Current perspectives in medical image perception

Great expectations: minor differences in initial instructions have a major impact on visual search in the absence of feedback

Professions such as radiology and aviation security screening that rely on visual search-the act of looking for targets among distractors-often cannot provide operators immediate feedback, which can create situations where performance may be largely driven by the searchers' own expectations. For example, if searchers do not expect relatively hard-to-spot targets to be present in a given search, they may find easy-to-spot targets but systematically quit searching before finding more difficult ones. Without feedback, searchers can create self-fulfilling prophecies where they incorrectly reinforce initial biases (e.g., first assuming and then, perhaps wrongly, concluding hard-to-spot targets are rare). In the current study, two groups of searchers completed an identical visual search task but with just a single difference in their initial task instructions before the experiment started; those in the "high-expectation" condition were told that each trial could have one or two targets present (i.e., correctly implying no target-absent trials) and those in the "low-expectation" condition were told that each trial would have up to two targets (i.e., incorrectly implying there could be target-absent trials). Compared to the high-expectation group, the low-expectation group had a lower hit rate, lower false alarm rate and quit trials more quickly, consistent with a lower quitting threshold (i.e., performing less exhaustive searches) and a potentially higher target-present decision criterion. The expectation effect was present from the start and remained across the experiment-despite exposure to the same true distribution of targets, the groups' performances remained divergent, primarily driven by the different subjective experiences caused by each groups' self-fulfilling prophecies. The effects were limited to the single-targets trials, which provides insights into the mechanisms affected by the initial expectations set by the instructions. In sum, initial expectations can have dramatic influences-searchers who do not expect to find a target, are less likely to find a target as they are more likely to quit searching earlier.

Speech-language therapists' perceptions of an eLearning program to support training in videofluoroscopic swallow studies

BACKGROUND

Speech-language therapists (SLTs) seek a range of educational opportunities for training in adult videofluoroscopic swallow studies (VFSS). However, variable training methods and/or unequal access to training can influence VFSS practice.

AIMS

To document current SLT needs and barriers to VFSS training and to determine if a new beginner-level VFSS eLearning program would assist to meet their training needs. The program incorporated multimedia modules on preparing, conducting, interpreting and reporting VFSS.

METHODS & PROCEDURES

SLTs with limited experience in adult VFSS completed surveys relating to VFSS training experience and barriers, and perceived changes in knowledge, skills and confidence on core VFSS module topics pre- (n = 36) and post- (n = 32) eLearning training.

OUTCOMES & RESULTS

Inconsistent access to VFSS training opportunities and time-related work pressures were reported as the greatest training barriers. SLTs viewed the eLearning program as a suitable option for VFSS training. Post-training, participants perceived they gained confidence, as well as improved knowledge and skills in all VFSS aspects along with generalised benefits for dysphagia management. SLTs indicated that key benefits of the eLearning program were its comprehensive content and self-directed learning with multimedia tools, which afforded theoretical and practical learning opportunities.

CONCLUSIONS & IMPLICATIONS

The eLearning program offered SLTs free access to beginner-level adult VFSS training, meeting many identified training needs and providing a foundation from which to develop further practical knowledge and skills within a VFSS clinic setting. What this paper adds What is already known on the subject SLTs demonstrate variable knowledge and skill in conducting and interpreting VFSS, which can impact dysphagia diagnosis and management. While access to VFSS training can be challenging, the barriers to training for SLTs have not been clearly documented. Research has confirmed that eLearning can be used effectively in healthcare education, and in some aspects of VFSS training; however, it is yet to be applied to address the broad range of VFSS training needs. What this paper adds to existing knowledge This study describes the SLT reported barriers to VFSS training which include limited access to formal and practical training, workload-related time pressures and the complexity of learning the VFSS skill set. The findings highlight that an eLearning program, was an accepted mode of learning for VFSS training. SLTs reported the online program met their learning needs by improving access to training, the multimedia program features supported their understanding of complex anatomical and physiological concepts, and training frameworks assisted their clinical reasoning and VFSS interpretation. What are the potential or actual clinical implications of this work? eLearning can assist in overcoming many VFSS training barriers identified by SLTs and the multimedia aspects of eLearning can effectively support VFSS beginner-level education to complement and expedite in-clinic practical training. Given that VFSS results inform decisions regarding commencement and progression of oral intake and swallow rehabilitation, enhanced VFSS training has the potential to positively influence dysphagia outcomes and quality of life.

Eye movements reflect expertise development in hybrid search

Domain-specific expertise changes the way people perceive, process, and remember information from that domain. This is often observed in visual domains involving skilled searches, such as athletics referees, or professional visual searchers (e.g., security and medical screeners). Although existing research has compared expert to novice performance in visual search, little work has directly documented how accumulating experiences change behavior. A longitudinal approach to studying visual search performance may permit a finer-grained understanding of experience-dependent changes in visual scanning, and the extent to which various cognitive processes are affected by experience. In this study, participants acquired experience by taking part in many experimental sessions over the course of an academic semester. Searchers looked for 20 categories of targets simultaneously (which appeared with unequal frequency), in displays with 0-3 targets present, while having their eye movements recorded. With experience, accuracy increased and response times decreased. Fixation probabilities and durations decreased with increasing experience, but saccade amplitudes and visual span increased. These findings suggest that the behavioral benefits endowed by expertise emerge from oculomotor behaviors that reflect enhanced reliance on memory to guide attention and the ability to process more of the visual field within individual fixations.

Expert radiologist review at a hepatobiliary multidisciplinary tumor board: impact on patient management

PURPOSE

To identify the frequency, source, and management impact of discrepancies between the initial radiology report and expert reinterpretation occurring in the context of a hepatobiliary multidisciplinary tumor board (MTB).

METHODS

This retrospective study included 974 consecutive patients discussed at a weekly MTB at a large tertiary care academic medical center over a 2-year period. A single radiologist with dedicated hepatobiliary imaging expertise attended all conferences to review and discuss the relevant liver imaging and rated the concordance between original and re-reads based on RADPEER scoring criteria. Impact on management was based on the conference discussion and reflected changes in follow-up imaging, recommendations for biopsy/surgery, or liver transplant eligibility.

RESULTS

Image reinterpretation was discordant with the initial report in 19.9% (194/974) of cases (59.8%, 34.5%, 5.7% RADPEER 2/3/4 discrepancies, respectively). A change in LI-RADS category occurred in 59.8% of discrepancies. Most common causes of discordance included re-classification of a lesion as benign rather than malignant (16.0%) and missed tumor recurrence (13.9%). Impact on management occurred in 99.0% of discordant cases and included loco-regional therapy instead of follow-up imaging (19.1%), follow-up imaging instead of treatment (17.5%), and avoidance of biopsy (12.4%). 11.3% received OPTN exception scores due to the revised interpretation, and 8.8% were excluded from listing for orthotopic liver transplant.

CONCLUSION

Even in a sub-specialized abdominal imaging academic practice, expert radiologist review in the MTB setting identified discordant interpretations and impacted management in a substantial fraction of patients, potentially impacting transplant allocation. The findings may impact how abdominal imaging sections best staff advanced MTBs.

