Computer-aided diagnosis as a second reader: spectrum of findings in CT studies of the chest interpreted as normal

Incorporation of CAD (computer-aided detection) with thin-slice lung CT in routine 18F-FDG PET/CT imaging read-out protocol for detection of lung nodules

OBJECTIVE

To evaluate the detection rate and performance of 18F-FDG PET alone (PET), the combination of PET and low-dose thick-slice CT (PET/lCT), PET and diagnostic thin-slice CT (PET/dCT), and additional computer-aided detection (PET/dCT/CAD) for lung nodules (LN)/metastases in tumor patients. Along with this, assessment of inter-reader agreement and time requirement for different techniques were evaluated as well.

METHODS

In 100 tumor patients (56 male, 44 female; age range: 22-93 years, mean age: 60 years) 18F-FDG PET images, low-dose CT with shallow breathing (5 mm slice thickness), and diagnostic thin-slice CT (1 mm slice thickness) in full inspiration were retrospectively evaluated by three readers with variable experience (junior, mid-level, and senior) for the presence of lung nodules/metastases and additionally analyzed with CAD. Time taken for each analysis and number of the nodules detected were assessed. Sensitivity, specificity, positive and negative predictive value, accuracy, and Receiver operating characteristic (ROC) analysis of each technique was calculated. Histopathology and/or imaging follow-up served as reference standard for the diagnosis of metastases.

RESULTS

Three readers, on an average, detected 40 LN in 17 patients with PET only, 121 LN in 37 patients using ICT, 283 LN in 60 patients with dCT, and 282 LN in 53 patients with CAD. On average, CAD detected 49 extra LN, missed by the three readers without CAD, whereas CAD overall missed 53 LN. There was very good inter-reader agreement regarding the diagnosis of metastases for all four techniques (kappa: 0.84-0.93). The average time required for the evaluation of LN in PET, lCT, dCT, and CAD was 25, 31, 60, and 40 s, respectively; the assistance of CAD lead to average 33% reduction in time requirement for evaluation of lung nodules compared to dCT. The time-saving effect was highest in the less experienced reader. Regarding the diagnosis of metastases, sensitivity and specificity combined of all readers were 47.8%/96.2% for PET, 80.0%/81.9% for PET/lCT, 100%/56.7% for PET/dCT, and 95.6%/64.3% for PET/CAD. No significant difference was observed regarding the ROC AUC (area under the curve) between the imaging methods.

CONCLUSION

Implementation of CAD for the detection of lung nodules/metastases in routine 18F-FDG PET/CT read-out is feasible. The combination of diagnostic thin-slice CT and CAD significantly increases the detection rate of lung nodules in tumor patients compared to the standard PET/CT read-out. PET combined with low-dose CT showed the best balance between sensitivity and specificity regarding the diagnosis of metastases per patient. CAD reduces the time required for lung nodule/metastasis detection, especially for less experienced readers.

CNN-based evaluation of bone density improves diagnostic performance to detect osteopenia and osteoporosis in patients with non-contrast chest CT examinations

PURPOSE

As osteoporosis is still underdiagnosed by clinicians and radiologists, the aim of the present study was to assess the performance of an Artificial intelligence (AI)-based Convolutional Neuronal Network (CNN)-Algorithm for the detection of low bone density on routine non-contrast chest CT in comparison to clinical reports using DEXA scans as reference.

METHOD

This retrospective cross-sectional study included patients who underwent non-contrast chest CT and DEXA between April 2018 and June 2018 (n = 109, 19 men, mean age: 67.7 years). CT studies were evaluated for thoracic vertebral bone pathologies using a CNN-Algorithm, which calculates the attenuation profile of the spine. The content of the radiological reports was evaluated for the description of osteoporosis or osteopenia. DEXA was used as the reference standard. To estimate correlation the Spearman test was used and the comparison of the different groups was performed using the Wilcoxon rank sum test. Diagnostic was evaluated by performing a receiver operating characteristic curve analysis.

RESULTS

The DEXA examination revealed normal bone density in 42 patients, while 49 patients had osteopenia and 7 osteoporosis. There was a statistically significant correlation between the mean CNN-based attenuation of the thoracic spine and the bone density measured on the DEXA in the hip (r = 0.51, p < 0.001) and lumbar spine (r = 0.34, p = 0.01). The mean attenuation was significantly higher in patients with normal bone density (172 ± 44.5 HU) compared to those with osteopenia or osteoporosis (125.2 ± 33.8 HU), (p < 0.0001). Diagnostic performance in distinguishing normal from abnormal bone density was higher using the CNN-based vertebral attenuation (accuracy 0.75, sensitivity: 0.93, specificity: 0.61) compared to clinical reports (accuracy 0.51, sensitivity: 0.14, specificity: 0.53).

CONCLUSION

CNN-based evaluation of bone density may provide additional value over standard clinical reports for the detection of osteopenia and osteoporosis in patients undergoing routine non-contrast chest CT scans. This additional value could improve identification of fracture risk and subsequent treatment.

Diagnostic Accuracy and Performance of Artificial Intelligence in Detecting Lung Nodules in Patients With Complex Lung Disease: A Noninferiority Study

OBJECTIVES

The aim of the study is to investigate the performance of artificial intelligence (AI) convolutional neural networks (CNN) in detecting lung nodules on chest computed tomography of patients with complex lung disease, and demonstrate its noninferiority when compared against an experienced radiologist through clinically relevant assessments.

METHODS

A CNN prototype was used to retrospectively evaluate 103 complex lung disease cases and 40 control cases without reported nodules. Computed tomography scans were blindly evaluated by an expert thoracic radiologist; a month after initial analyses, 20 positive cases were re-evaluated with the assistance of AI. For clinically relevant applications: (1) AI was asked to classify each patient into nodules present or absent and (2) AI results were compared against standard radiology reports. Standard statistics were performed to determine detection performance.

RESULTS

AI was, on average, 27 seconds faster than the expert and detected 8.4% of nodules that would have been missed. AI had a sensitivity of 67.7%, similar to an accuracy reported for experienced radiologists. AI correctly classified each patient (nodules present/absent) with a sensitivity of 96.1%. When matched against radiology reports, AI performed with a sensitivity of 89.4%. Control group assessment demonstrated an overall specificity of 82.5%. When aided by AI, the expert decreased the average assessment time per case from 2:44 minutes to 35.7 seconds, while reporting an overall increase in confidence.

