Innovation In Focus will be one of the main themes at ECR 2024. We spoke with Martina Pecoraro, co-coordinator of the program, about the challenges of integrating AI and digital tools, and how innovation can help radiology become greener.
What can you tell us about Innovation in Focus?
The program has been developed in conjunction with my colleague Prof. Daniel Pinto Dos Santos, to highlight the opportunities offered by new technologies, and how these will impact radiology and healthcare.
Aligning with the congress theme ‘Next Generation Radiology’, we tried to define the program selecting topics not strictly related to imaging per se, but that somehow already influence or will influence radiology practice in the near future.
The program has been thought out to address specific points:
(i) the large availability of multimodal and multidisciplinary data originated from different sources and collectively defining integrated diagnostics, and the role of imaging biomarkers within the Digital Twin approach;
(ii) the need for a strict regulatory control of these technologies, especially when considering patients;
(iii) the expected impact of these technologies in scientific research and patients’ health – how they will transform the delivery of state-of-the-art multidisciplinary care and the management of clinical studies;
(iv) the need to improve sustainability in radiology;
(v) Large language models (LLMs), chatbots such as ChatGPT, etc., to provide a clear understanding of the pros and cons of using such tools.
The program will feature seven sessions in Room M2. Lectures will be held by junior and senior radiologists experts in the field, as well as radiographers, patient representatives, epidemiologists and ESR staff.
In what ways do you see AI enhancing diagnostic processes?
AI has already transformed medical imaging and it is now improving accuracy and mostly efficiency in the diagnostic process. This is done in several ways addressing every step of the imaging process, from image acquisition to reporting, but also on a multi-level scale. For instance, concerning image acquisition, AI reduces the length of the exam with algorithms enabling faster acquisition, and improves image quality with image reconstruction models.
The pool of use cases is very large and what we perceive today is only the “tip of the iceberg”. I’m thriving to see what the future holds. I strongly believe that digital twin technologies, which are a domain-adapted multimodal AI-based modeling approach, will enhance diagnostic processes, improving also disease prognosis and care.
As radiology becomes more digitized, how are concerns related to data security and patient privacy being addressed?
Data security and patients’ privacy might be one of the most challenging barriers to overcome for new technologies to really be taken up in clinical work. Privacy regulations, and social and ethical hazards are significant drawbacks that affect acceptability in healthcare.
However, as new technologies are developed, cloud computing is absolutely necessary to enable secure storage and retrieval of vast amounts of data online, and the synergy of AI models’ applications and cloud technology opens new possibilities to advance research. Federated learning is a potential solution for developing machine-learning models to avoid data centralization and allow parties to have better control over the processing of their personal data. Federated learning could also help reduce the amount of personal data transferred and processed by third parties for model training, and facilitate models training with data coming from different jurisdictions.
This important issue will be tackled during the session “The collection of high-quality data to benefit the patient, the radiologist, and at a multidisciplinary level” on February 29.
Sustainability is also a key topic in the programme. How can we make radiology a greener profession?
This topic needs to be addressed in the community. The world is facing climate change, and we as radiologists have to take responsibility and realize the implications of going always harder, better, faster and stronger on our planet.
People have been saying “going back to the future’’, for example when referring to the use of low-field MRI and its promises regarding high quality imaging. My thoughts are to move forward to the future, by looking at what new technologies can offer to reduce our environmental impact. Perhaps low-field MRI can be a solution in specific cases, but so can contrast media recycling and waste management strategies.
These topics will be addressed on February 29, and I strongly encourage everyone to keep up with this discussion.
Why should ECR visitors attend Innovation in Focus sessions?
Radiology is a multimodal discipline that makes extensive use of different imaging methods, all of which produce a huge amount of longitudinal digital quantitative data at different time frames. Radiologists are thus at the center of the “technological revolution” of healthcare, and need tools and knowledge to be able to navigate among many buzz words, such as AI, big data, machine learning and robotics. By visiting the In Focus sessions, radiologists will improve their awareness of new technologies to harness their power, as they might shift the face of our discipline very soon.
Martina Pecoraro trained at the Sapienza University of Rome, where she graduated in 2022. She is currently a PhD student in the Network Oncology and Precision Medicine doctoral programme at the same university, with GU imaging as her main field of research. She is a member of the Research Committee and of the Radiology Trainees Forum (RTF) of the European Society of Radiology (ESR), among many other roles.
