The field of medical imaging and diagnostics is rapidly evolving, with a focus on developing innovative AI-powered solutions to improve diagnosis accuracy and patient outcomes. Recent developments have centered around creating more accurate and efficient systems for detecting and diagnosing various diseases, including lung cancer, thyroid cancer, and mesothelioma. The use of deep learning algorithms and large datasets has enabled researchers to create models that can outperform radiologists in certain tasks, such as nodule detection and diagnosis. Additionally, there is a growing interest in developing models that can work with limited or incomplete data, such as sparse-view CBCT reconstruction and incomplete multi-modal tumor segmentation. Noteworthy papers in this area include the AI system for lung cancer screening that outperforms radiologists and leading AI models, and the TARDis framework for incomplete multi-modal tumor segmentation and classification that achieves state-of-the-art results on large-scale datasets.
Advances in Medical Imaging and Diagnostics
Sources
Rethinking Lung Cancer Screening: AI Nodule Detection and Diagnosis Outperforms Radiologists, Leading Models, and Standards Beyond Size and Growth
Doppler-Enhanced Deep Learning: Improving Thyroid Nodule Segmentation with YOLOv5 Instance Segmentation
Neural Discrete Representation Learning for Sparse-View CBCT Reconstruction: From Algorithm Design to Prospective Multicenter Clinical Evaluation
MasHeNe: A Benchmark for Head and Neck CT Mass Segmentation using Window-Enhanced Mamba with Frequency-Domain Integration
Toward Content-based Indexing and Retrieval of Head and Neck CT with Abscess Segmentation
Cross-Domain Validation of a Resection-Trained Self-Supervised Model on Multicentre Mesothelioma Biopsies
TARDis: Time Attenuated Representation Disentanglement for Incomplete Multi-Modal Tumor Segmentation and Classification