CONCLUSION

In a group of patients with complex lung disease, the sensitivity of AI is similar to an experienced radiologist and the tool helps detect previously missed nodules. AI also helps experts analyze for lung nodules faster and more confidently, a feature that is beneficial to patients and favorable to hospitals due to increased patient load and need for shorter turnaround times.

Performance of an Artificial Intelligence-Based Platform Against Clinical Radiology Reports for the Evaluation of Noncontrast Chest CT

RATIONALE AND OBJECTIVES

Research on implementation of artificial intelligence (AI) in radiology workflows and its impact on reports remains scarce. In this study, we aim to assess if an AI platform would perform better than clinical radiology reports in evaluating noncontrast chest computed tomography (CT) scans.

MATERIALS AND METHODS

Consecutive patients who had undergone noncontrast chest CT were retrospectively identified. The radiology reports were reviewed in a binary fashion for reporting of pulmonary lesions, pulmonary emphysema, aortic dilatation, coronary artery calcifications (CAC), and vertebral compression fractures (VCF). CT scans were then processed using an AI platform. The reports' findings and the AI results were subsequently compared to a consensus read by two board-certificated radiologists as reference.

RESULTS

A total of 100 patients (mean age: 64.2 ± 14.8 years; 57% males) were included in this study. Aortic segmentation and calcium quantification failed to be processed by AI in 2 and 3 cases, respectively. AI showed superior diagnostic performance in identifying aortic dilatation (AI: sensitivity: 96.3%, specificity: 81.4%, AUC: 0.89) vs (Reports: sensitivity: 25.9%, specificity: 100%, AUC: 0.63), p <0.001; and CAC (AI: sensitivity: 89.8%, specificity: 100, AUC: 0.95) vs (Reports: sensitivity: 75.4%, specificity: 94.9%, AUC: 0.85), p = 0.005. Reports had better performance than AI in identifying pulmonary lesions (Reports: sensitivity: 97.6%, specificity: 100%, AUC: 0.99) vs (AI: sensitivity: 92.8%, specificity: 82.4%, AUC: 0.88), p = 0.024; and VCF (Reports: sensitivity:100%, specificity: 100%, AUC: 1.0) vs (AI: sensitivity: 100%, specificity: 63.7%, AUC: 0.82), p <0.001. A comparable diagnostic performance was noted in identifying pulmonary emphysema on AI (sensitivity: 80.6%, specificity: 66.7%. AUC: 0.74) and reports (sensitivity: 74.2%, specificity: 97.1%, AUC: 0.86), p = 0.064.

CONCLUSION

Our results demonstrate that incorporating AI support platforms into radiology workflows can provide significant added value to clinical radiology reporting.

Performance and reading time of lung nodule identification on multidetector CT with or without an artificial intelligence-powered computer-aided detection system

AIM

To compare the performance and reading time of different readers using automatic artificial intelligence (AI)-powered computer-aided detection (CAD) to detect lung nodules in different reading modes.

MATERIALS AND METHODS

One hundred and fifty multidetector computed tomography (CT) datasets containing 340 nodules ≤10 mm in diameter were collected retrospectively. A CAD with vessel-suppressed function was used to interpret the images. Three junior and three senior readers were assigned to read (1) CT images without CAD, (2) second-read using CAD in which CAD was applied only after initial unassisted assessment, and (3) a concurrent read with CAD in which CAD was applied at the start of assessment. Diagnostic performances and reading times were compared using analysis of variance.

RESULTS

For all readers, the mean sensitivity improved from 64% (95% confidence interval [CI]: 62%, 66%) for the without-CAD mode to 82% (95% CI: 80%, 84%) for the second-reading mode and to 80% (95% CI: 79%, 82%) for the concurrent-reading mode (p<0.001). There was no significant difference between the two modes in terms of the mean sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) for both junior and senior readers and all readers (p>0.05). The reading time of all readers was significantly shorter for the concurrent-reading mode (124 ± 25 seconds) compared to without CAD (156 ± 34 seconds; p<0.001) and the second-reading mode (197 ± 46 seconds; p<0.001).

CONCLUSION

In CAD for lung nodules at CT, the second-reading mode and concurrent-reading mode may improve detection performance for all readers in both screening and clinical routine practice. Concurrent use of CAD is more efficient for both junior and senior readers.

Evaluation of a deep learning-based computer-aided detection algorithm on chest radiographs: Case-control study

Along with recent developments in deep learning techniques, computer-aided diagnosis (CAD) has been growing rapidly in the medical imaging field. In this work, we evaluate the deep learning-based CAD algorithm (DCAD) for detecting and localizing 3 major thoracic abnormalities visible on chest radiographs (CR) and to compare the performance of physicians with and without the assistance of the algorithm. A subset of 244 subjects (60% abnormal CRs) was evaluated. Abnormal findings included mass/nodules (55%), consolidation (21%), and pneumothorax (24%). Observer performance tests were conducted to assess whether the performance of physicians could be enhanced with the algorithm. The area under the receiver operating characteristic (ROC) curve (AUC) and the area under the jackknife alternative free-response ROC (JAFROC) were measured to evaluate the performance of the algorithm and physicians in image classification and lesion detection, respectively. The AUCs for nodule/mass, consolidation, and pneumothorax were 0.9883, 1.000, and 0.9997, respectively. For the image classification, the overall AUC of the pooled physicians was 0.8679 without DCAD and 0.9112 with DCAD. Regarding lesion detection, the pooled observers exhibited a weighted JAFROC figure of merit (FOM) of 0.8426 without DCAD and 0.9112 with DCAD. DCAD for CRs could enhance physicians' performance in the detection of 3 major thoracic abnormalities.

