Computer-aided detection in screening CT for pulmonary nodules

Performance Analysis in Children of Traditional and Deep Learning CT Lung Nodule Computer-Aided Detection Systems Trained on Adults

BACKGROUND. Although primary lung cancer is rare in children, chest CT is commonly performed to assess for lung metastases in children with cancer. Lung nodule computer-aided detection (CAD) systems have been designed and studied primarily using adult training data, and the efficacy of such systems when applied to pediatric patients is poorly understood. OBJECTIVE. The purpose of this study was to evaluate in children the diagnostic performance of traditional and deep learning CAD systems trained with adult data for the detection of lung nodules on chest CT scans and to compare the ability of such systems to generalize to children versus to other adults. METHODS. This retrospective study included pediatric and adult chest CT test sets. The pediatric test set comprised 59 CT scans in 59 patients (30 boys, 29 girls; mean age, 13.1 years; age range, 4-17 years), which were obtained from November 30, 2018, to August 31, 2020; lung nodules were annotated by fellowship-trained pediatric radiologists as the reference standard. The adult test set was the publicly available adult Lung Nodule Analysis (LUNA) 2016 subset 0, which contained 89 deidentified scans with previously annotated nodules. The test sets were processed through the traditional FlyerScan (github.com/rhardie1/FlyerScanCT) and deep learning Medical Open Network for Artificial Intelligence (MONAI; github.com/Project-MONAI/model-zoo/releases) lung nodule CAD systems, which had been trained on separate sets of CT scans in adults. Sensitivity and false-positive (FP) frequency were calculated for nodules measuring 3-30 mm; nonoverlapping 95% CIs indicated significant differences. RESULTS. Operating at two FPs per scan, on pediatric testing data FlyerScan and MONAI showed significantly lower detection sensitivities of 68.4% (197/288; 95% CI, 65.1-73.0%) and 53.1% (153/288; 95% CI, 46.7-58.4%), respectively, than on adult LUNA 2016 subset 0 testing data (83.9% [94/112; 95% CI, 79.1-88.0%] and 95.5% [107/112; 95% CI, 90.0-98.4%], respectively). Mean nodule size was smaller (p < .001) in the pediatric testing data (5.4 ± 3.1 [SD] mm) than in the adult LUNA 2016 subset 0 testing data (11.0 ± 6.2 mm). CONCLUSION. Adult-trained traditional and deep learning-based lung nodule CAD systems had significantly lower sensitivity for detection on pediatric data than on adult data at a matching FP frequency. The performance difference may relate to the smaller size of pediatric lung nodules. CLINICAL IMPACT. The results indicate a need for pediatric-specific lung nodule CAD systems trained on data specific to pediatric patients.

Diagnostic performance of artificial intelligence for pediatric pulmonary nodule detection on chest computed tomography: comparison of simulated lower radiation doses

The combination of low dose CT and AI performance in the pediatric population has not been explored. Understanding this relationship is relevant for pediatric patients given the potential radiation risks. Here, the objective was to determine the diagnostic performance of commercially available Computer Aided Detection (CAD) for pulmonary nodules in pediatric patients at simulated lower radiation doses. Retrospective chart review of 30 sequential patients between 12-18 years old who underwent a chest CT on the Siemens SOMATOM Force from December 20, 2021, to April 12, 2022. Simulated lower doses at 75%, 50%, and 25% were reconstructed in lung kernel at 3 mm slice thickness using ReconCT and imported to Syngo CT Lung CAD software for analysis. Two pediatric radiologists reviewed the full dose CTs to determine the reference read. Two other pediatric radiologists compared the Lung CAD results at 100% dose and each simulated lower dose level to the reference on a nodule by nodule basis. The sensitivity (Sn), positive predictive value (PPV), and McNemar test were used for comparison of Lung CAD performance based on dose. As reference standard, 109 nodules were identified by the two radiologists. At 100%, and simulated 75%, 50%, and 25% doses, lung CAD detected 60, 62, 58, and 62 nodules, respectively; 28, 28, 29, and 26 were true positive (Sn = 26%, 26%, 27%, 24%), 30, 32, 27, and 34 were false positive (PPV = 48%, 47%, 52%, 43%). No statistically significance difference of Lung CAD performance at different doses was found, with p-values of 1.0, 1.0, and 0.7 at simulated 75%, 50%, and 25% doses compared to standard dose.

