Start Your Search
PL01 - New Questions with Imaginative Answers (ID 88)
- Event: WCLC 2019
- Type: Plenary Session
- Track: Advanced NSCLC
- Presentations: 1
- Now Available
- Moderators:Enriqueta Felip, Giorgio Vittorio Scagliotti
- Coordinates: 9/08/2019, 08:15 - 09:45, Barcelona (2005)
PL01.04 - Artificial Intelligence, Big Data and Lung Cancer: Ready to Implement? (Now Available) (ID 3583)
08:15 - 09:45 | Presenting Author(s): Hugo Aerts
A critical barrier present in cancer research and treatment today is when and how to act based on the information provided from tumor data. One important reason for the slow progress in the fight against cancer, is the fact that cancer is a “moving target”. It is constantly evolving and diversifying, changing its phenotype, its genomic composition, and through metastatic spread, even its location. This is even more true when subjected to the pressure of therapeutic intervention, where cancer evolution rapidly explores and exploits resistance mechanisms, potentially even aided by the mutagenic nature of cancer treatments, leaving the treating oncologist chasing a constantly changing disease.
Artificial Intelligence (AI) and Deep Learning technologies have recently led to revolutionary advances in areas ranging from computer vision to speech recognition - tasks that up to a few years ago could only be done by humans. AI has the potential to fundamentally alter the way medicine is practiced, as it excels in recognizing complex patterns in medical data and provides a quantitative, rather than qualitative, assessment of clinical conditions. AI-powered radiographic-biomarkers (“radiomics”) may quantify non-invasive information of the cancer phenotype that is clinically actionable, and may further improve diagnosis, characterization, and longitudinal tracking through therapy. AI methods are precise and allow specific quantification of features not otherwise quantifiable by human experts. Radiomic-analysis is performed on the entire tumor as compared to just a small sample for molecular analysis and provides a non-invasive window into internal growth pattern of the tumor (including internal textural heterogeneity, macroscopic necrosis, and viable tumor mass). Radiomics can thus quantify the phenotypic state of a tumor within its evolutionary process, thereby sidestepping issues relating to biopsies.
This is particularly important for patients with cancer, where different cancer lesions can express different microenvironments that could ultimately lead to heterogeneous response patterns. Despite the remarkable success of novel cancer therapies, the clinical benefit remains limited to a subset. Cancer therapies are often expensive and could bring unnecessary toxicity, there is a direct need to identify beneficial patients, but this remains difficult in the clinic today. Radiomics biomarkers could provide this information on a lesion and patient level using standard-of-care CT scans. Unlike biopsy assays that - by definition - only represent a sample within the tumor, imaging can depict a full picture of the entire tumor burden, providing information of each cancer lesion within a single non-invasive examination.
Another field that will be impacted by AI and big data is radiation oncology. Radiation oncology as a therapeutic specialty presents itself as an exemplary field that will be impacted by AI automation. Especially as much of the current radiation therapy work flow requires time-consuming, manual labor by both radiation oncologists and a team of medical staff including medical physicists, certified medical dosimetrists, and radiation therapists. The growing complexity of the human-machine and human-software interactions in conjunction with the increasing incidences of cancer have created a workforce shortage throughout the world. In fact, variations in the radiation treatment planning process can lead to significant differences in the quality of care, and negatively impact overall survival even in clinical settings where extra care is given to standardizing segmentation and planning approaches. Furthermore, the knowledge and experience gap between more developed and under-resourced health care environments poses an enormous public health challenge and represents one of the great global inequities in cancer care.
In this talk, Dr. Aerts will discuss recent developments from his group and collaborators performing research at the intersection of artificial intelligence big data, and oncology. Also, he will discuss recent work of building a computational image analysis system to extract deep learning algorithms and use these to build radiomic signatures. The presentation will conclude with a discussion of future work on building integrative systems incorporating both molecular and phenotypic data to improve cancer therapies.
Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.