Breast Cancer Mass Detection in DCE–MRI Using Deep-Learning Features Followed by Discrimination of Infiltrative vs. In Situ Carcinoma through a Machine-Learning Approach

Breast cancer is the leading cause of cancer deaths worldwide in women. This aggressive tumor can be categorized into two main groups—in situ and infiltrative, with the latter being the most common malignant lesions. The current use of magnetic resonance imaging (MRI) was shown to provide the highest sensitivity in the detection and discrimination between benign vs. malignant lesions, when interpreted by expert radiologists. In this article, we present the prototype of a computer-aided detection/diagnosis (CAD) system that could provide valuable assistance to radiologists for discrimination between in situ and infiltrating tumors. The system consists of two main processing levels—(1) localization of possibly tumoral regions of interest (ROIs) through an iterative procedure based on intensity values (ROI Hunter), followed by a deep-feature extraction and classification method for false-positive rejection; and (2) characterization of the selected ROIs and discrimination between in situ and invasive tumor, consisting of Radiomics feature extraction and classification through a machine-learning algorithm. The CAD system was developed and evaluated using a DCE–MRI image database, containing at least one confirmed mass per image, as diagnosed by an expert radiologist. When evaluating the accuracy of the ROI Hunter procedure with respect to the radiologist-drawn boundaries, sensitivity to mass detection was found to be 75%. The AUC of the ROC curve for discrimination between in situ and infiltrative tumors was 0.70.

3D deep learning for detecting pulmonary nodules in CT scans

Objective

To demonstrate and test the validity of a novel deep-learning-based system for the automated detection of pulmonary nodules.

Materials and Methods

The proposed system uses 2 3D deep learning models, 1 for each of the essential tasks of computer-aided nodule detection: candidate generation and false positive reduction. A total of 888 scans from the LIDC-IDRI dataset were used for training and evaluation.

Results

Results for candidate generation on the test data indicated a detection rate of 94.77% with 30.39 false positives per scan, while the test results for false positive reduction exhibited a sensitivity of 94.21% with 1.789 false positives per scan. The overall system detection rate on the test data was 89.29% with 1.789 false positives per scan.

Discussion

An extensive and rigorous validation was conducted to assess the performance of the proposed system. The system demonstrated a novel combination of 3D deep neural network architectures and demonstrates the use of deep learning for both candidate generation and false positive reduction to be evaluated with a substantial test dataset. The results strongly support the ability of deep learning pulmonary nodule detection systems to generalize to unseen data. The source code and trained model weights have been made available.

Conclusion

A novel deep-neural-network-based pulmonary nodule detection system is demonstrated and validated. The results provide comparison of the proposed deep-learning-based system over other similar systems based on performance.

Computer-assisted detection of acute pulmonary embolism at CT pulmonary angiography in children and young adults: a diagnostic performance analysis

Background

To diagnose pulmonary embolism (PE) in children and adults since evaluating tiny pulmonary vasculature beyond segmental level is a challenging and demanding task with thousands of images.

Purpose

To evaluate the effect of computer-assisted detection (CAD) on acute PE on CTPA in children and young adults by readers with varying experience levels.

Material and Methods

Six radiologists were retrospectively divided into three groups according to experience levels and assessed the CTPA studies on a per-emboli basis. All readers identified independently the PE presence, and ranked diagnostic confidence on a 5-point scale with and without CAD. Reading time, sensitivities, specificities, accuracies, positive predictive values (PPVs), and negative predictive values (NPVs) were calculated for each reading.

Results

The sensitivities and NPVs differed significantly in most readers ( P = 0.004, 0.001, 0.010, 0.010, and 0.012 for sensitivities and P = 0.011, 0.003, 0.016, 0.017, and 0.019 for NPVs) except for reader 6 ( P = 0.148 and 0.165, respectively), and the accuracies of all readers differed significantly (all P < 0.05) in peripheral PE (beyond segmental level) detection readings with CAD versus without CAD between two reading methods. The overall time using CAD was longer than those without CAD (76.6 ± 54.4 s vs. 49.4 ± 17.7 s, P = 0.000) for all readers. Significant differences were found for confidence scores in inter-group measurements with CAD ( P = 0.045) and without CAD ( P < 0.001).

Conclusion

At the expense of longer reading time, the use of the CAD algorithms improves sensitivities, NPVs, and the accuracies of readers in peripheral PE detection, especially for readers with a poor level of interpretation experience.

Lung cancer screening: nodule identification and characterization

The accurate identification and characterization of small pulmonary nodules at low-dose CT is an essential requirement for the implementation of effective lung cancer screening. Individual reader detection performance is influenced by nodule characteristics and technical CT parameters but can be improved by training, the application of CT techniques, and by computer-aided techniques. However, the evaluation of nodule detection in lung cancer screening trials differs from the assessment of individual readers as it incorporates multiple readers, their inter-observer variability, reporting thresholds, and reflects the program accuracy in identifying lung cancer. Understanding detection and interpretation errors in screening trials aids in the implementation of lung cancer screening in clinical practice. Indeed, as CT screening moves to ever lower radiation doses, radiologists must be cognisant of new technical challenges in nodule assessment. Screen detected lung cancers demonstrate distinct morphological features from incidentally or symptomatically detected lung cancers. Hence characterization of screen detected nodules requires an awareness of emerging concepts in early lung cancer appearances and their impact on radiological assessment and malignancy prediction models. Ultimately many nodules remain indeterminate, but further imaging evaluation can be appropriate with judicious utilization of contrast enhanced CT or MRI techniques or functional evaluation by PET-CT.

Added value of double reading in diagnostic radiology,a systematic review

Objectives

Double reading in diagnostic radiology can find discrepancies in the original report, but a systematic program of double reading is resource consuming. There are conflicting opinions on the value of double reading. The purpose of the current study was to perform a systematic review on the value of double reading.

Methods

A systematic review was performed to find studies calculating the rate of misses and overcalls with the aim of establishing the added value of double reading by human observers.

Results

The literature search resulted in 1610 hits. After abstract and full-text reading, 46 articles were selected for analysis. The rate of discrepancy varied from 0.4 to 22% depending on study setting. Double reading by a sub-specialist, in general, led to high rates of changed reports.

Conclusions

The systematic review found rather low discrepancy rates. The benefit of double reading must be balanced by the considerable number of working hours a systematic double-reading scheme requires. A more profitable scheme might be to use systematic double reading for selected, high-risk examination types. A second conclusion is that there seems to be a value of sub-specialisation for increased report quality. A consequent implementation of this would have far-reaching organisational effects.

Key Points

• In double reading, two or more radiologists read the same images.

• A systematic literature review was performed.

• The discrepancy rates varied from 0.4 to 22% in various studies.

• Double reading by sub-specialists found high discrepancy rates.