What do radiologists look for? Advances and limitations of perceptual learning in radiologic search

Supported by guidance from training during residency programs, radiologists learn clinically relevant visual features by viewing thousands of medical images. Yet the precise visual features that expert radiologists use in their clinical practice remain unknown. Identifying such features would allow the development of perceptual learning training methods targeted to the optimization of radiology training and the reduction of medical error. Here we review attempts to bridge current gaps in understanding with a focus on computational saliency models that characterize and predict gaze behavior in radiologists. There have been great strides toward the accurate prediction of relevant medical information within images, thereby facilitating the development of novel computer-aided detection and diagnostic tools. In some cases, computational models have achieved equivalent sensitivity to that of radiologists, suggesting that we may be close to identifying the underlying visual representations that radiologists use. However, because the relevant bottom-up features vary across task context and imaging modalities, it will also be necessary to identify relevant top-down factors before perceptual expertise in radiology can be fully understood. Progress along these dimensions will improve the tools available for educating new generations of radiologists, and aid in the detection of medically relevant information, ultimately improving patient health.

Interobserver variability in quality assessment of magnetic resonance images

Background

The perceptual quality of magnetic resonance (MR) images influences diagnosis and may compromise the treatment. The purpose of this study was to evaluate how the image quality changes influence the interobserver variability of their assessment.

Methods

For the variability evaluation, a dataset containing distorted MRI images was prepared and then assessed by 31 experienced medical professionals (radiologists). Differences between observers were analyzed using the Fleiss’ kappa. However, since the kappa evaluates the agreement among radiologists taking into account aggregated decisions, a typically employed criterion of the image quality assessment (IQA) performance was used to provide a more thorough analysis. The IQA performance of radiologists was evaluated by comparing the Spearman correlation coefficients, ρ, between individual scores with the mean opinion scores (MOS) composed of the subjective opinions of the remaining professionals.

Results

The experiments show that there is a significant agreement among radiologists (κ=0.12; 95% confidence interval [CI]: 0.118, 0.121; P<0.001) on the quality of the assessed images. The resulted κ is strongly affected by the subjectivity of the assigned scores, separately presenting close scores. Therefore, the ρ was used to identify poor performance cases and to confirm the consistency of the majority of collected scores (ρ_mean = 0.5706). The results for interns (ρ_mean = 0.6868) supports the finding that the quality assessment of MR images can be successfully taught.

Conclusions

The agreement observed among radiologists from different imaging centers confirms the subjectivity of the perception of MR images. It was shown that the image content and severity of distortions affect the IQA. Furthermore, the study highlights the importance of the psychosomatic condition of the observers and their attitude.

Liver imaging features by convolutional neural network to predict the metachronous liver metastasis in stage I-III colorectal cancer patients based on preoperative abdominal CT scan

Background

Introducing deep learning approach to medical images has rendered a large amount of un-decoded information into usage in clinical research. But mostly, it has been focusing on the performance of the prediction modeling for disease-related entity, but not on the clinical implication of the feature itself. Here we analyzed liver imaging features of abdominal CT images collected from 2019 patients with stage I – III colorectal cancer (CRC) using convolutional neural network (CNN) to elucidate its clinical implication in oncological perspectives.

Results

CNN generated imaging features from the liver parenchyma. Dimension reduction was done for the features by principal component analysis. We designed multiple prediction models for 5-year metachronous liver metastasis (5YLM) using combinations of clinical variables (age, sex, T stage, N stage) and top principal components (PCs), with logistic regression classification. The model using “1^st PC (PC1) + clinical information” had the highest performance (mean AUC = 0.747) to predict 5YLM, compared to the model with clinical features alone (mean AUC = 0.709). The PC1 was independently associated with 5YLM in multivariate analysis (beta = − 3.831, P < 0.001). For the 5-year mortality rate, PC1 did not contribute to an improvement to the model with clinical features alone. For the PC1, Kaplan-Meier plots showed a significant difference between PC1 low vs. high group. The 5YLM-free survival of low PC1 was 89.6% and the high PC1 was 95.9%. In addition, PC1 had a significant correlation with sex, body mass index, alcohol consumption, and fatty liver status.

Conclusion

The imaging features combined with clinical information improved the performance compared to the standardized prediction model using only clinical information. The liver imaging features generated by CNN may have the potential to predict liver metastasis. These results suggest that even though there were no liver metastasis during the primary colectomy, the features of liver imaging can impose characteristics that could be predictive for metachronous liver metastasis.

Evaluation of Image Quality for 7 Iterative Reconstruction Algorithms in Chest Computed Tomography Imaging: A Phantom Study

OBJECTIVES

This study aimed to evaluate the image quality of 7 iterative reconstruction (IR) algorithms in comparison to filtered back-projection (FBP) algorithm.

METHODS

An anthropomorphic chest phantom was scanned on 4 computed tomography scanners and reconstructed with FBP and IR algorithms. Image quality of anatomical details-large/medium-sized pulmonary vessels, small pulmonary vessels, thoracic wall, and small and large lesions-was scored. Furthermore, general impression of noise, image contrast, and artifacts were evaluated. Visual grading regression was used to analyze the data. Standard deviations were measured, and the noise power spectrum was calculated.

RESULTS

Iterative reconstruction algorithms showed significantly better results when compared with FBP for these criteria (regression coefficients/P values in parentheses): vessels (FIRST: -1.8/0.05, AIDR Enhanced: <-2.3/0.01, Veo: <-0.1/0.03, ADMIRE: <-2.1/0.04), lesions (FIRST: <-2.6/0.01, AIDR Enhanced: <-1.9/0.03, IMR1: <-2.7/0.01, Veo: <-2.4/0.02, ADMIRE: -2.3/0.02), image noise (FIRST: <-3.2/0.004, AIDR Enhanced: <-3.5/0.002, IMR1: <-6.1/0.001, iDose: <-2.3/0.02, Veo: <-3.4/0.002, ADMIRE: <-3.5/0.02), image contrast (FIRST: -2.3/0.01, AIDR Enhanced: -2.5/0.01, IMR1: -3.7/0.001, iDose: -2.1/0.02), and artifacts (FIRST: <-3.8/0.004, AIDR Enhanced: <-2.7/0.02, IMR1: <-2.6/0.02, iDose: -2.1/0.04, Veo: -2.6/0.02). The iDose algorithm was the only IR algorithm that maintained the noise frequencies.

CONCLUSIONS

Iterative reconstruction algorithms performed differently on all evaluated criteria, showing the importance of careful implementation of algorithms for diagnostic purposes.

Search of low-contrast liver lesions in abdominal CT: the importance of scrolling behavior

Purpose: Visual search using volumetric images is becoming the standard in medical imaging. However, we do not fully understand how eye movement strategies mediate diagnostic performance. A recent study on computed tomography (CT) images showed that the search strategies of radiologists could be classified based on saccade amplitudes and cross-quadrant eye movements [eye movement index (EMI)] into two categories: drillers and scanners. Approach: We investigate how the number of times a radiologist scrolls in a given direction during analysis of the images (number of courses) could add a supplementary variable to use to characterize search strategies. We used a set of 15 normal liver CT images in which we inserted 1 to 5 hypodense metastases of two different signal contrast amplitudes. Twenty radiologists were asked to search for the metastases while their eye-gaze was recorded by an eye-tracker device (EyeLink1000, SR Research Ltd., Mississauga, Ontario, Canada). Results: We found that categorizing radiologists based on the number of courses (rather than EMI) could better predict differences in decision times, percentage of image covered, and search error rates. Radiologists with a larger number of courses covered more volume in more time, found more metastases, and made fewer search errors than those with a lower number of courses. Our results suggest that the traditional definition of drillers and scanners could be expanded to include scrolling behavior. Drillers could be defined as scrolling back and forth through the image stack, each time exploring a different area on each image (low EMI and high number of courses). Scanners could be defined as scrolling progressively through the stack of images and focusing on different areas within each image slice (high EMI and low number of courses). Conclusions: Together, our results further enhance the understanding of how radiologists investigate three-dimensional volumes and may improve how to teach effective reading strategies to radiology residents.