CONCLUSION

The Lung CAD shows low sensitivity at all simulated lower doses for the detection of pulmonary nodules in this pediatric population. However, radiation dose may be reduced from standard without further compromise to the Lung CAD performance.

WHAT IS KNOWN

• High diagnostic performance of Lung CAD for detection of pulmonary nodules in adults. • Several imaging techniques are applied to reduce pediatric radiation dose.

WHAT IS NEW

• Low sensitivity at all simulated lower doses for the detection of pulmonary nodules in our pediatric population. • Radiation dose may be reduced from standard without further compromise to the Lung CAD performance.

"Will I change nodule management recommendations if I change my CAD system?"-impact of volumetric deviation between different CAD systems on lesion management

OBJECTIVES

To evaluate and compare the measurement accuracy of two different computer-aided diagnosis (CAD) systems regarding artificial pulmonary nodules and assess the clinical impact of volumetric inaccuracies in a phantom study.

METHODS

In this phantom study, 59 different phantom arrangements with 326 artificial nodules (178 solid, 148 ground-glass) were scanned at 80 kV, 100 kV, and 120 kV. Four different nodule diameters were used: 5 mm, 8 mm, 10 mm, and 12 mm. Scans were analyzed by a deep-learning (DL)-based CAD and a standard CAD system. Relative volumetric errors (RVE) of each system vs. ground truth and the relative volume difference (RVD) DL-based vs. standard CAD were calculated. The Bland-Altman method was used to define the limits of agreement (LOA). The hypothetical impact on LungRADS classification was assessed for both systems.

RESULTS

There was no difference between the three voltage groups regarding nodule volumetry. Regarding the solid nodules, the RVE of the 5-mm-, 8-mm-, 10-mm-, and 12-mm-size groups for the DL CAD/standard CAD were 12.2/2.8%, 1.3/ - 2.8%, - 3.6/1.5%, and - 12.2/ - 0.3%, respectively. The corresponding values for the ground-glass nodules (GGN) were 25.6%/81.0%, 9.0%/28.0%, 7.6/20.6%, and 6.8/21.2%. The mean RVD for solid nodules/GGN was 1.3/ - 15.2%. Regarding the LungRADS classification, 88.5% and 79.8% of all solid nodules were correctly assigned by the DL CAD and the standard CAD, respectively. 14.9% of the nodules were assigned differently between the systems.

CONCLUSIONS

Patient management may be affected by the volumetric inaccuracy of the CAD systems and hence demands supervision and/or manual correction by a radiologist.

KEY POINTS

• The DL-based CAD system was more accurate in the volumetry of GGN and less accurate regarding solid nodules than the standard CAD system. • Nodule size and attenuation have an effect on the measurement accuracy of both systems; tube voltage has no effect on measurement accuracy. • Measurement inaccuracies of CAD systems can have an impact on patient management, which demands supervision by radiologists.

Diagnostic performance of artificial intelligence for pediatric pulmonary nodule detection in computed tomography of the chest

PURPOSE

To test the performance of a commercially available adult pulmonary nodule detection artificial intelligence (AI) tool in pediatric CT chests.

METHODS

30 consecutive chest CTs with or without contrast of patients ages 12-18 were included. Images were retrospectively reconstructed at 3 mm and 1 mm slice thickness. AI for detection of lung nodules in adults (Syngo CT Lung Computer Aided Detection (CAD)) was evaluated. 3 mm axial images were retrospectively reviewed by two pediatric radiologists (reference read) who determined the location, type, and size of nodules. Lung CAD results at 3 mm and 1 mm slice thickness were compared to reference read by two other pediatric radiologists. Sensitivity (Sn) and positive predictive value (PPV) were analyzed.

RESULTS

The radiologists identified 109 nodules. At 1 mm, CAD detected 70 nodules; 43 true positive (Sn = 39 %), 26 false positive (PPV = 62 %), and 1 nodule which had not been identified by radiologists. At 3 mm, CAD detected 60 nodules; 28 true positive (Sn = 26 %), 30 false positive (PPV = 48 %) and 2 nodules which had not been identified by radiologists. There were 103 solid nodules (47 measuring < 3 mm) and 6 subsolid nodules (5 measuring < 5 mm). When excluding 52 nodules (solid < 3 mm and subsolid < 5 mm) based on algorithm conditions, the Sn increased to 68 % at 1 mm and 49 % at 3 mm but there was no significant change in the PPV measuring 60 % at 1 mm and 48 % at 3 mm.