Electronic supplementary material

The online version of this article (10.1007/s13244-018-0599-0) contains supplementary material, which is available to authorised users.

A comparison of axial versus coronal image viewing in computer-aided detection of lung nodules on CT

PURPOSE

To compare primarily viewing axial images (Axial mode) versus coronal reconstruction images (Coronal mode) in computer-aided detection (CAD) of lung nodules on multidetector computed tomography (CT) in terms of detection performance and reading time.

MATERIALS AND METHODS

Sixty CT data sets from two institutions were collected prospectively. Ten observers (6 radiologists, 4 pulmonologists) with varying degrees of experience interpreted the data sets using CAD as a second reader (performing nodule detection first without then with aid). The data sets were interpreted twice, once each for Axial and Coronal modes, in two sessions held 4 weeks apart. Jackknife free-response receiver-operating characteristic analysis was used to compare detection performances in the two modes.

RESULTS

Mean figure-of-merit values with and without aid were 0.717 and 0.684 in Axial mode and 0.702 and 0.671 in Coronal mode; use of CAD significantly increased the performance of observers in both modes (P < 0.01). Mean reading times for radiologists did not significantly differ between Axial (156 ± 74 s) and Coronal mode (164 ± 69 s; P = 0.08). Mean reading times for pulmonologists were significantly lower in Coronal (112 ± 53 s) than in Axial mode (130 ± 80 s; P < 0.01).

CONCLUSION

There was no statistically significant difference between Axial and Coronal modes for lung nodule detection with CAD.

Automatic detection of subsolid pulmonary nodules in thoracic computed tomography images

Subsolid pulmonary nodules occur less often than solid pulmonary nodules, but show a much higher malignancy rate. Therefore, accurate detection of this type of pulmonary nodules is crucial. In this work, a computer-aided detection (CAD) system for subsolid nodules in computed tomography images is presented and evaluated on a large data set from a multi-center lung cancer screening trial. The paper describes the different components of the CAD system and presents experiments to optimize the performance of the proposed CAD system. A rich set of 128 features is defined for subsolid nodule candidates. In addition to previously used intensity, shape and texture features, a novel set of context features is introduced. Experiments show that these features significantly improve the classification performance. Optimization and training of the CAD system is performed on a large training set from one site of a lung cancer screening trial. Performance analysis on an independent test from another site of the trial shows that the proposed system reaches a sensitivity of 80% at an average of only 1.0 false positive detections per scan. A retrospective analysis of the output of the CAD system by an experienced thoracic radiologist shows that the CAD system is able to find subsolid nodules which were not contained in the screening database.

Comparison of sensitivity of lung nodule detection between radiologists and technologists on low-dose CT lung cancer screening images

OBJECTIVES

The objective of this study was to compare the sensitivity of detection of lung nodules on low-dose screening CT images between radiologists and technologists.

METHODS

11 radiologists and 10 technologists read the low-dose screening CT images of 78 subjects. On images with a slice thickness of 5 mm, there were 60 lung nodules that were ≥5 mm in diameter: 26 nodules with pure ground-glass opacity (GGO), 7 nodules with mixed ground-glass opacity (GGO with a solid component) and 27 solid nodules. On images with a slice thickness of 2 mm, 69 lung nodules were ≥5 mm in diameter: 35 pure GGOs, 7 mixed GGOs and 27 solid nodules. The 21 observers read screening CT images of 5-mm slice thickness at first; then, 6 months later, they read screening CT images of 2-mm slice thickness from the 78 subjects.

RESULTS

The differences in the mean sensitivities of detection of the pure GGOs, mixed GGOs and solid nodules between radiologists and technologists were not statistically significant, except for the case of solid nodules; the p-values of the differences for pure GGOs, mixed GGOs and solid nodules on the CT images with 5-mm slice thickness were 0.095, 0.461 and 0.005, respectively, and the corresponding p-values on CT images of 2-mm slice thickness were 0.971, 0.722 and 0.0037, respectively.

CONCLUSION

Well-trained technologists may contribute to the detection of pure and mixed GGOs ≥5 mm in diameter on low-dose screening CT images.

Consensus versus disagreement in imaging research: a case study using the LIDC database

Traditionally, image studies evaluating the effectiveness of computer-aided diagnosis (CAD) use a single label from a medical expert compared with a single label produced by CAD. The purpose of this research is to present a CAD system based on Belief Decision Tree classification algorithm, capable of learning from probabilistic input (based on intra-reader variability) and providing probabilistic output. We compared our approach against a traditional decision tree approach with respect to a traditional performance metric (accuracy) and a probabilistic one (area under the distance-threshold curve-AuC(dt)). The probabilistic classification technique showed notable performance improvement in comparison with the traditional one with respect to both evaluation metrics. Specifically, when applying cross-validation technique on the training subset of instances, boosts of 28.26% and 30.28% were noted for the probabilistic approach with respect to accuracy and AuC(dt), respectively. Furthermore, on the validation subset of instances, boosts of 20.64% and 23.21% were noted again for the probabilistic approach with respect to the same two metrics. In addition, we compared our CAD system results with diagnostic data available for a small subset of the Lung Image Database Consortium database. We discovered that when our CAD system errs, it generally does so with low confidence. Predictions produced by the system also agree with diagnoses of truly benign nodules more often than radiologists, offering the possibility of reducing the false positives.

Performance of computer-aided detection of pulmonary nodules in low-dose CT: comparison with double reading by nodule volume

Objective

To evaluate performance of computer-aided detection (CAD) beyond double reading for pulmonary nodules on low-dose computed tomography (CT) by nodule volume.

Methods

A total of 400 low-dose chest CT examinations were randomly selected from the NELSON lung cancer screening trial. CTs were evaluated by two independent readers and processed by CAD. A total of 1,667 findings marked by readers and/or CAD were evaluated by a consensus panel of expert chest radiologists. Performance was evaluated by calculating sensitivity of pulmonary nodule detection and number of false positives, by nodule characteristics and volume.

Results

According to the screening protocol, 90.9 % of the findings could be excluded from further evaluation, 49.2 % being small nodules (less than 50 mm³). Excluding small nodules reduced false-positive detections by CAD from 3.7 to 1.9 per examination. Of 151 findings that needed further evaluation, 33 (21.9 %) were detected by CAD only, one of them being diagnosed as lung cancer the following year. The sensitivity of nodule detection was 78.1 % for double reading and 96.7 % for CAD. A total of 69.7 % of nodules undetected by readers were attached nodules of which 78.3 % were vessel-attached.

