Narus | Generative AI in Healthcare

The world of healthcare is changing, with generative AI (GenAI) coming into play across many different fields. Here, we’ll cover the benefits of GenAI in healthcare, as well as the limitations and challenges that must be considered for its ethical and responsible use.

The benefits of GenAI in healthcare

GenAI can be applied across various healthcare domains, offering many great opportunities to enhance patient care, optimise clinical workflows, and advance medical research. Here are some ways GenAI can benefit the field of healthcare:

Increase administrative efficiency

One of the most pressing issues in healthcare is the administrative burden placed on medical professionals. Forbes says that in the US, primary care physicians report spending 2.1 additional hours on paperwork for every 8 hours of patient visits.

GenAI offers solutions to streamline these processes, reducing the time spent on paperwork and administrative tasks. For instance, GenAI can automate clinical documentation, allowing healthcare providers to focus more on patient care.

One study, conducted by the nonprofit Phyx Primary Care, demonstrated the advantages of AI assistants for primary care providers, which reportedly decreased time spent on documentation by 41% and time spent on work after-hours by 37%.

Support with diagnoses and treatments

GenAI can assist in clinical diagnosis by generating high-quality medical images from poor-quality scans, enhancing diagnostic abilities. Advanced techniques like image synthesis, denoising, and reconstruction can provide clinicians with clearer, more detailed visuals, enabling healthcare professionals to analyse images such as X-rays, MRIs, ultrasounds, and more with greater accuracy and confidence.

GenAI can also create personalised treatment plans by analysing patient data, such as electronic health records, genomic information, and clinical notes. By sifting through vast and varied medical datasets and uncovering hidden patterns and insights, it can forecast how diseases might progress and tailor treatment recommendations to each patient's unique profile and medical background.

Advance research and development

GenAI can help speed up drug discovery by making it easier to identify targets, test compounds, find promising leads, and evaluate their effectiveness in the early stages. According to a report by McKinsey & Company, combining GenAI with other well-known AI methods like computer vision, virtual screening, and knowledge graphs could potentially help researchers find new medicines in about half the time it usually takes.

The report also found that GenAI could improve efficiency throughout the whole clinical development process, cutting costs by up to 50% through streamlined clinical trials and automated trial document drafting, as well as reducing trial timelines by over 12 months.

Improve the well-being of healthcare providers

The integration of GenAI into healthcare systems has shown promising results in reducing burnout among medical professionals. This was demonstrated in the study by Phyx Primary Care, where the surveyed physicians reported a 60% decrease in burnout after implementing the AI assistant.

Strengthen communication with patients

Interestingly, the same survey also demonstrated an improvement in the attentiveness of healthcare providers during patient visits, with physicians reporting a 32% decrease in visits that felt rushed.

According to one of the physicians surveyed by Phyx Primary Care, “I don't worry about getting details documented during the visit because I know the AI is doing it for me. I can sit and face the patient most of the time. They like that better.”

GenAI can also generate discharge summaries and instructions in patients' native languages, ensuring effective communication regardless of language barriers.

Limitations and ethical considerations

While GenAI is showing great promise in improving some areas of healthcare for both patients and providers, there are limitations that are vital to consider in order to use GenAI ethically and responsibly in this field. Let's go through them:

Bias

One of the foremost ethical concerns surrounding the use of AI in healthcare is the potential for bias, which occurs when AI models produce content that reflects systematic errors or distortions that often mirror human biases.

AI bias is especially dangerous when it can affect people's lives, such as their access to healthcare. This happened in the US when a major healthcare risk algorithm presented a racial bias which underestimated the health needs of Black patients.

It's crucial to ensure that AI algorithms are developed and implemented in a way that promotes equitable healthcare outcomes across diverse populations.

Hallucinations

While GenAI has shown remarkable capabilities, it's important to acknowledge the potential for these systems to generate inaccurate or misleading information, a phenomenon known as "hallucinations". In healthcare, where decisions can have life-altering consequences, the risk of AI hallucinations is particularly concerning.

To mitigate this risk, healthcare providers must implement robust validation processes. This includes cross-referencing AI-generated information with established medical literature, peer-reviewed studies, and expert opinions. Educating healthcare professionals about the limitations of AI, including the potential for hallucination, is also crucial.

Privacy and data security

The protection of patient privacy and data security is paramount when implementing AI in healthcare.

As pointed out in a review published in the US National Library of Medicine, “Unauthorized access to patient data can result in the breach of confidentiality, identity theft, or misuse of sensitive medical information, posing significant risks to patient autonomy and trust in the healthcare system.”

AI systems must comply with strict regulations like HIPAA and GDPR to protect sensitive patient information.

Maintaining trust

Nothing can replace the human touch when it comes to building and maintaining trust in the patient-healthcare provider relationship. While GenAI can assist in many areas of healthcare, it cannot replicate empathy, understanding, and personal connection.

Healthcare organisations should integrate AI tools thoughtfully, balancing innovation with a commitment to human-centred care. By educating patients about AI's role and addressing their concerns, providers can reinforce trust while ensuring that technology enhances rather than replaces human connection.

Future Outlook

GenAI is continuously developing more capabilities and (with the right regulations and considerations in place) it can continue to grow as a useful tool in healthcare, accelerating areas like research and freeing up more time for healthcare providers to focus on where human connection, experience, and expertise are needed.