Will Artificial Intelligence revolutionize medicine?
Artificial intelligence: the new pillar of modern medicine?
According to Eric Topol, author of the book “Deep Medicine”, the use of artificial intelligence in medicine is inevitable and necessary. AI can solve the many challenges : misdiagnosis, high cost of care, treatment errors, waste, overworked and overwhelmed caregivers. In fact, AI is already being used throughout the healthcare chain: from diagnosis to treatment, from MRI to the creation of new drugs.
Impact of AI on healthcare professions
In the coming years, AI is expected to impact many healthcare professions: from radiologists to pharmacists to nurses. For Eric Topol, AI should not replace these professions but help them to be more efficient.
Impact of AI on Medical Imaging
AI also improves the quality and accuracy of examinations such as the resolution of cardiac MRI images. Pathologies are detected faster and with greater accuracy. The impact on the management of pathologies is immediate.
Increasing the quality of cardiac MRI:
4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics-May 2020
Recent advances in super-resolution imaging using deep learning have shown that it is possible to improve image resolution, fill in missing details, and recover information.
Real-time monitoring of patient’s health
The use of AI combined with connected objects will allow for real-time monitoring and analysis of patients’ health. This is the goal of this toilet bowl equipped with sensors that should allow real-time analysis of data extracted from urine and bodily fluids. The “toilet bowl” print is better than the connected watch to follow the evolution of its health.
Replacing doctors with the toilet bowl :
A mountable toilet system for personalized health monitoring via the analysis of excreta-April 2020
Each toilet user is identified by his or her fingerprint and the distinctive characteristics of his or her anoderm. The data is securely stored, encrypted and analyzed in a server hosted on the cloud. The toilets can be used for screening, diagnosis and longitudinal monitoring of specific patient populations.
A virtual medical assistant
For Eric Topol, AI will allow the implementation of a virtual medical assistant: a sort of giant algorithm capable of taking into account variables as diverse as the genome, the microbiome, the family history, the physical activity, the sensors of the toilet bowl, etc… An augmented general practitioner consultable at any time and without waiting room.
The fields of application of artificial intelligence in healthcare are numerous but still limited. The lack of annotated data, the lack of collaboration between AI experts and health actors will be the main obstacles to overcome in order to put AI at the service of caregivers and patients.
Healing through Artificial Intelligence
Article written from this video : Pie & AI: Real-world AI Applications in Medicine. (4'21 to 20'16 : Eric Topol)