Conclusions

CAD is valuable in lung cancer screening to improve sensitivity of pulmonary nodule detection beyond double reading, at a low false-positive rate when excluding small nodules.

Key Points

• Computer-aided detection (CAD) has known advantages for computed tomography (CT).

• Combined CAD/nodule size cut-off parameters assist CT lung cancer screening.

• This combination improves the sensitivity of pulmonary nodule detection by CT.

• It increases the positive predictive value for cancer detection.

Can computer assisted diagnosis (CAD) be used as a screening tool in the detection of pulmonary nodules when using 64-slice multidetector computed tomography?

OBJECTIVES

To evaluate (1) whether or not the addition of computer-assisted diagnosis (CAD) to 64-slice multidetector computed tomography (CT) can be used as a screening tool for detection of pulmonary nodules in routine CT chest examinations and (2) whether or not to advocate the incorporation of CAD as a screening tool into our daily practice.

MATERIALS AND METHODS

A retrospective cross-sectional analysis of 109 consecutive patients who had all undergone routine contrast-enhanced CT chest examinations for indications other than lung cancer at the Radiology Department of Aga Khan University Hospital, Karachi, between November 2010 and January 2011. All examinations were evaluated in terms of the detection of pulmonary nodules by a consultant radiologist and CAD (ImageChecker CT Algorithm R2 Technology) software. The ability of CAD software to detect pulmonary nodules was evaluated against the reference standard. In addition, a chest radiologist also calculated the number of pulmonary nodules. The sensitivity and specificity of the CAD software were calculated against the reference standard by using a 2 × 2 table. The Mann-Whitney U test was applied to compare the performances of CAD and the radiologist.

RESULTS

CAD detected 610 pulmonary nodules while the radiologist detected only 113. The reference standard declared 198 pulmonary nodules to be true nodules. CAD detected 95% of all true nodules (189/198), whereas the radiologist detected only 57% (113/198). In the detection of true pulmonary nodules, CAD had 98% sensitivity compared with the radiologist who had 57% sensitivity; the statistical difference between their performances had a P value <0.001.

CONCLUSION

Considering the high sensitivity of CAD to detect nearly all true pulmonary nodules, we advocate its application as a screening tool in all CT chest examinations for the early detection of pulmonary nodules and lung carcinoma.

Evaluation of a method of computer-aided detection (CAD) of pulmonary nodules with computed tomography

PURPOSE

The authors sought to compare the sensitivity and reading time obtained using computer-aided detection (CAD) software as second reader (SR) or concurrent reader (CR) in the identification of pulmonary nodules.

MATERIALS AND METHODS

Unenhanced CT scans of 100 consecutive cancer patients were retrospectively reviewed by four readers to identify all solid, noncalcified pulmonary nodules ranging from 3 to 30 mm in diameter. The sensitivity and reading time of each reader and of CAD alone were calculated at 3-mm and 5-mm thresholds with respect to the reference standard, consisting of a consensus reading by the four radiologists involved in the study. The McNemar test was used to compare the sensitivities obtained by reading without CAD (readers 1 and 2), with CAD as SR (readers 1 and 2 with a 2-month delay), and with CAD as CR (readers 3 and 4). The paired Student's t test was used to compare reading times. A value of p<0.05 was considered statistically significant.

RESULTS

A total of 258 and 224 nodules were identified at 3-mm and 5-mm thresholds, respectively. The sensitivity of CAD alone was 62.79% and 67.41% at the 3-mm and 5-mm threshold values respectively, with 4.15 and 2.96 false-positive findings per examination. CAD as SR produced a significant increase in sensitivity (p<0.001) in nodule detection with respect to reading without CAD both at 3 mm (12.01%) and 5 mm (10.04%); the average increase in sensitivity obtained when comparing CAD as SR to CAD as CR was statistically significant (p<0.025) both at the 3-mm (5.35%) and 5-mm (4.68%) thresholds. CAD as CR produced a nonsignificant increase in sensitivity compared with reading without CAD (p>0.05). Mean reading time using CAD as SR (330 s) was significantly longer than reading without CAD (135 s, p<0.001) and reading with CAD as CR (195 s, p<0.025).

CONCLUSIONS

The use of CAD as CR, without any significant increase in reading time, produces no significant increase in sensitivity in pulmonary nodule detection when compared with reading without CAD (p>0.05); CAD as SR, at the cost of longer reading times, increases sensitivity when compared with reading without CAD (p<0.001) or with CAD as CR (p<0.025).

Methodology for automatic detection of lung nodules in computerized tomography images

Lung cancer is a disease with significant prevalence in several countries around the world. Its difficult treatment and rapid progression make the mortality rates among people affected by this illness to be very high. Aiming to offer a computational alternative for helping in detection of nodules, serving as a second opinion to the specialists, this work proposes a totally automatic methodology based on successive detection refining stages. The automated lung nodules detection scheme consists of six stages: thorax extraction, lung extraction, lung reconstruction, structures extraction, tubular structures elimination, and false positive reduction. In the thorax extraction stage all the artifacts external to the patient's body are discarded. Lung extraction stage is responsible for the identification of the lung parenchyma. The objective of the lung reconstruction stage is to prevent incorrect elimination of portions belonging to the parenchyma. Structures extraction stage comprises the selection of dense structures from inside the lung parenchyma. The next stage, tubular structures elimination eliminates a great part of the pulmonary trees. Finally, the false positive stage selects only structures with great probability to be nodule. Each of the several stages has very specific objectives in detection of particular cases of lung nodules, ensuring good matching rates even in difficult detection situations. We use 33 exams with diversified diagnosis and slices numbers for validating the methodology. We obtained a false positive per exam rate of 0.42 and false negative rate of 0.15. The total classification sensitivity obtained, measured out of the nodule candidates, was 84.84%. The specificity achieved was 96.15% and the total accuracy of the method was 95.21%.

Lung nodule computer-aided detection as a second reader: influence on radiology residents

OBJECTIVE

The purpose of this study was to evaluate the use of a computed tomographic lung nodule computer-aided detection (CAD) software as a second reader for radiology residents.