The effect of perceptual decision-making on the interpretation of twin fetal heart rate tracings

Objective: Decision-making is an integrative process during which multiple sources of available evidence are combined into a singular response. Importantly, subconscious processes occur in perceptual decisions that may influence interpretations of visually displayed data such as fetal heart rate tracings (FHRT), which are typically presented together for twins. To examine the potential impact of subconscious perceptual influences on fetal well-being, differences in assessments of FHRTs for twin gestations presented singly or paired were evaluated for baseline fetal heart rate, variability, accelerations, decelerations, and overall concern.Study design: Obstetrical nurses (N = 27) assessed FHRTs from 20 twin gestations (each of which had at least one live birth with a 5-min Apgar <7) presented either on the same tracing or as singletons on separate tracings. Nurses were naïve to the fact that the fetal heart rate tracings presented in the unpaired condition were the same as those presented in the paired condition. Assessments were then compared between the two conditions.Results: Each nurse participant completed ratings on five metrics for each of 20 twin gestations across two conditions (80 FHRT assessments, 400 metrics total per participant). The intraobserver impact of visual context was calculated as the frequency of changed opinions regarding an individual metric (e.g. variability) between the paired and unpaired contexts for each individual fetal heart rate. Assessments of variability (average Kappa = 0.59), decelerations (average Kappa = 0.34), and overall level of concern (average Kappa = 0.33) were moderately to heavily impacted by viewing condition (unpaired vs. paired FHRT). Analysis of interobserver agreement using intraclass correlations (two-way random effect, absolute agreement) indicates poor agreement on unpaired assessments for both accelerations (ICC = 0.01, 95% CI -0.01-0.04) and decelerations (ICC = 0.22, 95% CI 0.15-0.33). These results are mirrored in poor agreement on paired assessments for both accelerations (ICC = 0.00, 95% CI -0.01-0.03) and decelerations (ICC = 0.27, 95% CI 0.19-0.39). There was moderate agreement on overall level of concern for unpaired assessments (ICC = 0.55, 95% CI 0.44-0.67) and near moderate agreement for the paired condition (ICC = 0.45, 95% CI 0.34-0.58).Conclusions: The simultaneous presentation of fetal heart rate tracings in twin gestations introduces both intraobserver and interobserver variances in the interpretation of variability, accelerations, and decelerations, likely due to the influence of subconscious perceptual decision-making. This may theoretically affect outcomes in cases in which visual information is nuanced. More research is necessary to determine whether the standard protocol of simultaneous assessment of FHRT in twins is subliminally affected by perceptual decision-making.

Deep Learning to Automate Reference-Free Image Quality Assessment of Whole-Heart MR Images

Purpose

To develop and characterize an algorithm that mimics human expert visual assessment to quantitatively determine the quality of three-dimensional (3D) whole-heart MR images.

Materials and Methods

In this study, 3D whole-heart cardiac MRI scans from 424 participants (average age, 57 years ± 18 [standard deviation]; 66.5% men) were used to generate an image quality assessment algorithm. A deep convolutional neural network for image quality assessment (IQ-DCNN) was designed, trained, optimized, and cross-validated on a clinical database of 324 (training set) scans. On a separate test set (100 scans), two hypotheses were tested: (a) that the algorithm can assess image quality in concordance with human expert assessment as assessed by human-machine correlation and intra- and interobserver agreement and (b) that the IQ-DCNN algorithm may be used to monitor a compressed sensing reconstruction process where image quality progressively improves. Weighted κ values, agreement and disagreement counts, and Krippendorff α reliability coefficients were reported.

Results

Regression performance of the IQ-DCNN was within the range of human intra- and interobserver agreement and in very good agreement with the human expert (R ² = 0.78, κ = 0.67). The image quality assessment during compressed sensing reconstruction correlated with the cost function at each iteration and was successfully applied to rank the results in very good agreement with the human expert.

Conclusion

The proposed IQ-DCNN was trained to mimic expert visual image quality assessment of 3D whole-heart MR images. The results from the IQ-DCNN were in good agreement with human expert reading, and the network was capable of automatically comparing different reconstructed volumes.Supplemental material is available for this article.© RSNA, 2020.

How to develop a meaningful radiomic signature for clinical use in oncologic patients

During the last decade, there is an increasing usage of quantitative methods in Radiology in an effort to reduce the diagnostic variability associated with a subjective manner of radiological interpretation. Combined approaches where visual assessment made by the radiologist is augmented by quantitative imaging biomarkers are gaining attention. Advances in machine learning resulted in the rise of radiomics that is a new methodology referring to the extraction of quantitative information from medical images. Radiomics are based on the development of computational models, referred to as “Radiomic Signatures”, trying to address either unmet clinical needs, mostly in the field of oncologic imaging, or to compare radiomics performance with that of radiologists. However, to explore this new technology, initial publications did not consider best practices in the field of machine learning resulting in publications with questionable clinical value. In this paper, our effort was concentrated on how to avoid methodological mistakes and consider critical issues in the workflow of the development of clinically meaningful radiomic signatures.

Development and validation of the Lesion Synthesis Toolbox and the Perception Study Tool for quantifying observer limits of detection of lesions in positron emission tomography

Purpose: Accurate detection of cancer lesions in positron emission tomography (PET) is fundamental to achieving favorable clinical outcomes. Therefore, image reconstruction, processing, visualization, and interpretation techniques must be optimized for this task. The objective of this work was to (1) develop and validate an efficient method to generate well-characterized synthetic lesions in real patient data and (2) to apply these lesions in a human perception experiment to establish baseline measurements of the limits of lesion detection as a function of lesion size and contrast using current imaging technologies. Approach: A fully integrated software package for synthesizing well-characterized lesions in real patient PET was developed using a vendor provided PET image reconstruction toolbox (REGRECON5, General Electric Healthcare, Waukesha, Wisconsin). Lesion characteristics were validated experimentally for geometric accuracy, activity accuracy, and absence of artifacts. The Lesion Synthesis Toolbox was used to generate a library of 133 synthetic lesions of varying sizes ( n = 7 ) and contrast levels ( n = 19 ) in manually defined locations in the livers of 37 patient studies. A lesion-localization perception study was performed with seven observers to determine the limits of detection with regard to lesion size and contrast using our web-based perception study tool. Results: The Lesion Synthesis Toolbox was validated for accurate lesion placement and size. Lesion intensities were deemed accurate with slightly elevated activities (5% at 2:1 lesion-to-background contrast) in small lesions ( Ø = 15    mm spheres), and no bias in large lesions ( Ø = 22.5    mm ). Bed-stitching artifacts were not observed, and lesion attenuation correction bias was small ( - 1.6 ± 1.2 % ). The 133 liver lesions were synthesized in ∼ 50    h , and readers were able to complete the perception study of these lesions in 12 ± 3    min with consistent limits of detection amongst all readers. Conclusions: Our open-source utilities can be employed by nonexperts to generate well-characterized synthetic lesions in real patient PET images and for administering perception studies on clinical workstations without the need to install proprietary software.