CONCLUSION

The adult Lung CAD showed low sensitivity in pediatric patients, but better performance at thinner slice thickness and when smaller nodules were excluded.

Multi-scale detection of pulmonary nodules by integrating attention mechanism

The detection of pulmonary nodules has a low accuracy due to the various shapes and sizes of pulmonary nodules. In this paper, a multi-scale detection network for pulmonary nodules based on the attention mechanism is proposed to accurately predict pulmonary nodules. During data processing, the pseudo-color processing strategy is designed to enhance the gray image and introduce more contextual semantic information. In the feature extraction network section, this paper designs a basic module of ResSCBlock integrating attention mechanism for feature extraction. At the same time, the feature pyramid structure is used for feature fusion in the network, and the problem of the detection of small-size nodules which are easily lost is solved by multi-scale prediction method. The proposed method is tested on the LUNA16 data set, with an 83% mAP value. Compared with other detection networks, the proposed method achieves an improvement in detecting pulmonary nodules.

Radiology for Thoracic Conditions in Low- and Middle-Income Countries

With a disproportionately high burden of global morbidity and mortality caused by chronic respiratory diseases (CRDs) in low and middle-income countries (LMICs), access to radiological services is of critical importance for screening, diagnosis, and treatment guidance.

Clinical impact of a deep learning system for automated detection of missed pulmonary nodules on routine body computed tomography including the chest region

OBJECTIVES

To evaluate the clinical impact of a deep learning system (DLS) for automated detection of pulmonary nodules on computed tomography (CT) images as a second reader.

METHODS

This single-centre retrospective study screened 21,150 consecutive body CT studies from September 2018 to February 2019. Pulmonary nodules detected by the DLS on axial CT images but not mentioned in initial radiology reports were flagged. Flagged images were scored by four board-certificated radiologists each with at least 5 years of experience. Nodules with scores of 2 (understandable miss) or 3 (should not be missed) were then categorised as unlikely to be clinically significant (2a or 3a) or likely to be clinically significant (2b or 3b) according to the 2017 Fleischner guidelines for pulmonary nodules. The miss rate was defined as the total number of studies receiving scores of 2 or 3 divided by total screened studies.

RESULTS

Among 172 nodules flagged by the DLS, 60 (35%) missed nodules were confirmed by the radiologists. The nodules were further categorised as 2a, 2b, 3a, and 3b in 24, 14, 10, and 12 studies, respectively, with an overall positive predictive value of 35%. Missed pulmonary nodules were identified in 0.3% of all CT images, and one-third of these lesions were considered clinically significant.

CONCLUSIONS

Use of DLS-assisted automated detection as a second reader can identify missed pulmonary nodules, some of which may be clinically significant.

CLINICAL RELEVANCE/APPLICATION

Use of DLS to help radiologists detect pulmonary lesions may improve patient care.

KEY POINTS

• DLS-assisted automated detection as a second reader is feasible in a large consecutive cohort. • Performance of combined radiologists and DLS was better than DLS or radiologists alone. • Pulmonary nodules were missed more frequently in abdomino-pelvis CT than the thoracic CT.

The current state of knowledge on imaging informatics: a survey among Spanish radiologists

Background

There is growing concern about the impact of artificial intelligence (AI) on radiology and the future of the profession. The aim of this study is to evaluate general knowledge and concerns about trends on imaging informatics among radiologists working in Spain (residents and attending physicians). For this purpose, an online survey among radiologists working in Spain was conducted with questions related to: knowledge about terminology and technologies, need for a regulated academic training on AI and concerns about the implications of the use of these technologies.

Results

A total of 223 radiologists answered the survey, of whom 76.7% were attending physicians and 23.3% residents. General terms such as AI and algorithm had been heard of or read in at least 75.8% and 57.4% of the cases, respectively, while more specific terms were scarcely known. All the respondents consider that they should pursue academic training in medical informatics and new technologies, and 92.9% of them reckon this preparation should be incorporated in the training program of the specialty. Patient safety was found to be the main concern for 54.2% of the respondents. Job loss was not seen as a peril by 45.7% of the participants.

Conclusions

Although there is a lack of knowledge about AI among Spanish radiologists, there is a will to explore such topics and a general belief that radiologists should be trained in these matters. Based on the results, a consensus is needed to change the current training curriculum to better prepare future radiologists.