METHODS

The study involved 110 cases from 4 sites. Three expert radiologists identified nodules that were 4 to 30 mm in maximum diameter to form the ground truth. These cases were then interpreted by 6 board-certified radiologists and 6 radiology residents. The residents read each case without and then with a CAD software (Lung Nodule Assesment, Extended Brilliance Workspace; Philips Healthcare, Highlands Heights, OH) to identify nodules that were 4 to 30 mm in maximum diameter.

RESULTS

The experts identified 91 nodules as the ground truth for the study. The mean sensitivity of the 6 board-certified radiologists was 89%. The mean sensitivity of the residents was 85% without the CAD and 90% (P < 0.05) with the CAD as a second reader.

CONCLUSIONS

The CAD software can help improve the sensitivity of residents in the detection of pulmonary nodules on computed tomography, making them comparable with board-certified radiologists.

Computer-aided detection for the identification of pulmonary nodules in pediatric oncology patients: initial experience

BACKGROUND

Computer-aided detection (CAD) has been shown to increase the sensitivity for detection of pulmonary nodules in adults. This study reports initial findings utilizing a CAD system for the detection of pediatric pulmonary nodules.

OBJECTIVE

To assess the performance of CAD and pediatric radiologists in the detection of pediatric pulmonary nodules.

MATERIALS AND METHODS

CT scans from a series of pediatric patients with known primary tumors and lung nodules were analyzed by four radiologists and a commercially available CAD system. IRB approval was obtained. Sensitivities were calculated for detection according to nodule size and location.

RESULTS

In 24 children (age 3-18 years) 173 nodules were identified. Overall the sensitivity of CAD was 34%, but the sensitivity of CAD for detection of nodules 4.0 mm or larger was 80%. Overall radiologist sensitivity ranged from 68% to 79%. There were 0.9 CAD false-positives and 0.3-2.4 radiologist false-positives per study.

CONCLUSION

CAD in our pediatric oncology patients had good sensitivity for detection of lung nodules 4 mm and larger with a low number of false-positives. However, the sensitivity was considerably less for nodules smaller than 4 mm.

The use of computer-aided detection for the assessment of pulmonary arterial filling defects at computed tomographic angiography

PURPOSE

To validate a computer-aided detection (CAD) tool for the detection of pulmonary arterial filling defects at computed tomographic pulmonary angiography (CTPA) and to assess its benefit for readers of different levels of experience.

METHODS

One hundred consecutive CTPA studies were retrospectively evaluated by a chest radiologist for presence of emboli, serving as the reference standard. Subsequently, examinations were analyzed using commercially available second-generation CAD software (ImageChecker CT, version 2.1; R2 Technology, Inc., Sunnyvale, Calif). The staff radiologist assessed all CAD marks and classified them as true positive or false positive (FP), and any unmarked emboli were classified as false negative. Computer-aided detection software was also evaluated on a case basis compared with the reference standard.For the second part of the study, the 100 CTPAs were reviewed by 3 additional readers of different levels of experience, both without and with CAD, and findings correlated with the reference standard.

RESULTS

Twenty-one studies (21%) were positive for pulmonary embolism. Of these, 18 were true positive on a case basis, and 3 were false negative. Of the 79 negative studies, 16 were true negative with no CAD marks, and the remaining 63 were FP. On a case basis, CAD sensitivity was 86%, specificity was 20%, negative predictive value was 84%, and positive predictive value (PPV) was 22%.Overall, the CAD software yielded 318 marks, identifying 64 of 93 emboli with an additional 254 FP marks. On a mark basis, sensitivity was 69%, and PPV was 20%.Computer-aided detection did not influence the most experienced reader (a chest fellow). Although CAD improved the subjective confidence of the second-year resident in some cases, it had no influence on overall interpretation or accuracy. Computer-aided detection improved accuracy only for the most inexperienced reader, helping this reader to identify 9 emboli not initially appreciated.

CONCLUSIONS

Computer-aided detection specificity and PPV are poor due to expected FP marks, although, often, these can be easily dismissed. However, CAD software may play an important role as a second reader for residents or inexperienced readers.

Anniversary paper: History and status of CAD and quantitative image analysis: the role of Medical Physics and AAPM

The roles of physicists in medical imaging have expanded over the years, from the study of imaging systems (sources and detectors) and dose to the assessment of image quality and perception, the development of image processing techniques, and the development of image analysis methods to assist in detection and diagnosis. The latter is a natural extension of medical physicists' goals in developing imaging techniques to help physicians acquire diagnostic information and improve clinical decisions. Studies indicate that radiologists do not detect all abnormalities on images that are visible on retrospective review, and they do not always correctly characterize abnormalities that are found. Since the 1950s, the potential use of computers had been considered for analysis of radiographic abnormalities. In the mid-1980s, however, medical physicists and radiologists began major research efforts for computer-aided detection or computer-aided diagnosis (CAD), that is, using the computer output as an aid to radiologists-as opposed to a completely automatic computer interpretation-focusing initially on methods for the detection of lesions on chest radiographs and mammograms. Since then, extensive investigations of computerized image analysis for detection or diagnosis of abnormalities in a variety of 2D and 3D medical images have been conducted. The growth of CAD over the past 20 years has been tremendous-from the early days of time-consuming film digitization and CPU-intensive computations on a limited number of cases to its current status in which developed CAD approaches are evaluated rigorously on large clinically relevant databases. CAD research by medical physicists includes many aspects-collecting relevant normal and pathological cases; developing computer algorithms appropriate for the medical interpretation task including those for segmentation, feature extraction, and classifier design; developing methodology for assessing CAD performance; validating the algorithms using appropriate cases to measure performance and robustness; conducting observer studies with which to evaluate radiologists in the diagnostic task without and with the use of the computer aid; and ultimately assessing performance with a clinical trial. Medical physicists also have an important role in quantitative imaging, by validating the quantitative integrity of scanners and developing imaging techniques, and image analysis tools that extract quantitative data in a more accurate and automated fashion. As imaging systems become more complex and the need for better quantitative information from images grows, the future includes the combined research efforts from physicists working in CAD with those working on quantitative imaging systems to readily yield information on morphology, function, molecular structure, and more-from animal imaging research to clinical patient care. A historical review of CAD and a discussion of challenges for the future are presented here, along with the extension to quantitative image analysis.

