The evolution of artificial intelligence (AI) in healthcare is nothing short of extraordinary. It is becoming a dominant force in the healthcare industry and advancing at an unprecedented pace. With 2022 being a game-changing year for AI in healthcare, we wonder what’s more to come.
To understand how AI is reshaping the future of healthcare and what we can expect, healthcare experts from a16z, Vijay Pandey and Marc Andreessen, sat down for a wide-ranging discussion on AI as an additive superpower.
In this article, we’ll delve into expert opinions and discuss the key takeaways from the discussion.
Evolution of AI in healthcare
The past few months have seen dramatic—almost magical—generative AI applications released for public use.
Marc Andreessen, Co-founder of a16z, observed these remarkable advancements and drew parallels with Vladimir Lenin’s famous quote—” weeks in which decades happen.” He says just like software, which has become an integral part of every industry in the past decade, AI is on the verge of reshaping every aspect of healthcare.
There have been many breakthrough moments in recent years, such as the 2012 breakthrough in image recognition, which led to the development of self-driving cars. Then there was a natural language breakthrough three years ago, which catalysed a whole new range of applications, including GPT and text generation, better transcription, improved speech synthesis and an artistic revolution in the image and video creation.
And now it seems that every week there are fundamental breakthroughs, research papers and product releases coming out, creating a cascading effect that is fascinating to watch.
Current capabilities of AI in healthcare
Apart from GPT-3’s ability to take exams and write code, AI can generalise and apply knowledge in various domains. It has the potential to excel in areas where humans may have limitations, such as data analysis and pattern recognition.
With its ability to process and analyse large amounts of data quickly and accurately, AI can assist healthcare professionals identify patterns that may not be easily noticeable to humans. This can lead to improved diagnosis, effective treatment planning and more personalised care.
Limitations of the current healthcare system
In the current healthcare system, doctors face several challenges in diagnosing and treating patients, like:
Limited time
Research shows that the median healthcare experience is just 15 minutes, leading to rushed diagnoses and treatments, which may result in missed or incorrect diagnoses.
Availability and accuracy of data
Doctors rely on their own knowledge, experience and patient medical records to diagnose and prescribe treatments. However, accessing and interpreting medical records is time-consuming and complex. Also, the records may not always be complete or up-to-date, impacting the accuracy of diagnoses and treatment decisions.
Compliance with prescribed treatments
Healthcare providers don’t always know whether patients are adhering to their prescribed treatments, affecting the effectiveness of the treatment plan. Tracking drug interactions and evaluating treatment success is also a challenge due to limited data on patient compliance.
Marc compares the existing healthcare system with human car drivers. While regulations and standards are in place, limitations and challenges can result in errors and inefficiencies. He believes that the use of AI in healthcare, with access to comprehensive medical data and the ability to process and analyse it, could potentially improve the accuracy and efficiency of diagnoses and treatments.
AI, healthcare jobs and the future role of Doctors
Will AI replace doctors?
While there are concerns about the technological displacement of jobs, history has shown technology creates new jobs that were not possible before. Rather than simply automating existing tasks, AI technology creates new opportunities and transforms industries.
For instance, software developers, engineers, testers and project managers are a few of the many roles that emerged as a result of the growth of the software industry.
“AI is not replacing doctors, it’s augmenting doctors.”
When it comes to healthcare, Marc says AI will change the role of doctors. The use of technology will enable doctors to have more time with patients, access a wealth of data and focus on behavioural health issues such as mental health, addiction, and lifestyle modifications, which are often overlooked in traditional healthcare settings. This shift in roles can lead to improved patient care and better utilisation of human expertise.
Here are a few aspects of healthcare AI will reshape in the future:
AI in healthcare data analytics
AI-powered tools will assist doctors in gathering and analysing patient data more efficiently, leading to more accurate diagnoses and personalised treatment plans. By analysing vast amounts of data, AI tools can also identify patterns and risk factors, enabling early interventions and preventive measures.
AI for patient compliance
These tools and platforms can improve compliance with treatment plans and medication adherence, allowing behavioural health interventions.
For instance, mobile apps, wearables and telehealth platforms provide remote monitoring and support for patients, leading to better management of their conditions and improved outcomes.
AI in the healthcare bureaucratic process
Marc discusses how AI is revolutionising the bureaucratic process in healthcare. One example he highlights is doctors having to write letters to insurance companies for reimbursement, which can be time-consuming and labour-intensive.
AI can streamline this process by automating letter-writing and document submission, reducing errors and inconsistencies and improving efficiency. It will also reduce the administrative burden on healthcare providers and allow them to focus more on patient care.
He says hospitals and clinics that have already implemented AI-powered solutions for insurance claims have reported reduced processing times and increased reimbursement rates.
Regulatory challenges with AI in healthcare
Vijay and Marc believe that the regulatory landscape for AI in healthcare is different from the history of healthcare, as it presents unique challenges that need to be addressed. The challenges of regulating patient access to information on the internet, commonly referred to as Dr. Google. They are concerned about the reliability and accuracy of the information available online and the potential for misinformation or misinterpretation.
While patients being educated and engaged in the healthcare decision-making process can empower them and improve healthcare outcomes, it can also lead to increased expectations, demands and challenges for healthcare providers.
AI expectations and real-world scenarios
Patients, more often than not, expect perfection with AI in healthcare and set unrealistic expectations. However, Vijay and Marc highlight that AI is not infallible and can have limitations and potential risks.
It’s important to note that the current generation of AI is not creating new information or thinking on its own. It is trained on existing human knowledge and data, and its capabilities are based on the patterns and information it has been exposed to.
While AI can excel in many areas, it still lacks the nuanced understanding and context humans possess, making some tasks difficult for it to perform accurately.
There are also concerns about the accuracy and reliability of AI-generated documents, the ethical implications of using AI in healthcare, data privacy, security, biases and the loss of human touch.
Marc says negative headlines and moral panic can arise from a single instance of AI in healthcare going wrong, emphasising the need for responsible and cautious use of AI. He asserts the importance of properly containing expectations and not spinning off into fantasy narratives when discussing the possibilities of AI in healthcare.
Bottom line
The expert discussion on AI in healthcare highlights its benefits and challenges. The use of AI in healthcare has the potential to address current healthcare challenges and improve patient care, but further research, regulation and collaboration among stakeholders are needed to ensure responsible and ethical integration of AI into the healthcare system.
It’s essential to consider the complexities of the industry and associated concerns to ensure proper oversight and regulation to address any potential challenges. Moving forward, striking a balance between leveraging AI’s potential and upholding human values and ethics will be crucial.
“AI has the potential to be a true superpower for healthcare, unlocking new insights and capabilities that we never thought were possible.”
