of healthcare services in resource-constrained settings.

AI can transform healthcare in many ways, from discovering new medicines to improving diagnostic accuracy, helping healthcare workers become more efficient, productive, and precise. If used carefully, it could significantly reshape how the entire healthcare system operates.

Rizwan is a technology specialist and is currently pursuing an MSc in Social Innovation and Entrepreneurship at the London School of Economics and Political Science (LSE).

Views expressed are the author’s own and don’t necessarily reflect those of Down To Earth