Taking pigeons to heart: Birds proficiently diagnose human cardiac disease

In two experiments, we trained pigeons (Columba livia) to sort visual images (obtained by clinical myocardial perfusion imaging techniques) depicting different degrees of human cardiac disfunction (myocardial hypoperfusion of the left ventricle) into normal and abnormal categories by providing food reward only after correct choice responses. Pigeons proved to be highly proficient at categorizing pseudo-colorized images as well as highly sensitive to the degree of the perfusion deficit depicted in the abnormal images. In later testing, the pigeons completely transferred discriminative responding to novel stimuli, demonstrating that they had fully learned the normal and abnormal categories. Yet, these pigeons failed to transfer discriminative responding to grayscale images containing no color information. We therefore trained a second cohort of pigeons to categorize grayscale image sets from the outset. These birds required substantially more training to achieve similar levels of performance. Yet, they too completely transferred discriminative responding to novel stimuli by relying on both global and local disparities in brightness between the normal and abnormal images. These results confirm that pseudo-colorization can enhance pigeons' categorization of human cardiac images, a result also found with human observers. Overall, our findings further document the potential of the pigeon as a useful aide in studies of medical image perception.

Automatic Lung Nodule Detection Combined With Gaze Information Improves Radiologists' Screening Performance

Early diagnosis of lung cancer via computed tomography can significantly reduce the morbidity and mortality rates associated with the pathology. However, searching lung nodules is a high complexity task, which affects the success of screening programs. Whilst computer-aided detection systems can be used as second observers, they may bias radiologists and introduce significant time overheads. With this in mind, this study assesses the potential of using gaze information for integrating automatic detection systems in the clinical practice. For that purpose, 4 radiologists were asked to annotate 20 scans from a public dataset while being monitored by an eye tracker device, and an automatic lung nodule detection system was developed. Our results show that radiologists follow a similar search routine and tend to have lower fixation periods in regions where finding errors occur. The overall detection sensitivity of the specialists was 0.67±0.07, whereas the system achieved 0.69. Combining the annotations of one radiologist with the automatic system significantly improves the detection performance to similar levels of two annotators. Filtering automatic detection candidates only for low fixation regions still significantly improves the detection sensitivity without increasing the number of false-positives.

Perceptual learning in the identification of lung cancer in chest radiographs

Extensive research has shown that practice yields highly specific perceptual learning of simple visual properties such as orientation and contrast. Does this same learning characterize more complex perceptual skills? Here we investigated perceptual learning of complex medical images. Novices underwent training over four sessions to discriminate which of two chest radiographs contained a tumor and to indicate the location of the tumor. In training, one group received six repetitions of 30 normal/abnormal images, the other three repetitions of 60 normal/abnormal images. Groups were then tested on trained and novel images. To assess the nature of perceptual learning, test items were presented in three formats – the full image, the cutout of the tumor, or the background only. Performance improved across training sessions, and notably, the improvement transferred to the classification of novel images. Training with more repetitions on fewer images yielded comparable transfer to training with fewer repetitions on more images. Little transfer to novel images occurred when tested with just the cutout of the cancer region or just the background, but a larger cutout that included both the cancer region and some surrounding regions yielded good transfer. Perceptual learning contributes to the acquisition of expertise in cancer image perception.

The role of artificial intelligence in colon polyps detection

Over the past few decades, artificial intelligence (AI) has evolved dramatically and is believed to have a significant impact on all aspects of technology and daily life. The use of AI in the healthcare system has been rapidly growing, owing to the large amount of data. Various methods of AI including machine learning, deep learning and convolutional neural network (CNN) have been used in diagnostic imaging, which have helped physicians in the accurate diagnosis of diseases and determination of appropriate treatment for them. Using and collecting a huge number of digital images and medical records has led to the creation of big data over a time period. Currently, considerations regarding the diagnosis of various presentations in all endoscopic procedures and imaging findings are solely handled by endoscopists. Moreover, AI has shown to be highly effective in the field of gastroenterology in terms of diagnosis, prognosis, and image processing. Herein, this review aimed to discuss different aspects of AI use for early detection and treatment of gastroenterology diseases.

A Review of Perceptual Expertise in Radiology-How it develops, How we can test it, and Why humans still matter in the era of Artificial Intelligence

As the first step in image interpretation is detection, an error in perception can prematurely end the diagnostic process leading to missed diagnoses. Because perceptual errors of this sort-"failure to detect"-are the most common interpretive error (and cause of litigation) in radiology, understanding the nature of perceptual expertise is essential in decreasing radiology's long-standing error rates. In this article, we review what constitutes a perceptual error, the existing models of radiologic image perception, the development of perceptual expertise and how it can be tested, perceptual learning methods in training radiologists, and why understanding perceptual expertise is still relevant in the era of artificial intelligence. Adding targeted interventions, such as perceptual learning, to existing teaching practices, has the potential to enhance expertise and reduce medical error.

Evaluation of Gaze Tracking Calibration for Longitudinal Biomedical Imaging Studies

Gaze tracking is a promising technology for studying the visual perception of clinicians during image-based medical exams. It could be used in longitudinal studies to analyze their perceptive process, explore human-machine interactions, and develop innovative computer-aided imaging systems. However, using a remote eye tracker in an unconstrained environment and over time periods of weeks requires a certain guarantee of performance to ensure that collected gaze data are fit for purpose. We report the results of evaluating eye tracking calibration for longitudinal studies. First, we tested the performance of an eye tracker on a cohort of 13 users over a period of one month. For each participant, the eye tracker was calibrated during the first session. The participants were asked to sit in front of a monitor equipped with the eye tracker, but their position was not constrained. Second, we tested the performance of the eye tracker on sonographers positioned in front of a cart-based ultrasound scanner. Experimental results show a decrease of accuracy between calibration and later testing of 0.30° and a further degradation over time at a rate of 0.13°. month^-1. The overall median accuracy was 1.00° (50.9 pixels) and the overall median precision was 0.16° (8.3 pixels). The results from the ultrasonography setting show a decrease of accuracy of 0.16° between calibration and later testing. This slow degradation of gaze tracking accuracy could impact the data quality in long-term studies. Therefore, the results we present here can help in planning such long-term gaze tracking studies.

An Eye Tracking Analysis on Diagnostic Performance of Digital Fundus Photography Images between Ophthalmologists and Optometrists

To investigate the parameters of eye movement between ophthalmologists and optometrists while diagnosing digital fundus photographs, sixteen participants (eight ophthalmologists and eight optometrists) were recruited in this study. Every participant’s eye movement during diagnosis of a randomized set of fundus photographs displayed on an eye tracker were recorded. Fixation metrics (duration, count and rate) and scan path patterns were extracted from the eye tracker. These parameters of eye movement and correct diagnosis score were compared between both groups. Correlation analyses between fixation metrics and correct diagnosis score were also performed. Although fixation metrics between ophthalmologists and optometrists were not statistically different (p > 0.05), these parameters were statistically different when compared between different area of interests. Both participant groups had a similar correct diagnosis score. No correlation was found between fixation metrics and correct diagnosis score between both groups, except for total fixation duration and ophthalmologists’ diagnosis score of diabetic retinopathy photographs. The ophthalmologists’ scan paths were simpler, with larger saccades, and were distributed at the middle region of the photographs. Conversely, optometrists’ scan paths were extensive, with shorter saccades covering wider fundus areas, and were accumulated in some unrelated fundus areas. These findings indicated comparable efficiency and systematic visual search patterns between both the groups. Understanding visual search strategy could expedite the creation of a novel training routine for interpretation of ophthalmic diagnostic imaging.

