Back to the list
Congress: ECR25
Poster Number: C-17756
Type: Poster: EPOS Radiologist (scientific)
Authorblock: B. Dalkıran, G. Durhan, H. Avci, N. Kertmen, F. Demirkazik; Ankara/TR
Disclosures:
Burak Dalkıran: Nothing to disclose
Gamze Durhan: Nothing to disclose
Hanife Avci: Nothing to disclose
Neyran Kertmen: Nothing to disclose
Figen Demirkazik: Nothing to disclose
Keywords: Artificial Intelligence, Computer applications, Oncology, CT, CT-Quantitative, CAD, Computer Applications-Detection, diagnosis, Cancer
Purpose Metastatic pulmonary nodules of undefined primary origin can be incidentally detected on chest computed tomography (CT). Radiologists evaluate metastatic nodules based on a limited set of imaging characteristics, such as size, shape, and contrast enhancement. These characteristics of the nodules are mostly insufficient to determine the primary tumor. Identifying the primary tumor requires comprehensive diagnostic evaluations, including laboratory tests, radiological imaging, biopsies, and endoscopic procedures. These assessments are costly, time-consuming, and associated with risks due to invasive interventions. Early detection...
Read more Methods and materials Patients diagnosed with metastatic breast cancer, colorectal cancer, and renal cell carcinoma between 2015 and 2023 at two hospitals affiliated with a single university were retrospectively reviewed. The exclusion criteria included metastatic lesions that developed during treatment, CT scans with motion artifact, and non-contrast CT scans. While patients from one hospital constituted the training cohort, those from the other served as the external validation cohort. The training cohort included 50 breast cancer, 55 colorectal cancer, and 55 renal cell carcinoma...
Read more Results The macro-averaged and micro-averaged AUC values in the validation set were 0.694/0.623 for the clinical-radiological model, 0.616/0.616 for the radiomics model, and 0.714/0.712 for the combined model in triple classification (Figure 2).
Read more Conclusion The radiomics model alone did not demonstrate sufficient performance in distinguishing lung metastases. However, the combined model outperformed both the radiomics and clinical-radiological models, highlighting its potential as a promising tool for differentiating lung metastases.
Read more References Taylor MB, Bromham NR, Arnold SE. Carcinoma of unknown primary: key radiological issues from the recent National Institute for Health and Clinical Excellence guidelines. Br J Radiol. 2012;85(1014):661-71. Stella GM, Kolling S, Benvenuti S, Bortolotto C. Lung-Seeking Metastases. Cancers (Basel). 2019;11(7):1010. Gillies RJ, Kinahan PE, Hricak H. Radiomics: Images Are More than Pictures, They Are Data. Radiology. 2016;278(2):563-77.
Read more
GALLERY