Objective: To evaluate whether artificial intelligence (AI) for detecting breast cancer on mammography can improve the performance and time efficiency of radiologists reading mammograms.
Materials And Methods: A commercial deep learning-based software for mammography was validated using external data collected from 200 patients, 100 each with and without breast cancer (40 with benign lesions and 60 without lesions) from one hospital. Ten readers, including five breast specialist radiologists (BSRs) and five general radiologists (GRs), assessed all mammography images using a seven-point scale to rate the likelihood of malignancy in two sessions, with and without the aid of the AI-based software, and the reading time was automatically recorded using a web-based reporting system. Two reading sessions were conducted with a two-month washout period in between. Differences in the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and reading time between reading with and without AI were analyzed, accounting for data clustering by readers when indicated.
Results: The AUROC of the AI alone, BSR (average across five readers), and GR (average across five readers) groups was 0.915 (95% confidence interval, 0.876-0.954), 0.813 (0.756-0.870), and 0.684 (0.616-0.752), respectively. With AI assistance, the AUROC significantly increased to 0.884 (0.840-0.928) and 0.833 (0.779-0.887) in the BSR and GR groups, respectively ( = 0.007 and < 0.001, respectively). Sensitivity was improved by AI assistance in both groups (74.6% vs. 88.6% in BSR, < 0.001; 52.1% vs. 79.4% in GR, < 0.001), but the specificity did not differ significantly (66.6% vs. 66.4% in BSR, = 0.238; 70.8% vs. 70.0% in GR, = 0.689). The average reading time pooled across readers was significantly decreased by AI assistance for BSRs (82.73 vs. 73.04 seconds, < 0.001) but increased in GRs (35.44 vs. 42.52 seconds, < 0.001).
Conclusion: AI-based software improved the performance of radiologists regardless of their experience and affected the reading time.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9081685 | PMC |
| http://dx.doi.org/10.3348/kjr.2021.0476 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Dynamic Video Assessment of Axial Postural Abnormalities in Parkinson's Disease: A Pilot Study.
Mov Disord Clin Pract
January 2025
Department of Computer Science, University of Verona, Verona, Italy.
Background: Axial postural abnormalities (APAs) are frequent and disabling axial symptoms of Parkinson's disease (PD). Image-based measurement is considered the gold standard but may not accurately detect the true severity of APAs because these symptoms can appear or get worse under dynamic conditions.
Objective: The aim was to evaluate quantitative changes in APAs degree during prolonged standing and walking in both single- and dual-task conditions (motor + cognitive).
Neural Correlates of Narrative Reading Development: A Comparative fMRI Study of Adults and Children Using Time-Locked Inter-Subject Correlation Analyses.
Psychophysiology
January 2025
Beijing Key Lab of Learning and Cognition, School of Psychology, Capital Normal University, Beijing, China.
The naturalistic paradigm and analytical methods present new approaches that are particularly suitable for research concentrating on narrative reading development. We analyzed fMRI data from 44 adults and 42 children engaged in story reading using time-locked inter-subject correlation (ISC), inter-subject representation similarity analysis (IS-RSA), and inter-subject functional correlation (ISFC). The ISC results indicated that for both children and adults, narrative reading recruited not only traditional reading areas but also regions that are sensitive to long-time-scale information, such as the medial prefrontal cortex and hippocampus, which increased involvement from children to adults.
View Article and Find Full Text PDFThe work of the human mind: Time and symbolization (in honour of Dana's work and thought).
Int J Psychoanal
December 2024
SBPSP (Brazilian Society of Psychoanalysis, São Paulo), São Paulo, Brazil.
In order to honour the work of Dana Birksted-Breen I will attempt to show how the reading of her work has impacted on my own thinking and changed it or modified it. I would like to begin by highlighting one of Dana's claims in . I will focus on two areas: (1) the development of symbolic capacity and its connection with temporality, and (2) the concept of reverie as a .
View Article and Find Full Text PDFEye Movement and Recall of Visual Elements in Eco-friendly Product.
J Eye Mov Res
December 2024
Department of Marine Design Convergence Engineering, Pukyong National University, Korea.
This study aims to explore the distribution of visual attention on sustainability graphics when viewing an eco-friendly product and the recall of sustainability information afterward. Twenty-five students majoring in environmental studies and twenty-five students from non-environmental majors participated in the study. They were further divided into a higher group and a lower group based on their sustainability level.
View Article and Find Full Text PDFPreferences of nurses in the United Kingdom for attributes of pediatric hexavalent vaccines: a discrete-choice experiment.
Expert Rev Pharmacoecon Outcomes Res
January 2025
Merck & Co. Inc, Rahway, NJ, USA.
Background: We evaluated UK nurses' preferences for pediatric hexavalent vaccine attributes.
Research Design And Methods: In a discrete-choice experiment study, 150 nurses chose between 2 hypothetical pediatric hexavalent vaccines with varying attribute levels (device type, plastic in packaging, time on the market, and time the vaccine can stay safely at room temperature) in a series of choice questions. Using random-parameters logit-model estimates, conditional relative attribute importance (CRAI) and odds ratios (ORs) were calculated.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!