Digital in the hospital is making great strides. Some algorithms are already able to diagnose lung cancer or analyze a mammogram better than a doctor. This technology promises huge advancements in the future, provided certain challenges are met. Several players and analysts from the field gathered during the round table”Digital, artificial intelligence, doctors and patientsof the Global Health conference in Strasbourg, analyze the future challenges that the world of technology and the world of health will have to face together.
Data, yes, but quality
“To have a good artificial intelligence, it must work with a good algorithm. And a good algorithm is built with good dataexplains Bernard Nordlinger, a member of the National Academy of Medicine and a specialist in artificial intelligence in health. Health data, collected by hospitals and health structures, must be calibrated, relevant and “cleansed” of any parasitic details that would distort their interpretation.”We remember the artificial intelligence Watson, created by IBM, which had disappointed many, especially since the health data used was not ideal.”
Scientists working with this program explained that they had great difficulty understanding patients’ medical records: acronyms to be detailed, errors to correct, abbreviated sentences. Each piece of information must first be formatted in order to be properly analyzed by the system. “The problem was the same during the Covid-19 crisis. Artificial intelligence could not find its place because the data generated at the time was not of high quality“, regrets the specialist. On the other hand, if the data are accurate and well-formulated, they can even get their place in a clinical trial. Today there are control groups, that is, the arm of a clinical trial that does not take treatment, composed only of synthetic data performed with artificial intelligence.A promising head start.
After their development and before their use in health structures, the algorithms are subjected to rigorous evaluation in clinical trials, such as during the development of a drug. “Validating these algorithms requires international collaborationsaid Irene Buvat, director of the Translational Imaging in Oncology Laboratory at the CNRS and director of research at the Institut Curie.We carry out many projects with European partners to build databases on which the algorithms will be developed. But the validation step with other countries is also crucialThe specialist explains that the different protocols and criteria in force in each country provide an opportunity to perfect artificial intelligence so that it can be used in hospitals internationally.
Some algorithms are built on health data collected in hospitals, at the patient’s bedside. “It is therefore necessary to be able to collect data in the hospital, but sometimes also obtained by remotely connected objects“explains Jean Sibilia, PU-PH in rheumatology at the University Hospitals of Strasbourg and the Faculty of Medicine of Strasbourg.”Doctors also need to be up to date to ensure ethics to patients and reassure them. And finally, they must be able to analyze this dataObjectives that can be included during initial training, as well as later, as part of ongoing training.Ethics and data protection are topics that should be discussed with both experienced doctors and the young generation that has grown up in the digital world.“, resumes Jean Sibilia.
Cyber security must be there
Artificial intelligence means generating health data within healthcare centers. But these are now under threat. In recent years, the number of ransomware targeting hospitals has multiplied. These computer attacks paralyze healthcare structures and threaten the privacy of patients, whose “data” can be resold on the darknet.
“We don’t wonder if it’s going to happen, but when”, confides Antoine Geissbühler, chief physician of the cyber health and telemedicine service of Geneva University Hospitals.Faced with vulnerable systems, we must reduce the attack surface available to cybercriminals. We have ethical hackers on our side and intelligent monitoring systems that can detect suspicious behavior before a massive attack. A bit like banks that can check when credit card usage is abnormalComputer updates to stay up-to-date and networking between hospitals are also among the essential tools to know how to mitigate the risk of computer attacks.