Many studies have shown that using a computer-aided detection (CAD) system does not significantly improve diagnostic accuracy in radiology, possibly because radiologists fail to interpret the CAD results properly. We tested this possibility using screening mammography as an illustrative example. We carried out two experiments, one using 28 practicing radiologists, and a second one using 25 non-professional subjects. During each trial, subjects were shown the following four pieces of information necessary for evaluating the actual probability of cancer in a given unseen mammogram: the binary decision of the CAD system as to whether the mammogram was positive for cancer, the true-positive and false-positive rates of the system, and the prevalence of breast cancer in the relevant patient population. Based only on this information, the subjects had to estimate the probability that the unseen mammogram in question was positive for cancer. Additionally, the non-professional subjects also had to decide, based on the same information, whether to recall the patients for additional testing. Both groups of subjects similarly (and significantly) overestimated the cancer probability regardless of the categorical CAD decision, suggesting that this effect is not peculiar to either group. The misestimations were not fully attributable to causes well-known in other contexts, such as base rate neglect or inverse fallacy. Non-professional subjects tended to recall the patients at high rates, even when the actual probably of cancer was at or near zero. Moreover, the recall rates closely reflected the subjects’ estimations of cancer probability. Together, our results show that subjects interpret CAD system output poorly when only the probabilistic information about the underlying decision parameters is available to them. Our results also highlight the need for making the output of CAD systems more readily interpretable, and for providing training and assistance to radiologists in evaluating the output.
Background Mixed data exist regarding the association between hyperglycemia and functional outcome after acute ischemic stroke when accounting for the impact of leptomeningeal collateral flow. We sought to determine whether collateral status modifies the association between treatment group and functional outcome in a subset of patients with large vessel occlusion enrolled in the Stroke Hyperglycemia Insulin Network Effort (SHINE) trial. Methods In this post-hoc analysis, we analyzed patients enrolled into the SHINE trial with anterior circulation large vessel occlusion who underwent imaging with CT angiography prior to glucose control treatment group assignment. The primary analysis assessed the degree to which collateral status modified the effect between treatment group and functional outcome as defined by the 90-day modified Rankin Scale score. Logistic regression was used to model the data, with adjustments made for thrombectomy status, age, post-perfusion thrombolysis in cerebral infarction (TICI) score, tissue plasminogen activator (tPA) use, and baseline National Institutes of Health Stroke Scale (NIHSS) score. Five SHINE trial centers contributed data for this analysis. Statistical significance was defined as a p-value < 0.05. Results Among the 1151 patients in the SHINE trial, 57 with angiographic data were included in this sub-analysis, of whom 19 had poor collaterals and 38 had good collaterals. While collateral status had no effect (p = 0.855) on the association between glucose control treatment group and functional outcome, patients with good collaterals were more likely to have a favorable functional outcome (p = 0.001, OR 5.02; 95% CI 1.37–16.0). Conclusions In a post-hoc analysis using a subset of patients with angiographic data enrolled in the SHINE trial, collateral status did not modify the association between glucose control treatment group and functional outcome. However, consistent with prior studies, there was a significant association between good collateral status and favorable outcome in patients with large vessel occlusion stroke. Trial registration ClinicalTrials.gov Identifier is NCT01369069. Registration date is June 8, 2011.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.