Diagnosis and management of solitary pulmonary nodules

The advent of computed tomography (CT) screening with or without the help of computer-aided detection systems has increased the detection rate of solitary pulmonary nodules (SPNs), including that of early peripheral lung cancer. Helical dynamic (HD)CT, providing the information on morphologic and hemodynamic characteristics with high specificity and reasonably high accuracy, can be used for the initial assessment of SPNs. (18)F-fluorodeoxyglucose PET/CT is more sensitive at detecting malignancy than HDCT. Therefore, PET/CT may be selectively performed to characterize SPNs when HDCT gives an inconclusive diagnosis. Serial volume measurements are currently the most reliable methods for the tissue characterization of subcentimeter nodules. When malignant nodule is highly suspected for subcentimeter nodules, video-assisted thoracoscopic surgery nodule removal after nodule localization using the pulmonary nodule-marker system may be performed for diagnosis and treatment.

Computer-aided diagnosis in lung nodule assessment

Computed tomography (CT) imaging is playing an increasingly important role in cancer detection, diagnosis, and lesion characterization, and it is the most sensitive test for lung nodule detection. Interpretation of lung nodules involves characterization and integration of clinical and other imaging information. Advances in lung nodule management using CT require optimization of CT data acquisition, postprocessing tools, and computer-aided diagnosis (CAD). The goal of CAD systems being developed is to both assist radiologists in the more sensitive detection of nodules and noninvasively differentiate benign from malignant lesions; the latter is important given that malignant lesions account for between 1% and 11% of pulmonary nodules. The aim of this review is to summarize the current state of the art regarding CAD techniques for the detection and characterization of solitary pulmonary nodules and their potential applications in the clinical workup of these lesions.

Computer-aided diagnosis of lung cancer and pulmonary embolism in computed tomography-a review

Computer-aided detection (CADe) and computer-aided diagnosis (CADx) have been important areas of research in the last two decades. Significant progress has been made in the area of breast cancer detection, and CAD techniques are being developed in many other areas. Recent advances in multidetector row computed tomography have made it an increasingly common modality for imaging of lung diseases. A thoracic examination using thin-section computed tomography contains hundreds of images. Detection of lung cancer and pulmonary embolism on computed tomographic (CT) examinations are demanding tasks for radiologists because they have to search for abnormalities in a large number of images, and the lesions can be subtle. If successfully developed, CAD can be a useful second opinion to radiologists in thoracic CT interpretation. In this review, we summarize the studies that have been reported in these areas, discuss some challenges in the development of CAD, and identify areas that deserve particular attention in future research.

Lung nodule CAD software as a second reader: a multicenter study

RATIONALE AND OBJECTIVES

The purpose of this multicenter, multireader study was to evaluate the performance of computed tomography (CT) lung nodule computer-aided detection (CAD) software as a second reader.

METHODS AND MATERIALS

The study involved 109 patients from four sites. The data were collected from a variety of multidetector CT scanners and had different scan parameters. Each chest CT scan was divided into four quadrants. A group of three expert thoracic radiologists identified nodules between 4 and 30 mm in maximum diameter within each quadrant. The standard of reference was established by a consensus read of these experienced radiologists. The cases were then interpreted by 10 other radiologist readers with varying degrees of experience, without and then with CAD software. These readers identified nodules and assigned an actionability rating to each quadrant before and after using CAD software. Receiver operating characteristic curves were used to measure the performance of the readers without and with CAD software.

RESULTS

The average increase in area under the curve for the 10 readers with CAD software was 1.9% for a 95% confidence interval (0.8-8.0%). The area under the curve without CAD software was 86.7% and with CAD software was 88.7%. A nonsignificant correlation was observed between the improvement in sensitivity and experience of the radiologists. The readers also showed a greater improvement in patients with cancer as compared to those without cancer.

CONCLUSIONS

In this multicenter trial, CAD software was shown to be effective as a second reader by improving the sensitivity of the radiologists in detecting pulmonary nodules.

Errors in cancer diagnosis: current understanding and future directions

PURPOSE

Errors in cancer diagnosis are likely the most harmful and expensive types of diagnostic errors. We reviewed the literature to understand the prevalence, origins, and prevention of errors in cancer diagnosis, focusing on common cancers for which early diagnosis offers clear benefit (melanoma and cancers of the breast, colon, and lung).

METHODS

We searched the Cochrane Library and PubMed from 1966 until April 2007 for publications that met our review criteria and manually searched references of key publications. Our search yielded 110 studies, of which nine were prospective studies and the remaining were retrospective studies.

RESULTS

Errors in cancer diagnosis were not uncommon in autopsy studies and were associated with significant harm and expense in malpractice claims. Literature on prevalence was scant. For each type of cancer, we classified preventable errors according to their origins in patient-physician encounters in the clinic setting, diagnostic test or procedure performance, pathologic confirmation of diagnosis, follow-up of patient or test result, or patient-related delays.

CONCLUSION

The literature reflects advanced knowledge of contributory factors and prevention for diagnostic errors related to the performance of procedures and imaging tests and emerging understanding of pathology errors. However, prospective studies are few, as are studies of diagnostic errors arising from the clinical encounter and patient follow-up. Future research should examine further the system and cognitive problems that lead to the many contributory factors we identified, and address interdisciplinary interventions to prevent errors in cancer diagnosis.

Computer-aided detection of solid lung nodules on follow-up MDCT screening: evaluation of detection, tracking, and reading time

OBJECTIVE

The purpose of this article is to assess detection, tracking, and reading time of solid lung nodules > or = 4 mm on pairs of MDCT chest screening examinations using a computer-aided detection (CAD) system.

MATERIALS AND METHODS

Of 54 pairs of low-dose MDCT chest examinations (1.25-mm collimation), two chest radiologists in consensus established that 25 examinations contained 52 nodules > or = 4 mm. All paired examinations were interpreted on the CAD workstation--first without and then with CAD input--for the detection and tracking of lung nodules. A subset of 33 examination pairs was later read on the clinical workstation used in daily practice, and the results were compared for reading time with those on the CAD workstation.

