CAD in mammography: lesion-level versus case-level analysis of the effects of prompts on human decisions

Alberdi, Eugenio; Povyakalo, Andrey; Strigini, Lorenzo; Ayton, Peter; Given-Wilson, Rosalind

doi:10.1007/s11548-008-0213-x

Cited by 23 publications

(24 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors called this an error of complacency (Skitka et al, 1999), in that participants trusted the automated assistance more than they should have, given its actual reliability. The same pattern of results was obtained by Alberdi, Povyakalo, Strigini, Ayton, and Given-Wilson (2008) in a study involving automated assistance for lesion detection in mammography. Again, the participants' detection performances increased when the assistance provided relevant information but declined when it provided irrelevant information.…”

Section: Introductionsupporting

confidence: 75%

See 1 more Smart Citation

Application of the Resources Model to the Supervision of an Automated Process

Fleury

Jamet

Ghorbel

et al. 2014

Human–Computer Interaction

View full text Add to dashboard Cite

International audienceThe present study was designed to ascertain how far flagging up potential errors can improve the automatic interpretation of technical documents. We used the resources model to analyze the supervised retro-conversion of architectural floor plans, from the perspective of distributed cognition. Results showed that automated assistance helps users to correct errors spotted by the system and also saves time. Surprisingly, they also showed that flagging up possible errors may make users less effective in identifying and correcting errors that go unnoticed by the system. Responses to a questionnaire probing the participants’ confidence in the system suggested that they were so trusting that they lowered their vigilance in those areas that had not been signaled by the system, leading to the identification of fewer errors there. Thus, while the participants’ confidence in the automated assistance system led to improved performances in those areas it highlighted, it also meant that areas to which the system did not draw attention were less thoroughly checked

show abstract

Section: Introductionsupporting

confidence: 75%

“…Again, the participants' detection performances increased when the assistance provided relevant information but declined when it provided irrelevant information. However, neither Skitka et al (1999) nor Alberdi et al (2008) actually measured participant confidence in order to confirm their interpretations of the results.…”

Section: Introductionmentioning

confidence: 86%

Application of the Resources Model to the Supervision of an Automated Process

Fleury

Jamet

Ghorbel

et al. 2014

Human–Computer Interaction

View full text Add to dashboard Cite

show abstract

“…In fact, operators' performance could be worse with computer support even for demands for which the tool provides correct advice. Evidence from the case study on CAD in breast screening strongly supports this view [35,36]. Nodes 18 and 16, in conjunction with node 6, illustrate the "cry wolf" situation that may explain phenomena like this.…”

Section: A Case Study: Computer Aided Detection (Cad) For Mammographymentioning

confidence: 80%

“…Many of the considerations we present originate from a case study we conducted in the area of CAD for breast cancer screening [35][36][37][38][39][40]. In breast screening, expert clinicians ("readers") examine mammograms (X-ray images of a woman's breasts), and decide whether the patient should be "recalled" for further tests because they suspect cancer.…”

Section: A Case Study: Computer Aided Detection (Cad) For Mammographymentioning

confidence: 99%

Why Are People’s Decisions Sometimes Worse with Computer Support?

Alberdi

Strigini

Povyakalo

et al. 2009

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

This is the unspecified version of the paper.This version of the publication may differ from the final published version. Abstract. In many applications of computerised decision support, a recognised source of undesired outcomes is operators' apparent over-reliance on automation. For instance, an operator may fail to react to a potentially dangerous situation because a computer fails to generate an alarm. However, the very use of terms like "over-reliance" betrays possible misunderstandings of these phenomena and their causes, which may lead to ineffective corrective action. For instance, training or procedural changes are favored responses, but they do not address all causes of apparently "over-reliant" behaviour. We review relevant literature in the area of "automation bias" and describe the diverse mechanisms that may be involved in human errors when using computer support. We discuss these mechanisms, with reference to errors of omission when using "alerting systems", with the help of examples of novel counterintuitive findings we obtained from a case study in a health care application, as well as other examples from the literature. Permanent repository link

show abstract

“…Sarter & Woods, 1992;Sarter & Woods, 1994;) or examined the possible implications of a new tool (e.g. Loft, Smith, & Bhaskara, 2011;Alberdi et al, 2008) and have not directly compared different levels of automation in terms of workload, performance and operator behaviour. This experiment achieved this and also specifically examined the impact of disruption, or non-normal system operation, an aspect which is frequently omitted in the existing body of research (Sauer, Nickel, & Wastell, 2013).…”

Section: Effects Of Automationmentioning

confidence: 99%

Impact of automation: Measurement of performance, workload and behaviour in a complex control environment

2015

View full text Add to dashboard Cite

This paper describes an experiment that was undertaken to compare three levels of automation in rail signalling; a high level in which an automated agent set routes for trains using timetable information, a medium level in which trains were routed along pre-defined paths, and a low level where the operator (signaller) was responsible for the movement of all trains. These levels are described in terms of a rail automation model based on previous automation theory (Parasuraman, Sheridan, & Wickens, 2000). Performance, subjective workload, and signaller activity were measured for each level of automation running under both normal operating conditions and abnormal, or disrupted, conditions. The results indicate that perceived workload, during both normal and disrupted phases of the experiment, decreased as the level of automation increased and performance was most consistent (i.e. showed the least variation between participants) with the highest level of automation. The results give a strong case in favour of automation, particularly in terms of demonstrating the potential for automation to reduce workload, but also suggest much benefit can achieved from a mid-level of automation potentially at a lower cost and complexity. Impact StatementResearch in the area of automation, and in particular in the examination of human interaction with different levels of automation, has normally been undertaken in Balfe et al, Impact of Automation: Measurement of Performance, Workload and Behaviour in a Complex Control Environment 2 laboratory settings using simple tasks and naïve participants where the level of automation can be easily manipulated. This research was undertaken with expert participants using complex simulation of three ecologically valid levels of automation and provides empirical field validation of some of the results found in laboratory studies.

show abstract

CAD in mammography: lesion-level versus case-level analysis of the effects of prompts on human decisions

Cited by 23 publications

References 16 publications

Application of the Resources Model to the Supervision of an Automated Process

Application of the Resources Model to the Supervision of an Automated Process

Why Are People’s Decisions Sometimes Worse with Computer Support?

Impact of automation: Measurement of performance, workload and behaviour in a complex control environment

Contact Info

Product

Resources

About