Artificial Intelligence Algorithm Improves Radiologist Performance in Skeletal Age Assessment: A Prospective Multicenter Randomized Controlled Trial

Background Previous studies suggest that use of artificial intelligence (AI) algorithms as diagnostic aids may improve the quality of skeletal age assessment, though these studies lack evidence from clinical practice. Purpose To compare the accuracy and interpretation time of skeletal age assessment on hand radiograph examinations with and without the use of an AI algorithm as a diagnostic aid. Materials and Methods In this prospective randomized controlled trial, the accuracy of skeletal age assessment on hand radiograph examinations was performed with (n = 792) and without (n = 739) the AI algorithm as a diagnostic aid. For examinations with the AI algorithm, the radiologist was shown the AI interpretation as part of their routine clinical work and was permitted to accept or modify it. Hand radiographs were interpreted by 93 radiologists from six centers. The primary efficacy outcome was the mean absolute difference between the skeletal age dictated into the radiologists' signed report and the average interpretation of a panel of four radiologists not using a diagnostic aid. The secondary outcome was the interpretation time. A linear mixed-effects regression model with random center- and radiologist-level effects was used to compare the two experimental groups. Results Overall mean absolute difference was lower when radiologists used the AI algorithm compared with when they did not (5.36 months vs 5.95 months; P = .04). The proportions at which the absolute difference exceeded 12 months (9.3% vs 13.0%, P = .02) and 24 months (0.5% vs 1.8%, P = .02) were lower with the AI algorithm than without it. Median radiologist interpretation time was lower with the AI algorithm than without it (102 seconds vs 142 seconds, P = .001). Conclusion Use of an artificial intelligence algorithm improved skeletal age assessment accuracy and reduced interpretation times for radiologists, although differences were observed between centers. Clinical trial registration no. NCT03530098 © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Rubin in this issue.

Identification of non-ribosomal peptide synthetase in Ganoderma boninense Pat. that was expressed during the interaction with oil palm

Basal stem rot (BSR) of oil palm is a disastrous disease caused by a white-rot fungus Ganoderma boninense Pat. Non-ribosomal peptides (NRPs) synthesized by non-ribosomal peptide synthetases (NRPSs) are a group of secondary metabolites that act as fungal virulent factors during pathogenesis in the host. In this study, we aimed to isolate NRPS gene of G. boninense strain UPMGB001 and investigate the role of this gene during G. boninense-oil palm interaction. The isolated NRPS DNA fragment of 8322 bp was used to predict the putative peptide sequence of different domains and showed similarity with G. sinense (85%) at conserved motifs of three main NRPS domains. Phylogenetic analysis of NRPS peptide sequences demonstrated that NRPS of G. boninense belongs to the type VI siderophore family. The roots of 6-month-old oil palm seedlings were artificially inoculated for studying NRPS gene expression and disease severity in the greenhouse. The correlation between high disease severity (50%) and high expression (67-fold) of G. boninense NRPS gene at 4 months after inoculation and above indicated that this gene played a significant role in the advancement of BSR disease. Overall, these findings increase our knowledge on the gene structure of NRPS in G. boninense and its involvement in BSR pathogenesis as an effector gene.

Lung computed tomography image segmentation based on U-Net network fused with dilated convolution

PURPOSE

In order to solve the problem of accurate and effective segmentation of the patient's lung computed tomography (CT) images, so as to improve the efficiency of treating lung cancer.

METHOD

We propose a U-Net network (DC-U-Net) fused with dilated convolution, and compare the results of segmented lung CT with DC-U-Net, Otsu and region growth. We use Intersection over Union (IOU), Dice coefficient, Precision and Recall to evaluate the performance of the three algorithms.

RESULTS

Compared with the common segmentation algorithm Otsu and region growing, the segmented image of DC-U-Net is closer to the Ground truth. The IOU of DC-U-Net is 0.9627, and the Dice coefficient is 0.9743, which is close to 1 and much higher than the other two algorithms.

CONCLUSION

We propose that the model can directly segment the original image automatically, and the segmentation effect is good. This model speeds up the segmentation, simplifies the steps of medical image segmentation, and provides better segmentation for subsequent lung blood vessels, trachea and other tissues.

Prospective Study of Spatial Distribution of Missed Lung Nodules by Readers in CT Lung Screening Using Computer-assisted Detection

PURPOSE

To evaluate the spatial patterns of missed lung nodules in a real-life routine screening environment.

