Molecular medical imaging, with technologies like PET and SPECT, plays a crucial role in oncology for diagnosis and treatment tracking. Yet, interpreting these images, especially in gauging immunotherapy responses, presents challenges due to new patterns of response and progression as well as inherent subjectivity in image interpretation.
While there’s a clear drive to harness artificial intelligence (AI) for improved clinical care, it’s vital to ensure transparency, avoiding the ‘black box’ scenario. Hence, our study delves into the current landscape of AI research in PET and SPECT imaging, seeking areas for refinement and clarity.
Our findings underscore the potential of Radiomics and AI in quantifying image biomarkers and streamlining algorithm development. However, the current surge in studies lacks robust validation, hampering their applicability across different scenarios.
Examining existing literature, we note a predominant focus on lung cancer predictions using PET, sidelining the untapped potential of SPECT and PET in other tumor types. Many studies introduce novel models without thorough validation, indicating a nascent field that demands rigorous scrutiny.
We envision our study as shedding light on areas for improvement for our medical community in both methodology and content, while charting a course for future investigation, development, and clinical integration of AI in radiology.
Key points:
- Immunotherapy has drastically altered the cancer treatment landscape but is known to precipitate response patterns that are not accurately accounted for by traditional imaging methods.
- There is an unmet need for better tools to not only facilitate in-treatment evaluation but also to predict, a priori, which patients are likely to achieve a good response with a certain treatment as well as those who are likely to develop side effects.
- Artificial intelligence applied to PET/SPECT imaging of immunotherapy-treated patients is mainly used to make predictions about prognosis or tumor phenotype and is built from baseline, pre-treatment images. Further testing is required before a true transition to clinical application can be realized.
Article: Artificial intelligence in immunotherapy PET/SPECT imaging
Authors: Jeremy P. McGale, Delphine L. Chen, Stefano Trebeschi, Michael D. Farwell, Anna M. Wu, Cathy S. Cutler, Lawrence H. Schwartz & Laurent Dercle