Chest Radiograph Interpretation with Deep Learning Models: Assessment with Radiologist-adjudicated Reference Standards and Population-adjusted Evaluation

BackgroundDeep learning has the potential to augment the use of chest radiography in clinical radiology, but challenges include poor generalizability, spectrum bias, and difficulty comparing across studies.PurposeTo develop and evaluate deep learning models for chest radiograph interpretation by using radiologist-adjudicated reference standards.Materials and MethodsDeep learning models were developed to detect four findings (pneumothorax, opacity, nodule or mass, and fracture) on frontal chest radiographs. This retrospective study used two data sets. Data set 1 (DS1) consisted of 759 611 images from a multicity hospital network and ChestX-ray14 is a publicly available data set with 112 120 images. Natural language processing and expert review of a subset of images provided labels for 657 954 training images. Test sets consisted of 1818 and 1962 images from DS1 and ChestX-ray14, respectively. Reference standards were defined by radiologist-adjudicated image review. Performance was evaluated by area under the receiver operating characteristic curve analysis, sensitivity, specificity, and positive predictive value. Four radiologists reviewed test set images for performance comparison. Inverse probability weighting was applied to DS1 to account for positive radiograph enrichment and estimate population-level performance.ResultsIn DS1, population-adjusted areas under the receiver operating characteristic curve for pneumothorax, nodule or mass, airspace opacity, and fracture were, respectively, 0.95 (95% confidence interval [CI]: 0.91, 0.99), 0.72 (95% CI: 0.66, 0.77), 0.91 (95% CI: 0.88, 0.93), and 0.86 (95% CI: 0.79, 0.92). With ChestX-ray14, areas under the receiver operating characteristic curve were 0.94 (95% CI: 0.93, 0.96), 0.91 (95% CI: 0.89, 0.93), 0.94 (95% CI: 0.93, 0.95), and 0.81 (95% CI: 0.75, 0.86), respectively.ConclusionExpert-level models for detecting clinically relevant chest radiograph findings were developed for this study by using adjudicated reference standards and with population-level performance estimation. Radiologist-adjudicated labels for 2412 ChestX-ray14 validation set images and 1962 test set images are provided.© RSNA, 2019Online supplemental material is available for this article.See also the editorial by Chang in this issue.

Effects of time of day on radiological interpretation

Accurate interpretation of radiological images involves a complex visual search that relies on several cognitive processes, including selective attention, working memory, and decision-making. Patient outcomes often depend on the accuracy of image interpretations, and yet research has revealed that conclusions vary significantly from one radiologist to another. A myriad of factors has been shown to contribute to the likelihood of interpretative errors and discrepancies, including the radiologist's level of experience and fatigue, and these factors are well reported elsewhere; however, a potentially important factor that has been given little previous consideration is how radiologists' interpretations might be impacted by the time of day at which the reading takes place, a factor that other disciplines have shown to be a determinant of competency. The available literature shows that while the time of day is known to significantly impact some cognitive functions that likely relate to reading competence, including selective visual attention and visual working memory, little is known about the impact of the time of day on radiology interpretation performance. This review explores the evidence regarding the relationship between time of day and performance, with a particular emphasis on radiological activities.

Introduction of Ultrasound Simulation in Medical Education: Exploratory Study

BACKGROUND

Ultrasound is ubiquitous across all disciplines of medicine; it is one of the most commonly used noninvasive, painless diagnostic tools. However, not many are educated and trained well enough in its use. Ultrasound requires not only theoretical knowledge but also extensive practical experience. The simulated setting offers the safest environment for health care professionals to learn and practice using ultrasound.

OBJECTIVE

This study aimed to (1) assess health care professionals' need for and enthusiasm toward practicing using ultrasound via simulation and (2) gauge their perception and acceptance of simulation as an integral element of ultrasound education in medical curricula.

METHODS

A day-long intervention was organized at the American University of Beirut Medical Center (AUBMC) to provide a free-of-charge interactive ultrasound simulation workshop-using CAE Vimedix high-fidelity simulator-for health care providers, including physicians, nurses, ultrasound technicians, residents, and medical students. Following the intervention, attendees completed an evaluation, which included 4 demographic questions and 16 close-ended questions based on a Likert scale agree-neutral-disagree. The results presented are based on this evaluation form.

RESULTS

A total of 41 participants attended the workshop (46% [19/41] physicians, 30% [12/41] residents, 19% [8/41] sonographers, and 5% [2/41] medical students), mostly from AUBMC (88%, 36/41), with an average experience of 2.27 (SD 3.45) years and 30 (SD 46) scans per attendee. Moreover, 15 out of 41 (36%) participants were from obstetrics and gynecology, 11 (27%) from internal medicine, 4 (10%) from pediatrics, 4 (10%) from emergency medicine, 2 (5%) from surgery and family medicine, and 5 (12%) were technicians. The majority of participants agreed that ultrasound provided a realistic setting (98%, 40/41) and that it allowed for training and identification of pathologies (88%, 36/41). Furthermore, 100% (41/41) of the participants agreed that it should be part of the curriculum either in medical school or residency, and most of the participants approved it for training (98%, 40/41) and teaching (98%, 40/41).

CONCLUSIONS

All attendees were satisfied with the intervention. There was a positive perception toward the use of simulation for training and teaching medical students and residents in using ultrasound, and there was a definite need and enthusiasm for its integration into curricula. Simulation offers an avenue not only for teaching but also for practicing the ultrasound technology by both medical students and health care providers.

Real-time three-dimensional MRI for the assessment of dynamic carpal instability

Background

Carpal instability is defined as a condition where wrist motion and/or loading creates mechanical dysfunction, resulting in weakness, pain and decreased function. When conventional methods do not identify the instability patterns, yet clinical signs of instability exist, the diagnosis of dynamic instability is often suggested to describe carpal derangement manifested only during the wrist’s active motion or stress. We addressed the question: can advanced MRI techniques provide quantitative means to evaluate dynamic carpal instability and supplement standard static MRI acquisition? Our objectives were to (i) develop a real-time, three-dimensional MRI method to image the carpal joints during their active, uninterrupted motion; and (ii) demonstrate feasibility of the method for assessing metrics relevant to dynamic carpal instability, thus overcoming limitations of standard MRI.

Methods

Twenty wrists (bilateral wrists of ten healthy participants) were scanned during radial-ulnar deviation and clenched-fist maneuvers. Images resulting from two real-time MRI pulse sequences, four sparse data-acquisition schemes, and three constrained image reconstruction techniques were compared. Image quality was assessed via blinded scoring by three radiologists and quantitative imaging metrics.

Results

Real-time MRI data-acquisition employing sparse radial sampling with a gradient-recalled-echo acquisition and constrained iterative reconstruction appeared to provide a practical tradeoff between imaging speed (temporal resolution up to 135 ms per slice) and image quality. The method effectively reduced streaking artifacts arising from data undersampling and enabled the derivation of quantitative measures pertinent to evaluating dynamic carpal instability.

Conclusion

This study demonstrates that real-time, three-dimensional MRI of the moving wrist is feasible and may be useful for the evaluation of dynamic carpal instability.

