Habitat-Radiomics combining multichannel 2.5D deep learning for differentiating adrenal adenomas from metastases using automatic segmentation: a multicenter study - Scorecard - MDSpire
Advertisement
Habitat-Radiomics combining multichannel 2.5D deep learning for differentiating adrenal adenomas from metastases using automatic segmentation: a multicenter study
Clinical Scorecard: Integrating Habitat-Radiomics with Multichannel 2.5D Deep Learning for the Automated Segmentation and Differentiation of Adrenal Adenomas from Metastatic Lesions: A Multicenter Investigation
At a Glance
Category
Detail
Condition
Key Mechanisms
Habitat-radiomics and 2.5D deep learning for automated segmentation and differentiation
Target Population
Care Setting
Key Highlights
390 patients involved in the study
Fusion model achieved AUCs of 0.983, 0.913, and 0.886
Standalone models showed AUCs ranging from 0.830–0.945 and 0.825–0.970
Utilization of DenseNet-121 architecture for deep learning
Guideline-Based Recommendations
Diagnosis
Automated segmentation using Medical SAM
Utilization of habitat-radiomics and conventional radiomics features
Management
Monitoring & Follow-up
Assessment of model performance through ROC curve analysis
Risks
Challenges in qualitative diagnosis of lipid-poor adrenal adenomas and metastases
