By Nick Dobrzelecki, MBA, BSN, RN, Managing Partner and co-founder
Artificial Intelligence (AI) is a hot topic, especially with the recent launch of ChatGPT’s public platform. It’s generated a lot of buzz and conversations with friends and family about the amazing projects being developed specifically in the healthcare industry that most of the public are not aware of. NLP Logix for example, a leading company in this field, has been working on AI advancements for over a decade, and their work never fails to leave me awestruck when I visit. Although AI advancements in healthcare have been happening for years, it seems that we are now on the brink of a major breakthrough. The impact of AI on healthcare is poised to change how healthcare is managed and delivered, but with such power comes the need for regulation to ensure patient safety, privacy, and ethical considerations.
AI is currently being used in healthcare in various ways, including:
- Medical image analysis – AI algorithms, specifically Deep Learning approaches, have been shown to help radiologists improve diagnoses and patient outcomes by providing highly accurate interpretations of medical images, such as magnetic resonance imaging (MRI), computed tomography (CT), and X-rays.
- Chatbots and virtual assistants – These artificial intelligence-based programs can help patients receive immediate medical advice, schedule appointments, and access medical information. Chatbots in healthcare offer a range of benefits including improved efficiency, reduced wait times, and quicker diagnosis.
- Predictive analytics – AI is helping healthcare providers identify patients at risk of developing certain conditions, giving them the ability to intervene early. For example, predictive analytics can be used to predict which patients are at risk of hospital readmission, allowing healthcare providers to intervene before the patient’s condition exacerbates. This technology has the potential to revolutionize healthcare by facilitating earlier and more accurate diagnoses, improving treatment outcomes, and reducing healthcare costs.
- Drug discovery – AI is being used to speed up the drug discovery process and identify new treatments for diseases. AI algorithms can quickly analyze vast amounts of data such as molecular structures, genomic data, and clinical trials, to identify promising drug candidates, predict their efficacy and toxicity, and optimize their properties. Machine learning techniques can help in identifying patient populations that may respond better to certain treatments, allowing for personalized medicines.
We expect that in the near future AI’s use in healthcare will become even more independent and more advanced, leading to what once seemed like science fiction, such as AI-powered wearables, precision prophylactic medicine based on genetics, predictive modeling that warns us of pending pandemics and outbreaks, and sophisticated ChatGPT engines to provide clinical decision support to a patient’s human medical teams.
A side note about ChatGPT: By following Moore’s Law, which posits that computational progress doubles roughly every two years, ChatGPT will be 32 times more powerful in a decade and over 1,000 times more powerful in two decades. This revolution will enable tomorrow’s technology to match the diagnostic skills of clinicians today by emulating how doctors make clinical decisions and by providing 24/7 medical assistance. AI doesn’t need to sleep.
Regulation is Needed
While there is a lot of promise and excitement around what AI can do, there is a lot of concern about AI becoming too powerful if it’s not regulated. In the world of analytics: “Bad data in, bad data out.” In the medical industry, using bad data in AI algorithms could result in catastrophic outcomes, including adverse events and deaths. Human medical teams must fully understand how the AI arrived at its conclusions and not simply follow its recommendations.
The development and implementation of AI in healthcare must be regulated to ensure patient safety, privacy, and ethical considerations. In June 2021, the World Health Organization (WHO) issued a report on the use of AI in healthcare and offered six guiding principles for AI regulation: protecting autonomy; promoting safety; ensuring transparency; fostering responsibility; ensuring equity; and promoting sustainable AI.
We agree that regulations on AI in medicine are necessary to ensure quality datasets, ensuring that the data used to train and test artificial intelligence models are unbiased and free from error. Before going down the path of providing clinical support, AI needs to be trained on best practices from only high quality data. Humans must never give up oversight of the datasets used in the healthcare industry. Additionally, AI models also need to represent a fair and accurate reflection of the real world, eliminating potential biases in the data that could lead to unfair medical treatment or inaccurate predictions. Lastly, the highest levels of cybersecurity are needed to ensure that systems are secure and patients’ data is protected.
With global agreement on AI regulation, the technology no doubt has the potential to ensure a positive, not negative, effect on patients around the world.