Multiple diagnostic task performance in CT examination of the chest

Artificial Intelligence (AI) for Lung Nodules, From the AJR Special Series on AI Applications

Interest in artificial intelligence (AI) applications for lung nodules continues to grow among radiologists, particularly with the expanding eligibility criteria and clinical utilization of lung cancer screening CT. AI has been heavily investigated for detecting and characterizing lung nodules and for guiding prognostic assessment. AI tools have also been used for image postprocessing (e.g., rib suppression on radiography or vessel suppression on CT) and for noninterpretive aspects of reporting and workflow, including management of nodule follow-up. Despite growing interest in and rapid development of AI tools and FDA approval of AI tools for pulmonary nodule evaluation, integration into clinical practice has been limited. Challenges to clinical adoption have included concerns about generalizability, regulatory issues, technical hurdles in implementation, and human skepticism. Further validation of AI tools for clinical use and demonstration of benefit in terms of patient-oriented outcomes also are needed. This article provides an overview of potential applications of AI tools in the imaging evaluation of lung nodules and discusses the challenges faced by practices interested in clinical implementation of such tools.

From "satisfaction of search" to "subsequent search misses": a review of multiple-target search errors across radiology and cognitive science

For over 50 years, the satisfaction of search effect has been studied within the field of radiology. Defined as a decrease in detection rates for a subsequent target when an initial target is found within the image, these multiple target errors are known to underlie errors of omission (e.g., a radiologist is more likely to miss an abnormality if another abnormality is identified). More recently, they have also been found to underlie lab-based search errors in cognitive science experiments (e.g., an observer is more likely to miss a target 'T' if a different target 'T' was detected). This phenomenon was renamed the subsequent search miss (SSM) effect in cognitive science. Here we review the SSM literature in both radiology and cognitive science and discuss: (1) the current SSM theories (i.e., satisfaction, perceptual set, and resource depletion theories), (2) the eye movement errors that underlie the SSM effect, (3) the existing efforts tested to alleviate SSM errors, and (4) the evolution of methodologies and analyses used when calculating the SSM effect. Finally, we present the attentional template theory, a novel mechanistic explanation for SSM errors, which ties together our current understanding of SSM errors and the attentional template literature.

Satisfaction of Search in Chest Radiography 2015

RATIONALE AND OBJECTIVES

Two decades have passed since the publication of laboratory studies of satisfaction of search (SOS) in chest radiography. Those studies were performed using film. The current investigation tests for SOS effects in computed radiography of the chest.

METHODS

Sixty-four chest computed radiographs half demonstrating various "test" abnormalities were read twice by 20 radiologists, once with and once without the addition of a simulated pulmonary nodule. Receiver-operating characteristic detection accuracy and decision thresholds were analyzed to study the effects of adding the nodule on detecting the test abnormalities. Results of previous studies were reanalyzed using similar modern techniques.

RESULTS

In the present study, adding nodules did not influence detection accuracy for the other abnormalities (P = .93), but did induce a reluctance to report them (P < .001). Adding nodules did not affect inspection time (P = .58) so the reluctance to report was not associated with reduced search. Reanalysis revealed a similar decision threshold shift that had not been recognized in the early studies of SOS in chest radiography (P < .01) in addition to reduced detection accuracy (P < .01).

CONCLUSIONS

The nature of SOS in chest radiography has changed, but it is not clear why.

ADVANCES IN KNOWLEDGE

SOS may be changing as a function of changes in radiology education and practice.

Satisfaction of search from detection of pulmonary nodules in computed tomography of the chest

RATIONALE AND OBJECTIVES

We tested whether satisfaction of search (SOS) effects that occur in computed tomography (CT) examination of the chest on detection of native abnormalities are produced by the addition of simulated pulmonary nodules.

MATERIALS AND METHODS

Two experiments were conducted. In the first experiment, 70 CT examinations, half that demonstrated diverse, subtle abnormalities and half that demonstrated no native lesions, were read by 18 radiology residents and fellows under two experimental conditions: presented with and without pulmonary nodules. In a second experiment, many of the examinations were replaced to include more salient native abnormalities. This set was read by 14 additional radiology residents and fellows. In both experiments, detection of the natural abnormalities was studied. Receiver operating characteristic (ROC) curve areas for each reader-treatment combination were estimated using empirical and proper ROC models. Additional analyses focused on decision thresholds and visual search time on abnormality-free CT slice ranges. Institutional review board approval and informed consent from 32 participants were obtained.

RESULTS

Observers more often missed diverse native abnormalities when pulmonary nodules were added, but also made fewer false-positive responses. There was no change in ROC area, but decision criteria grew more conservative. The SOS effect on decision thresholds was accompanied by a reduction in search time on abnormality-free CT slice ranges.

CONCLUSION

The SOS effect in CT examination of the chest is similar to that found in contrast examination of the abdomen, involving induced visual neglect.

Do long radiology workdays affect nodule detection in dynamic CT interpretation?

PURPOSE

A previous study demonstrated decreased diagnostic accuracy for finding fractures and decreased ability to focus on skeletal radiographs after a long working day. Skeletal radiographic examinations commonly have images that are displayed statically. The aim of this study was to investigate whether diagnostic accuracy for detecting pulmonary nodules on CT of the chest displayed dynamically would be similarly affected by fatigue.

METHODS

Twenty-two radiologists and 22 residents were given 2 tests searching CT chest sequences for a solitary pulmonary nodule before and after a day of clinical reading. To measure search time, 10 lung CT sequences, each containing 20 consecutive sections and a single nodule, were inspected using free search and navigation. To measure diagnostic accuracy, 100 CT sequences, each with 20 sections and half with nodules, were displayed at preset scrolling speed and duration. Accuracy was measured using receiver operating characteristic curve analysis. Visual strain was measured via dark vergence, an indicator of the ability to keep the eyes focused on the display.

RESULTS

Diagnostic accuracy was reduced after a day of clinical reading (P = .0246), but search time was not affected (P > .05). After a day of reading, dark vergence was significantly larger and more variable (P = .0098), reflecting higher levels of visual strain, and subjective ratings of fatigue were also higher.

CONCLUSIONS

After their usual workday, radiologists experience increased fatigue and decreased diagnostic accuracy for detecting pulmonary nodules on CT. Effects of fatigue may be mitigated by active interaction with the display.