Interventional Oncology
Arian Mansur, BA (he/him/his)
Medical Student
Harvard Medical School
Disclosure(s): No financial relationships to disclose
Tarig Elhakim, MD
Research Associate
MGH
John C. Panagides, BS (he/him/his)
Medical Student
Harvard Medical School
Sanjeeva Kalva, MD, FSIR, FCIRSE, FACR (he/him/his)
Professor
Massachusetts General Hospital
Dania Daye, MD, PhD
Assistant Professor of Radiology
Massachusetts General Hospital/ Harvard Medical School
Several AI-based algorithms, including convolutional neural networks, have started to aid in not only the diagnosis of HCC, using several imaging modalities (abdominal ultrasound, CT with contrast, MRI, 18F-FDG PET/CT, histopathological classifications) but also the selection of patients for the IR-based treatment options, including transarterial chemoembolization (TACE), transarterial radioembolization (TARE) and tumor ablation. Several studies have also shown promising findings of AI in predicting the response to treatment with TACE/ablation as well as in local tumor progression and recurrence. This exhibit reviews and discusses some of the major findings to date.
Conclusion and/or Teaching Points: AI is starting to play a pivotal role in the way we manage patients with HCC, particularly in the context of interventional radiology. After viewing this exhibit, attendees will learn the various image-guided percutaneous and transarterial therapies that have proven to be highly promising for HCC as well as the role of AI in selecting suitable candidates for IR-based modalities, predicting response to treatment, and in follow-up care.