RESUMEN
Background Artificial intelligence (AI) applications for cancer imaging conceptually begin with automated tumor detection, which can provide the foundation for downstream AI tasks. However, supervised training requires many image annotations, and performing dedicated post hoc image labeling is burdensome and costly. Purpose To investigate whether clinically generated image annotations can be data mined from the picture archiving and communication system (PACS), automatically curated, and used for semisupervised training of a brain MRI tumor detection model. Materials and Methods In this retrospective study, the cancer center PACS was mined for brain MRI scans acquired between January 2012 and December 2017 and included all annotated axial T1 postcontrast images. Line annotations were converted to boxes, excluding boxes shorter than 1 cm or longer than 7 cm. The resulting boxes were used for supervised training of object detection models using RetinaNet and Mask region-based convolutional neural network (R-CNN) architectures. The best-performing model trained from the mined data set was used to detect unannotated tumors on training images themselves (self-labeling), automatically correcting many of the missing labels. After self-labeling, new models were trained using this expanded data set. Models were scored for precision, recall, and F1 using a held-out test data set comprising 754 manually labeled images from 100 patients (403 intra-axial and 56 extra-axial enhancing tumors). Model F1 scores were compared using bootstrap resampling. Results The PACS query extracted 31 150 line annotations, yielding 11 880 boxes that met inclusion criteria. This mined data set was used to train models, yielding F1 scores of 0.886 for RetinaNet and 0.908 for Mask R-CNN. Self-labeling added 18 562 training boxes, improving model F1 scores to 0.935 (P < .001) and 0.954 (P < .001), respectively. Conclusion The application of semisupervised learning to mined image annotations significantly improved tumor detection performance, achieving an excellent F1 score of 0.954. This development pipeline can be extended for other imaging modalities, repurposing unused data silos to potentially enable automated tumor detection across radiologic modalities. © RSNA, 2022 Online supplemental material is available for this article.
Asunto(s)
Inteligencia Artificial , Redes Neurales de la Computación , Encéfalo , Humanos , Imagen por Resonancia Magnética , Estudios RetrospectivosRESUMEN
Cancer-associated venous thromboembolism (VTE) is a major source of oncologic cost, morbidity and mortality. Identifying high-risk patients for prophylactic anticoagulation is challenging and adds to clinician burden. Circulating tumor DNA (ctDNA) sequencing assays ('liquid biopsies') are widely implemented, but their utility for VTE prognostication is unknown. Here we analyzed three plasma sequencing cohorts: a pan-cancer discovery cohort of 4,141 patients with non-small cell lung cancer (NSCLC) or breast, pancreatic and other cancers; a prospective validation cohort consisting of 1,426 patients with the same cancer types; and an international generalizability cohort of 463 patients with advanced NSCLC. ctDNA detection was associated with VTE independent of clinical and radiographic features. A machine learning model trained on liquid biopsy data outperformed previous risk scores (discovery, validation and generalizability c-indices 0.74, 0.73 and 0.67, respectively, versus 0.57, 0.61 and 0.54 for the Khorana score). In real-world data, anticoagulation was associated with lower VTE rates if ctDNA was detected (n = 2,522, adjusted hazard ratio (HR) = 0.50, 95% confidence interval (CI): 0.30-0.81); ctDNA- patients (n = 1,619) did not benefit from anticoagulation (adjusted HR = 0.89, 95% CI: 0.40-2.0). These results provide preliminary evidence that liquid biopsies may improve VTE risk stratification in addition to clinical parameters. Interventional, randomized prospective studies are needed to confirm the clinical utility of liquid biopsies for guiding anticoagulation in patients with cancer.