With the rise of artificial intelligence in medicine, this technology is becoming a valuable ally to human expertise. Diagnosis, research, surgical intervention, AI is everywhere. To the point of redefining, or even eventually replacing, the role of doctors?
By Philippe Emy.
In 1950, medical knowledge doubled every 50 years. In 2010, this duration was reduced to 3 and a half years, while in 2020, this cycle fell to 73 days. The explosion of medical knowledge in recent years is linked to the development of artificial intelligence (AI), which today marks a decisive advance in the field of health. As evidenced by the recent awarding of the Nobel Prize in Chemistry to Demis Hassabis and John Jumper (directors of DeepMind, Google’s AI subsidiary), distinguished for the development of AlphaFold2, an AI model capable of predicting the structure of proteins , which could accelerate research in biotechnology and pharmacology.
From prevention to surgical assistance
The promises of new technologies using artificial intelligence in the field of health are multiple: prevention-prognosis, diagnosis, treatment, research, etc. The applications of AI in health are already visible on a daily basis. For example, connected watches can now detect heart rhythm disorders and trigger alerts if there is a risk of stroke. Thanks to these devices, it will be possible to establish disease prediction factors for each individual based on individual characteristics (lifestyle, genetics, environment).
Diagnostic assistance is also developing rapidly: new high-resolution medical imaging machines make it possible to detect small tumors invisible to the human eye more reliably early and to better target treatments. A startup has even developed an algorithm capable of detecting diabetic retinopathy early. Clinical decision support systems provide rapid and accurate interpretation of results such as the interpretation of an electrocardiogram. Remote consultations are developing: a simple photo of a skin lesion is sent to the machine which will analyze it and, if necessary, trigger an appointment with a dermatologist, very useful in medical deserts.
AI has also, in recent years, entered operating theaters: surgical interventions using robots are becoming widespread, improving the comfort of the surgeon and the patient and simplifying post-operative procedures: the surgeon operates in front of a 3D screen allowing better vision. It controls articulated arms which carry out the operation with great precision. Thanks to teletransmission, the intervention can also be carried out remotely. AI advances medical research by making it possible to discover new therapeutic targets, develop new molecules, and validate scientific hypotheses.
Reliable and safe technology?
However, the integration of AI in medicine raises ethical and practical questions. If the decision-making of the algorithms is autonomous, all deviations are possible, as illustrated by a recent controversial study which aimed to detect the homosexuality of people on the basis of photos of faces. Human supervision is essential, but will it always be possible? Just like the issue around confidentiality and security of collected data, not immune to the growing phenomenon of cyberattacks. The lack of regulation of AI can lead to legal, ethical and professional risks in the event of misuse or malfunction of the technology.
If AI is at the heart of the medicine of the future, it will not replace the doctor-patient relationship, nor the human factor in the success of a treatment. It must be seen as effective assistance for human skills and not as a substitute.
More information otherwise on Magcentre:
New anti-obesity drugs: miracle treatments?
