The impact of artificial intelligence in healthcare: opportunities and challenges

As humanity evolves, new technologies are constantly being created. The 1990s were marked by the popularization of the internet, web browsers, and cell phones. In 2007, Apple’s launch of the iPhone revolutionized mobile technology and smartphones. During this same period, cloud computing and social networks emerged. Currently, ChatGPT, launched two years ago, is considered one of the greatest recent technological advancements, popularizing artificial intelligence (AI) – a technology that has driven significant advances in areas such as healthcare, transforming diagnoses, treatments, and data management. And, like everything revolutionary, new challenges also arise, such as privacy and data security.

According to a mapping conducted in 2023 by the National Association of Private Hospitals (Anahp) and the Brazilian Association of Health Startups (ABSS), which interviewed representatives from hospitals across all regions of Brazil except the North region, 62.5% of institutions claimed to use AI in some way. Of this total, 10% use AI to support clinical decision-making and 8% in the analysis of medical images. AI has improved accuracy in image diagnosis, helping to detect diseases such as cancer, cardiovascular diseases, and eye diseases. AI-based tools, such as those developed by Google Health, can analyze X-rays, CT scans, and MRIs with high precision.

Regarding early disease detection, machine learning algorithms have been used to identify early signs of diseases in medical images, often with comparable or superior accuracy to human radiologists. In addition, other exams use AI to assess the likelihood of someone having cancer or identify tumors in early stages, increasing the cure rate and enabling more specific, less aggressive treatments, and occasionally with less financial impact.

Another opportunity that goes hand in hand with AI – and that is already being widely used – is the approach to personalized medicine, which is based on the analysis of large data sets, including genomic information, clinical data, medical history, and test results. AI can not only process and analyze this data quickly and efficiently, identifying patterns and correlations that may not be apparent to humans, but also help doctors select the most appropriate and personalized treatment for each patient. This includes predicting the effectiveness of different therapies and identifying potential side effects.

Challenges

The advances and transformations generated by AI are undeniable, but at the same time, new challenges and areas of attention can already be observed. The mapping by Anahp and ABSS also raised the challenges encountered by the interviewees, some of which include: the engagement of open clinical bodies is a barrier to transformation; health insurers do not always keep pace with hospitals’ technological evolution; there is the challenge of maintaining attention and care for patients outside the hospital, and data security and trust in the market are still low.

Recommendations

According to the research on quality, patient safety, and the importance of clinical decision support tools conducted in 2023 by Anahp in partnership with Wolters Kluwer, among the 74 responding hospitals, 47.39% consider telemedicine important in the patient care flow. Within this group, 31.08% believe it is a technology applicable in remote patient monitoring, and 43.24% see it as a means of conducting remote training for the clinical team. Thus, telemedicine can be used as a tool to improve clinical staff engagement, as well as maintain patient attention and care both inside and outside the hospital.

In order to enhance data security and market trust, it is essential to implement robust data protection policies and regulations, as well as promote transparency and accountability in the use of AI in healthcare. At the beginning of 2024, the World Health Organization (WHO) released the Guide for Multimodal AI Models for Health, which provides, among other guidelines, directions for the ethical use of AI in accordance with data protection laws. In Brazil, it is necessary for relevant bodies such as the Ministry of Health and the National Health Surveillance Agency (Anvisa) to dedicate themselves to creating regulations, possibly based on the released guide, since the country has been a WHO member since its foundation. In this way, AI can be fully exploited, ensuring user safety.

Giovanna Braga

Giovanna Braga, Healthcare and Advocacy consultant at LLYC Brasil