Visual search in breast imaging

Breast cancer is the most common cancer among females worldwide and large volumes of breast images are produced and interpreted annually. As long as radiologists interpret these images, the diagnostic accuracy will be limited by human factors and both false-positive and false-negative errors might occur. By understanding visual search in breast images, we may be able to identify causes of diagnostic errors, find ways to reduce them, and also provide a better education to radiology residents. Many visual search studies in breast radiology have been devoted to mammography. These studies showed that 70% of missed lesions on mammograms attract radiologists' visual attention and that a plethora of different reasons, such as satisfaction of search, incorrect background sampling, and incorrect first impression can cause diagnostic errors in the interpretation of mammograms. Recently, highly accurate tools, which rely on both eye-tracking data and the content of the mammogram, have been proposed to provide feedback to the radiologists. Improving these tools and determining the optimal pathway to integrate them in the radiology workflow could be a possible line of future research. Moreover, in the past few years deep learning has led to improving diagnostic accuracy of computerized diagnostic tools and visual search studies will be required to understand how radiologists interact with the prompts from these tools, and to identify the best way to utilize them. Visual search in other breast imaging modalities, such as breast ultrasound and digital breast tomosynthesis, have so far received less attention, probably due to associated complexities of eye-tracking monitoring and analysing the data. For example, in digital breast tomosynthesis, scrolling through the image results in longer trials, adds a new factor to the study's complexity and makes calculation of gaze parameters more difficult. However, considering the wide utilization of three-dimensional imaging modalities, more visual search studies involving reading stack-view examinations are required in the future. To conclude, in the past few decades visual search studies provided extensive understanding about underlying reasons for diagnostic errors in breast radiology and characterized differences between experts' and novices' visual search patterns. Further visual search studies are required to investigate radiologists' interaction with relatively newer imaging modalities and artificial intelligence tools.

Risk Factors for Perceptual-versus-Interpretative Errors in Diagnostic Neuroradiology

BACKGROUND AND PURPOSE

Diagnostic errors in radiology are classified as perception or interpretation errors. This study determined whether specific conditions differed when perception or interpretation errors occurred during neuroradiology image interpretation.

MATERIALS AND METHODS

In a sample of 254 clinical error cases in diagnostic neuroradiology, we classified errors as perception or interpretation errors, then characterized imaging technique, interpreting radiologist's experience, anatomic location of the abnormality, disease etiology, time of day, and day of the week. Interpretation and perception errors were compared with hours worked per shift, cases read per shift, average cases read per shift hour, and the order of case during the shift when the error occurred.

RESULTS

Perception and interpretation errors were 74.8% (n = 190) and 25.2% (n = 64) of errors, respectively. Logistic regression analyses showed that the odds of an interpretation error were 2 times greater (OR, 2.09; 95% CI, 1.05-4.15; P = .04) for neuroradiology attending physicians with ≤5 years of experience. Interpretation errors were more likely with MR imaging compared with CT (OR, 2.10; 95% CI, 1.09-4.01; P = .03). Infectious/inflammatory/autoimmune diseases were more frequently associated with interpretation errors (P = .04). Perception errors were associated with faster reading rates (6.01 versus 5.03 cases read per hour; P = .004) and occurred later during the shift (24th-versus-18th case; P = .04).

CONCLUSIONS

Among diagnostic neuroradiology error cases, interpretation-versus-perception errors are affected by the neuroradiologist's experience, technique, and the volume and rate of cases read. Recognition of these risk factors may help guide programs for error reduction in clinical neuroradiology services.

A novel framework for MR image segmentation and quantification by using MedGA

BACKGROUND AND OBJECTIVES

Image segmentation represents one of the most challenging issues in medical image analysis to distinguish among different adjacent tissues in a body part. In this context, appropriate image pre-processing tools can improve the result accuracy achieved by computer-assisted segmentation methods. Taking into consideration images with a bimodal intensity distribution, image binarization can be used to classify the input pictorial data into two classes, given a threshold intensity value. Unfortunately, adaptive thresholding techniques for two-class segmentation work properly only for images characterized by bimodal histograms. We aim at overcoming these limitations and automatically determining a suitable optimal threshold for bimodal Magnetic Resonance (MR) images, by designing an intelligent image analysis framework tailored to effectively assist the physicians during their decision-making tasks.

METHODS

In this work, we present a novel evolutionary framework for image enhancement, automatic global thresholding, and segmentation, which is here applied to different clinical scenarios involving bimodal MR image analysis: (i) uterine fibroid segmentation in MR guided Focused Ultrasound Surgery, and (ii) brain metastatic cancer segmentation in neuro-radiosurgery therapy. Our framework exploits MedGA as a pre-processing stage. MedGA is an image enhancement method based on Genetic Algorithms that improves the threshold selection, obtained by the efficient Iterative Optimal Threshold Selection algorithm, between the underlying sub-distributions in a nearly bimodal histogram.

RESULTS

The results achieved by the proposed evolutionary framework were quantitatively evaluated, showing that the use of MedGA as a pre-processing stage outperforms the conventional image enhancement methods (i.e., histogram equalization, bi-histogram equalization, Gamma transformation, and sigmoid transformation), in terms of both MR image enhancement and segmentation evaluation metrics.

CONCLUSIONS

Thanks to this framework, MR image segmentation accuracy is considerably increased, allowing for measurement repeatability in clinical workflows. The proposed computational solution could be well-suited for other clinical contexts requiring MR image analysis and segmentation, aiming at providing useful insights for differential diagnosis and prognosis.

Analysis of Perceptual Expertise in Radiology - Current Knowledge and a New Perspective

Radiologists rely principally on visual inspection to detect, describe, and classify findings in medical images. As most interpretive errors in radiology are perceptual in nature, understanding the path to radiologic expertise during image analysis is essential to educate future generations of radiologists. We review the perceptual tasks and challenges in radiologic diagnosis, discuss models of radiologic image perception, consider the application of perceptual learning methods in medical training, and suggest a new approach to understanding perceptional expertise. Specific principled enhancements to educational practices in radiology promise to deepen perceptual expertise among radiologists with the goal of improving training and reducing medical error.

Eye Movements in Medical Image Perception: A Selective Review of Past, Present and Future

The eye movements of experts, reading medical images, have been studied for many years. Unlike topics such as face perception, medical image perception research needs to cope with substantial, qualitative changes in the stimuli under study due to dramatic advances in medical imaging technology. For example, little is known about how radiologists search through 3D volumes of image data because they simply did not exist when earlier eye tracking studies were performed. Moreover, improvements in the affordability and portability of modern eye trackers make other, new studies practical. Here, we review some uses of eye movements in the study of medical image perception with an emphasis on newer work. We ask how basic research on scene perception relates to studies of medical 'scenes' and we discuss how tracking experts' eyes may provide useful insights for medical education and screening efficiency.

Enhancing Parathyroid Gland Visualization Using a Near Infrared Fluorescence-Based Overlay Imaging System

BACKGROUND

Misidentifying parathyroid glands (PGs) during thyroidectomies or parathyroidectomies could significantly increase postoperative morbidity. Imaging systems based on near infrared autofluorescence (NIRAF) detection can localize PGs with high accuracy. These devices, however, depict NIRAF images on remote display monitors, where images lack spatial context and comparability with actual surgical field of view. In this study, we designed an overlay tissue imaging system (OTIS) that detects tissue NIRAF and back-projects the collected signal as a visible image directly onto the surgical field of view instead of a display monitor, and tested its ability for enhancing parathyroid visualization.