RESULTS

After CAD input, the sensitivity for nodule detection increased statistically significantly for both readers (9.6% and 23%; p < or = 0.025). One cancer initially missed by one radiologist was correctly identified with CAD input. The overall reading time on the CAD workstation and clinical workstation was comparable for both radiologists. On average, readers spent 4-5 minutes per case to read the paired examinations on the CAD workstation and 6-8 seconds per CAD mark. The CAD system successfully matched 91.3% of nodules detected in both examinations. The overall rate of available CAD growth assessment was 54.9% of all nodule pairs.

CONCLUSION

In the context of temporal comparison of MDCT screening examinations, the sensitivity of radiologists for detecting lung nodules > or = 4 mm increased significantly (p < or = 0.025) with CAD input without compromising reading time.

Computer-aided detection of pulmonary nodules in computed tomography: analysis and review of the literature

PURPOSE

To evaluate diagnostic sensitivity of the pulmonary nodules computer-aided detection (CAD) in computed tomography. To analyze parameters that modify CAD performance. We made a critical analysis of the literature, and we described CAD sensitivity. Moreover, we compared CAD and CAD plus radiologist sensitivity in detection of pulmonary nodules, and we compared different acquisition techniques (thin slice vs thick slice and low dose vs normal dose).

MATERIALS AND METHODS

We used as major data sources the medical literature database of PubMed and MEDLINE, where we searched for articles in English language published from January 2001 to November 2006. We included studies that used spiral or multidetector row CT for CAD.

RESULTS

Twenty studies met the inclusion criteria containing a total of more than 827 patients and 2717 pulmonary nodules detected by CAD. We observed an overall sensitivity of 79% for the CAD and of 92% for CAD plus radiologist; CAD sensitivity was 80% and 74% for thin slice and thick slice protocols, respectively.

CONCLUSIONS

Results of our study suggest that CAD technique is an accurate tool in detection of pulmonary nodules, by working as useful second look for the physician. Sensitivity becomes higher by using it together with radiologist. Actually, the main limitation about the use of CAD to be solved is represented by the persistent high false-positive rate.

[Computer-aided detection of lung nodules on thin collimation MDCT: impact on radiologists' performance]

OBJECTIVES

Evaluate the improvement in detecting lung nodules when using multidetector CT (MDCT) computer-assisted diagnosis (CAD).

MATERIAL AND METHODS

Three radiologists (R1, R2, R3) with different levels of experience independently interpreted 30 MDCT examinations of the thorax taken for screening purposes, first without and then with CAD. The diagnosis was established by two of the three radiologists interpreting the images together, assisted by the CAD.

RESULTS

The consensus reading identified 133 nodules, 61 (46%) of which were 4 mm or larger. The sensitivity values in the detection of nodules before and after using the CAD were 54% and 80% (R1), 38% and 71% (R2), and 70% and 88% (R3), respectively. When considering only the nodules that were 4 mm or larger, the sensitivity values varied before and after using the CAD, from 62% to 95% (R1), from 41% to 84% (R2), and from 74% to 92% (R3). By combining two by two the three radiologists' results obtained without the CAD, the sensitivity values were 65%, 83%, and 77%, respectively, for all the nodules, and 70%, 85%, and 77% for the nodules that were 4 mm or larger. The CAD induced a total of 105 false-positive results, with a mean of 3.5 per examination.

CONCLUSION

The lung nodules missed by the radiologist can be detected if the CAD is used as a second reader. The CAD can be at least as beneficial as the use of a second independent reader.

GridCAD: Grid-based Computer-aided Detection System

Grid computing-the use of a distributed network of electronic resources to cooperatively perform subsets of computationally intensive tasks-may help improve the speed and accuracy of radiologic image interpretation by enabling collaborative computer-based and human readings. GridCAD, a software application developed by using the National Cancer Institute Cancer Biomedical Informatics Grid architecture, implements the fundamental elements of grid computing and demonstrates the potential benefits of grid technology for medical imaging. It allows users to query local and remote image databases, view images, and simultaneously run multiple computer-assisted detection (CAD) algorithms on the images selected. The prototype CAD systems that are incorporated in the software application are designed for the detection of lung nodules on thoracic computed tomographic images. GridCAD displays the original full-resolution images with an overlay of nodule candidates detected by the CAD algorithms, by human observers, or by a combination of both types of readers. With an underlying framework that is computer platform independent and scalable to the task, the software application can support local and long-distance collaboration in both research and clinical practice through the efficient, secure, and reliable sharing of resources for image data mining, analysis, and archiving.

Partnerships in oncology and radiology: the role of radiology in the detection, staging, and follow-up of lung cancer

In this review, I examine the multifaceted role of radiology in the diagnosis, staging, and management of lung cancer, highlighting new applications and modalities such as computer-aided detection of lung nodules and positron emission tomography/computed tomography for staging and monitoring response to therapy. Lung cancer screening is also discussed.

Computer-aided detection of solid lung nodules in lossy compressed multidetector computed tomography chest exams

RATIONALE AND OBJECTIVES

To assess the effect of three-dimensional (3D) lossy image compression of multidetector computed tomography chest scans on computer-aided detection (CAD) of solid lung nodules greater than 4 mm in size.

MATERIALS AND METHODS

A total of 120 cases, acquired with 1.25-mm collimation, were collected from 5 different sites, of which 66/120 were low-dose cases. Two chest radiologists established that 37 cases had no actionable lung nodules; the remaining 83 cases contained 169 nodules (range 3.8-35.0 mm, mean 5.8 mm +/- 3.0 [SD]). All cases were compressed using the 3D Set Partitioning in Hierarchical Trees algorithm to 24:1, 48:1, and 96:1 levels. A study of the effect of compression on computer-aided detection (CAD) sensitivity was performed at operating points of 2.5 false marks (FM), 5 FM, and 10 FM per case using McNemar's test. Logistic regression models were used to evaluate the impact on CAD sensitivity by compression level on nodule and image characteristics.

RESULTS

Compared with no compression, there was no significant degradation in CAD sensitivity found at any of the studied compression levels and operating points. However, between compression levels, there was marginal association with sensitivity. Specifically, 24:1 level was significantly better than 96:1 at all operating points, and occasionally better than no compression at 10 FM/case. Based on multivariate analysis, nodule location was found to be a significant predictor (P = .01) with a lower sensitivity associated with juxtapleural nodules. Nodule size, dose, reconstruction filter, and contrast medium were not significant predictors.

CONCLUSION

CAD detection performance of solid lung nodules did not suffer until 48:1 compression.