An artificial intelligence tool outperforms human readers in diagnostic mammography but shows lower accuracy at the lesion level than at the breast level, a study shows.
showed a high sensitivity and specificity, outperforming human readers in overall performance, a study found. However, the performance of the AI tool was slightly lower at the lesion level than at the breast level.
Researchers retrospectively analysed mammograms from the UK's NHS Breast Screening Programme by using a commercial AI tool and human readers to evaluate 882 non-malignant and 318 malignant breasts with 328 lesions. Human readers , including radiologists, radiographers, and breast clinicians, reviewed 1200 mammograms. The same cases were independently reviewed by the AI tool. Human and AI decisions to clear or recall breasts or lesions were compared with real outcomes on the basis of pathology or a 3-year follow-up. Sensitivity, specificity, and area under the curve were calculated at both breast and lesion levels.The AI tool outperformed human readers on the basis of the AUC at the breast level and lesion level (0.929 vs 0.851;At the developer-recommended recall threshold, the AI tool achieved a significantly higher specificity than human readers at the breast level (87.4% vs 79.2%;When calibrated to match the human specificity, the AI tool had a higher sensitivity than human readers at the breast level (92.1% vs 87.5%;"Our findings support the notion of implementing AI into a prospective screening workflow, where the localisation of malignancies is beneficial to patients and the screening process," the authors wrote. "To improve human-AI collaboration, AI should be assessed at the lesion level; poor accuracy here may lead to automation bias and unnecessary patient procedures," they added.This study was led by Adnan Gani Taib and George John William Partridge, University of Nottingham, Nottingham, England. It wasThis study could not assess the real-time effect of AI on human decision-making due to its retrospective design. The use of cancer-enriched test sets may have led to an overestimation of human performance. Additionally, prior mammograms were not available for comparison, which are often used in routine clinical practice to aid detection.This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our
Deep Learning AI NPL Machine Learning ML Natural Language Processing Artificial Neural Networks Computer-Aided Detection For Mammography Breast Cancer Malignant Breast Neoplasm Breast Carcinoma Breast Mammary Gland Mammography Mammogram UK National Health Service United Kingdom National Health Service UK NHS NHS National Health Service Cancer Malignant Neoplasia Carcinoma Malignant Neoplasm Healthcare And Medical Technology Health And Medical Tech Health And Med Tech Health And Medical Technology Healthcare Technology Medical Technology Medical Imaging | Diagnostic Medical Imaging
United States Latest News, United States Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Artificial Intelligence For Business: Another Good Intro To The IssuesArtificial Intelligence for Business, Kamales Lardi, is another good introductory book to the subject.
Read more »
Libraries open their stacks up as training data for artificial intelligence platformsEverything ever said on the internet was just the start of teaching artificial intelligence about humanity. Data-hungry tech companies are now tapping into an older repository of knowledge: the library stacks.
Read more »
Artificial intelligence is very much like human intelligence—it makes mistakes.Whether AI matches or surpasses human intelligence is an odd question. When AI systems finally mimic human performance, we blame them for humanlike cognitive fallacies.
Read more »
Could Artificial Intelligence solve construction worker accidents?Position: Multimedia Journalist
Read more »
Why Artificial Integrity Must Overtake Artificial IntelligenceEvidence of AI models’ integrity lapses is not anecdotal or speculative. So-called intelligence alone is no longer the benchmark. Integrity is.
Read more »
Artificial Intelligence Model Shows High Accuracy in Early Detection of Pancreatic Cancer on CT ScansA pretrained artificial intelligence model demonstrates a high sensitivity in detecting pancreatic cancer on CT scans, a new study finds.
Read more »