98%
921
2 minutes
20
Artificial intelligence (AI) is being explored for a growing range of applications in radiology, including image reconstruction, image segmentation, synthetic image generation, disease classification, worklist triage, and examination scheduling. However, training accurate AI models typically requires substantial amounts of expert-labeled data, which can be time-consuming and expensive to obtain. Active learning offers a potential strategy for mitigating the impacts of such labeling requirements. In contrast with other machine-learning approaches used for data-limited situations, active learning aims to produce labeled datasets by identifying the most informative or uncertain data for human annotation, thereby reducing labeling burden to improve model performance under constrained datasets. This Review explores the application of active learning to radiology AI, focusing on the role of active learning in reducing the resources needed to train radiology AI models while enhancing physician-AI interaction and collaboration. We discuss how active learning can be incorporated into radiology workflows to promote physician-in-the-loop AI systems, presenting key active learning concepts and use cases for radiology-based tasks, including through literature-based examples. Finally, we provide summary recommendations for the integration of active learning in radiology workflows while highlighting relevant opportunities, challenges, and future directions.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.2214/AJR.25.33364 | DOI Listing |
Nano Lett
September 2025
School of Materials and Chemistry, University of Shanghai for Science & Technology, Shanghai 200093, China.
Developing low-temperature gas sensors for parts per billion-level acetone detection in breath analysis remains challenging for non-invasive diabetes monitoring. We implement dual-defect engineering via one-pot synthesis of Al-doped WO nanorod arrays, establishing a W-O-Al catalytic mechanism. Al doping induces lattice strain to boost oxygen vacancy density by 31.
View Article and Find Full Text PDFBr J Clin Pharmacol
September 2025
University of South Carolina, School of Medicine Greenville, Greenville, SC, USA.
Aims: We implemented changes to a medical school curriculum aimed at boosting active learning and integrated instruction. Using the second level of Kirkpatrick's model, we describe the impact of the curricular revision on student performance in pharmacology assessments.
Methods: The analysis was divided into legacy (n = 105) and new (n = 110) curriculum students.
Biochem Pharmacol
September 2025
Department of Anesthesiology, Tongren Hospital, Shanghai Jiao Tong University School of Medicine, 1111 XianXia Road, Shanghai 200336, China; Hongqiao International Institute of Medicine, Tongren Hospital, Shanghai Jiao Tong University School of Medicine, 1111 XianXia Road, Shanghai 200336, China. El
Hypoxic-ischemic brain damage (HIBD) is a severe condition leading to extensive neuronal loss and functional impairments, representing a significant challenge in neonatal care. PFGA12, a peptide derived from fibrinogen alpha chain (FGA), which is notably downregulated in the umbilical cord blood of hypoxic-ischemic encephalopathy (HIE) infants. We demonstrate that PFGA12 significantly enhances cell viability and mitigates oxygen-glucose deprivation/reperfusion (OGD/R)-induced neuronal cell death.
View Article and Find Full Text PDFJ Vis Exp
August 2025
Chitkara University Institute of Engineering & Technology, Chitkara University.
Emotion annotation in code-mixed languages like Hinglish (Hindi-English) presents unique challenges due to linguistic complexity and resource constraints. This study introduces a hybrid active learning framework that combines lexical rules, machine learning, and iterative expert feedback to achieve cost-efficient, high-accuracy emotion annotation. Grounded in psychological theories of emotion, including Discrete Emotions Theory and Cognitive Appraisal Theory, the framework employs bilingual emotion dictionaries (e.
View Article and Find Full Text PDFCell Rep Methods
July 2025
Hubei Key Laboratory of Agricultural Bioinformatics, College of Informatics, Huazhong Agricultural University, Wuhan 430070, P.R. China; Key Laboratory of Smart Farming for Agricultural Animals, Ministry of Agriculture and Rural Affairs, Beijing, P.R. China; College of Informatics, Huazhong Agricult
We introduce a cell-free DNA (cfDNA) fragmentation pattern: the fragment dispersity index (FDI), which integrates information on the distribution of cfDNA fragment ends with the variation in fragment coverage, enabling precise characterization of chromatin accessibility in specific regions. The FDI shows a strong correlation with chromatin accessibility and gene expression, and regions with high FDI are enriched in active regulatory elements. Using whole-genome cfDNA data from five datasets, we developed and validated the FDI-oncology model, which demonstrates robust performance in early cancer diagnosis, subtyping, and prognosis.
View Article and Find Full Text PDF