STUDY DESIGN

The OTIS was first calibrated with a fluorescent ink grid and initially tested with parathyroid, thyroid, and lymph node tissues ex vivo. For in vivo measurements, the surgeon's opinion on tissue of interest was first ascertained. After the surgeon looked away, the OTIS back-projected visible green light directly onto the tissue of interest, only if the device detected relatively high NIRAF as observed in PGs. System accuracy was determined by correlating NIRAF projection with surgeon's visual confirmation for in situ PGs or histopathology report for excised PGs.

RESULTS

The OTIS yielded 100% accuracy when tested ex vivo with parathyroid, thyroid, and lymph node specimens. Subsequently, the device was evaluated in 30 patients who underwent thyroidectomy and/or parathyroidectomy. Ninety-seven percent of exposed tissue of interest was visualized correctly as PGs by the OTIS, without requiring display monitors or contrast agents.

CONCLUSIONS

Although OTIS holds novel potential for enhancing label-free parathyroid visualization directly within the surgical field of view, additional device optimization is required for eventual clinical use.

Development and Evaluation of a Novel Point-of-Use Quality Assurance Tool for Digital Pathology

Context.—

Flexible working at diverse or remote sites is a major advantage when reporting using digital pathology, but currently there is no method to validate the clinical diagnostic setting within digital microscopy.

Objective.—

To develop a preliminary Point-of-Use Quality Assurance (POUQA) tool designed specifically to validate the diagnostic setting for digital microscopy.

Design.—

We based the POUQA tool on the red, green, and blue (RGB) values of hematoxylin-eosin. The tool used 144 hematoxylin-eosin–colored, 5×5-cm patches with a superimposed random letter with subtly lighter RGB values from the background color, with differing levels of difficulty. We performed an initial evaluation across 3 phases within 2 pathology departments: 1 in the United Kingdom and 1 in Sweden.

Results.—

In total, 53 experiments were conducted across all phases resulting in 7632 test images viewed in all. Results indicated that the display, the user's visual system, and the environment each independently impacted performance. Performance was improved with reduction in natural light and through use of medical-grade displays.

Conclusions.—

The use of a POUQA tool for digital microscopy is essential to afford flexible working while ensuring patient safety. The color-contrast test provides a standardized method of comparing diagnostic settings for digital microscopy. With further planned development, the color-contrast test may be used to create a “Verified Login” for diagnostic setting validation.

[Measurements errors in screening mammogram interpretation by radiologists]

The timely detection of breast cancer is achieved through mammography; however, the quality of the procedure should be addressed for proper performance and interpretation. Despite recent improvements in quality assurance in mammography, interpretation still depends on each reader; therefore, errors can be made when interpreting screening mammograms, leading to unnecessary biopsies and/or overdiagnosis, with sustained physical, economic and psychological consequences. Since interpretation is related to the perceptive and cognitive ability of the radiologist, it is necessary to have extensive knowledge about the possible errors that may occur during interpretation, as well as of the way how they can be reduced, prevented and/or corrected to provide the patient with the highest possible level of safety.

Chest X-ray evaluation training: impact of normal and abnormal image ratio and instructional sequence

CONTEXT

Medical image perception training generally focuses on abnormalities, whereas normal images are more prevalent in medical practice. Furthermore, instructional sequences that let students practice prior to expert instruction (inductive) may lead to improved performance compared with methods that give students expert instruction before practice (deductive). This study investigates the effects of the proportion of normal images and practice-instruction order on learning to interpret medical images. It is hypothesised that manipulation of the proportion of normal images will lead to a sensitivity-specificity trade-off and that students in practice-first (inductive) conditons need more time per practice case but will correctly identify more test cases.

METHODS

Third-year medical students (n = 103) learned radiograph interpretation by practising cases with, respectively, 30% or 70% normal radiographs prior to expert instruction (practice-first order) or after expert instruction (instruction-first order). After training, students performed a test (60% normal) and sensitivity (% of correctly identified abnormal radiographs), specificity (% of correctly identified normal radiographs), diagnostic performance (% of correct diagnoses) and case duration were measured.

RESULTS

The conditions with 30% of normal images scored higher on sensitivity but the conditions with 70% of normal images scored higher on specificity, indicating a sensitivity and specificity trade-off. Those who participated in inductive conditions took less time per practice case but more per test case. They had similar test sensitivity, but scored lower on test specificity.

CONCLUSIONS

The proportion of normal images impacted the sensitivity-specificity trade-off. This trade-off should be an important consideration for the alignment of training with future practice. Furthermore, the deductive conditions unexpectedly scored higher on specificity when participants took less time per case. An inductive approach did not lead to higher diagnostic performance, possibly because participants might already have relevant prior knowledge. Deductive approaches are therefore advised for the training of advanced learners.

Analysis of the French national evaluation of radiology residents

PURPOSE

In France, a national evaluation is given annually to radiology residents. The aim of this study was to perform both a docimological analysis of the quality of the questionnaire and a statistical analysis of the results.

MATERIALS AND METHODS

This retrospective study, which included French radiology residents from Year 1 to Year 5 of residency, was performed from 2015 to 2017 across 25 medical universities in France. Both qualitative and quantitative docimological analyses were performed as assessed by the Cronbach alpha coefficient, the difficulty of question (PDI), and the coefficient of discrimination (Rir). Results to the questionnaire were compared between years of residency.

RESULTS

The results of the analysis confirmed the quality of the questionnaire (Cronbach alpha coefficient=0.71, mean [PDI=0.40]) though the majority of questions could be answered by memory rather than cognitive ability. The mean Rir was 0.02, indicating that students could not be certified using only the questionnaire. The results measuring resident level of knowledge were moderate, with mean results ranging from 9.2/20 at the first year to 11.3/20 at the fifth year of residency (P<0.001). There were no significant differences in results obtained between the third, fourth, and fifth year of residency but results were significantly different among university hospitals.

CONCLUSION

Even if close interactions exist between learning and pedagogic environment, our results suggest that it may be useful to further develop an evaluation process in relation with pedagogic instructions in order to provide more optimal training.

Volumetric image interpretation in radiology: scroll behavior and cognitive processes

The interpretation of medical images is a primary task for radiologists. Besides two-dimensional (2D) images, current imaging technologies allow for volumetric display of medical images. Whereas current radiology practice increasingly uses volumetric images, the majority of studies on medical image interpretation is conducted on 2D images. The current study aimed to gain deeper insight into the volumetric image interpretation process by examining this process in twenty radiology trainees who all completed four volumetric image cases. Two types of data were obtained concerning scroll behaviors and think-aloud data. Types of scroll behavior concerned oscillations, half runs, full runs, image manipulations, and interruptions. Think-aloud data were coded by a framework of knowledge and skills in radiology including three cognitive processes: perception, analysis, and synthesis. Relating scroll behavior to cognitive processes showed that oscillations and half runs coincided more often with analysis and synthesis than full runs, whereas full runs coincided more often with perception than oscillations and half runs. Interruptions were characterized by synthesis and image manipulations by perception. In addition, we investigated relations between cognitive processes and found an overall bottom-up way of reasoning with dynamic interactions between cognitive processes, especially between perception and analysis. In sum, our results highlight the dynamic interactions between these processes and the grounding of cognitive processes in scroll behavior. It suggests, that the types of scroll behavior are relevant to describe how radiologists interact with and manipulate volumetric images.

256 Shades of gray: uncertainty and diagnostic error in radiology

Radiologists practice in an environment of extraordinarily high uncertainty, which results partly from the high variability of the physical and technical aspects of imaging, partly from the inherent limitations in the diagnostic power of the various imaging modalities, and partly from the complex visual-perceptual and cognitive processes involved in image interpretation. This paper reviews the high level of uncertainty inherent to the process of radiological imaging and image interpretation vis-à-vis the issue of radiological interpretive error, in order to highlight the considerable degree of overlap that exists between these. The scope of radiological error, its many potential causes and various error-reduction strategies in radiology are also reviewed.

Accelerating expertise: Perceptual and adaptive learning technology in medical learning

RATIONALE

Recent advances in the learning sciences offer remarkable potential for improving medical learning and performance. Difficult to teach pattern recognition skills can be systematically accelerated using techniques of perceptual learning (PL). The effectiveness of PL interventions is amplified when they are combined with adaptive learning (AL) technology in perceptual-adaptive learning modules (PALMs).

INNOVATION

Specifically, PALMs incorporate the Adaptive Response Time-based Sequencing (ARTS) system, which leverages learner performance (accuracy and speed) in interactive learning episodes to guide the course of factual, perceptual, or procedural learning, optimize spacing, and lead learners to comprehensive mastery. Here we describe elements and scientific foundations of PL and its embodiment in learning technology. We also consider evidence that AL systems utilizing both accuracy and speed enhance learning efficiency and provide a unified account and potential optimization of spacing effects in learning, as well as supporting accuracy, transfer, and fluency as goals of learning.

RESULTS

To illustrate this process, we review some results of earlier PALMs and present new data from a PALM designed to accelerate and improve diagnosis in electrocardiography.

CONCLUSIONS

Through relatively short training interventions, PALMs produce large and durable improvements in trainees' abilities to accurately and fluently interpret clinical signs and tests, helping to bridge the gap between novice and expert clinicians.

The impact of speed and bias on the cognitive processes of experts and novices in medical image decision-making

Training individuals to make accurate decisions from medical images is a critical component of education in diagnostic pathology. We describe a joint experimental and computational modeling approach to examine the similarities and differences in the cognitive processes of novice participants and experienced participants (pathology residents and pathology faculty) in cancer cell image identification. For this study we collected a bank of hundreds of digital images that were identified by cell type and classified by difficulty by a panel of expert hematopathologists. The key manipulations in our study included examining the speed-accuracy tradeoff as well as the impact of prior expectations on decisions. In addition, our study examined individual differences in decision-making by comparing task performance to domain general visual ability (as measured using the Novel Object Memory Test (NOMT) (Richler et al. Cognition 166:42–55, 2017). Using signal detection theory and the diffusion decision model (DDM), we found many similarities between experts and novices in our task. While experts tended to have better discriminability, the two groups responded similarly to time pressure (i.e., reduced caution under speed instructions in the DDM) and to the introduction of a probabilistic cue (i.e., increased response bias in the DDM). These results have important implications for training in this area as well as using novice participants in research on medical image perception and decision-making.

Renal cancer at unenhanced CT: imaging features, detection rates, and outcomes

PURPOSE

To describe and quantify the rate of detection of renal cancer on unenhanced CT.

METHODS

This retrospective, HIPAA-compliant study was approved by the Institutional Review Board. Electronic health records for all patients who underwent unenhanced abdominal CT at our institution between 2000 and 2005 were reviewed to identify patients subsequently diagnosed with renal cancer during a follow-up period of up to 12 years. Images were reviewed to determine if the cancer was visible at index (first) unenhanced CT and their findings recorded. Original radiology reports were reviewed to determine whether the renal cancer was reported; Fisher's Exact Test compared imaging features of detected and missed cancers. Clinical outcomes including time until diagnosis and stage at diagnosis were used to assess the potential impact of missed cancers.

RESULTS

Of 15,695 patients, 82 (0.52%) were diagnosed with renal cancer. Of these, 43/82 (52%) cancers were retrospectively detectable on index unenhanced CT. Among retrospectively detectable cancers, 63% (27/43) were originally detected and reported on index CT and 37% (16/43) were missed. Size was the only feature associated with detection; 83% (20/24) of cancers > 3.0 cm were detected versus 37% (7/19) of cancers ≤ 3.0 cm (p = 0.0036). Although none of the 16 missed renal cancers developed metastases between index CT and time of diagnosis (median 33.5 months), 4 (25%) progressed in stage.

CONCLUSIONS

Renal cancer was rare in patients undergoing unenhanced abdominal CT. Over one-third of potentially detectable cancers were missed prospectively. However, missed cancers did not metastasize and infrequently progressed in stage before being diagnosed.

Evolutionary image simplification for lung nodule classification with convolutional neural networks

PURPOSE

Understanding decisions of deep learning techniques is important. Especially in the medical field, the reasons for a decision in a classification task are as crucial as the pure classification results. In this article, we propose a new approach to compute relevant parts of a medical image. Knowing the relevant parts makes it easier to understand decisions.

METHODS

In our approach, a convolutional neural network is employed to learn structures of images of lung nodules. Then, an evolutionary algorithm is applied to compute a simplified version of an unknown image based on the learned structures by the convolutional neural network. In the simplified version, irrelevant parts are removed from the original image.

RESULTS

In the results, we show simplified images which allow the observer to focus on the relevant parts. In these images, more than 50% of the pixels are simplified. The simplified pixels do not change the meaning of the images based on the learned structures by the convolutional neural network. An experimental analysis shows the potential of the approach. Besides the examples of simplified images, we analyze the run time development.

CONCLUSIONS

Simplified images make it easier to focus on relevant parts and to find reasons for a decision. The combination of an evolutionary algorithm employing a learned convolutional neural network is well suited for the simplification task. From a research perspective, it is interesting which areas of the images are simplified and which parts are taken as relevant.

New approaches to the analysis of eye movement behaviour across expertise while viewing brain MRIs

Brain tumour detection and diagnosis requires clinicians to inspect and analyse brain magnetic resonance images. Eye-tracking is commonly used to examine observers' gaze behaviour during such medical image interpretation tasks, but analysis of eye movement sequences is limited. We therefore used ScanMatch, a novel technique that compares saccadic eye movement sequences, to examine the effect of expertise and diagnosis on the similarity of scanning patterns. Diagnostic accuracy was also recorded. Thirty-five participants were classified as Novices, Medics and Experts based on their level of expertise. Participants completed two brain tumour detection tasks. The first was a whole-brain task, which consisted of 60 consecutively presented slices from one patient; the second was an independent-slice detection task, which consisted of 32 independent slices from five different patients. Experts displayed the highest accuracy and sensitivity followed by Medics and then Novices in the independent-slice task. Experts showed the highest level of scanning pattern similarity, with medics engaging in the least similar scanning patterns, for both the whole-brain and independent-slice task. In the independent-slice task, scanning patterns were the least similar for false negatives across all expertise levels and most similar for experts when they responded correctly. These results demonstrate the value of using ScanMatch in the medical image perception literature. Future research adopting this tool could, for example, identify cases that yield low scanning similarity and so provide insight into why diagnostic errors occur and ultimately help in training radiologists.