MATERIALS AND METHODS

In a screening institute, 4,822 consecutive adults underwent chest CT, and each image set was independently interpreted by two radiologists in three steps: (1) independently interpreted without computer-assisted detection (CAD) software, (2) independently referred to the CAD results, (3) determined by the consensus of the two radiologists. The locations of nodules and the detection performance data were semi-automatically collected using a CAD server integrated into the reporting system. Fisher's exact test was employed for evaluating findings in different lung divisions. Probability maps were drawn to illustrate the spatial distribution of radiologists' missed nodules.

RESULTS

Radiologists significantly tended to miss lung nodules in the bilateral hilar divisions (p < 0.01). Some radiologists had their own spatial pattern of missed lung nodules.

CONCLUSION

Radiologists tend to miss lung nodules present in the hilar regions significantly more often than in the rest of the lung.

Artificial intelligence-based vessel suppression for detection of sub-solid nodules in lung cancer screening computed tomography

Background

Lung cancer screening (LCS) with low-dose computed tomography (LDCT) helps early lung cancer detection, commonly presenting as small pulmonary nodules. Artificial intelligence (AI)-based vessel suppression (AI-VS) and automatic detection (AI-AD) algorithm can improve detection of subsolid nodules (SSNs) on LDCT. We assessed the impact of AI-VS and AI-AD in detection and classification of SSNs [ground-glass nodules (GGNs) and part-solid nodules (PSNs)], on LDCT performed for LCS.

Methods

Following regulatory approval, 123 LDCT examinations with sub-solid pulmonary nodules (average diameter ≥6 mm) were processed to generate three image series for each examination-unprocessed, AI-VS, and AI-AD series with annotated lung nodules. Two thoracic radiologists in consensus formed the standard of reference (SOR) for this study. Two other thoracic radiologists (R1 and R2; 5 and 10 years of experience in thoracic CT image interpretation) independently assessed the unprocessed images alone, then together with AI-VS series, and finally with AI-AD for detecting all ≥6 mm GGN and PSN. We performed receiver operator characteristics (ROC) and Cohen's Kappa analyses for statistical analyses.

Results

On unprocessed images, R1 and R2 detected 232/310 nodules (R1: 114 GGN, 118 PSN) and 255/310 nodules (R2: 122 GGN, 133 PSN), respectively (P>0.05). On AI-VS images, they detected 249/310 nodules (119 GGN, 130 PSN) and 277/310 nodules (128 GGN, 149 PSN), respectively (P≥0.12). When compared to the SOR, accuracy (AUC) for detection of PSN on the AI-VS images (AUC 0.80-0.81) was greater than on the unprocessed images (AUC 0.70-0.76). AI-VS images enabled detection of solid components in five nodules deemed as GGN on the unprocessed images. Accuracy of AI-AD was lower than both the radiologists (AUC 0.60-0.72).

Conclusions

AI-VS improved the detection and classification of SSN into GGN and PSN on LDCT of the chest for the two radiologist (R1 and R2) readers.

Combining machine learning and texture analysis to differentiate mediastinal lymph nodes in lung cancer patients

Evaluate whether texture analysis associated with machine learning approaches could differentiate between malignant and benign lymph nodes. A total 18 patients with lung cancer were selected, with 39 lymph nodes, being 15 malignant and 24 benign. Retrospective computed tomography scans were utilized both with and without contrast medium. The great differential of this work was the use of 15 textures from mediastinal lymph nodes, with five different physicians as operators. First and second order statistical textures such as gray level run length and co-occurrence matrix were extracted and applied to three different machine learning classifiers. The best machine learning classifier demonstrated a variability of less than 5% among operators. The support vector machine (SVM) classifier presented 95% of the area under the ROC curve (AUC) and 89% of sensitivity for sequences without contrast medium. SVM classifier presented 93% of AUC and 86% of sensitivity for sequences with contrast medium. Texture analysis and machine learning may be helpful in the differentiation between malign and benign lymph nodes. This study can aid the physician in diagnosis and staging of lymph nodes and potentially reduce the number of invasive analysis to histopathological confirmation.

Computer-Aided System Application Value for Assessing Hip Development

Purpose

A computer-aided system was used to semiautomatically measure Tönnis angle, Sharp angle, and center-edge (CE) angle using contours of the hip bones to establish an auxiliary measurement model for developmental screening or diagnosis of hip joint disorders.

Methods

We retrospectively analyzed bilateral hip x-rays for 124 patients (41 men and 83 women aged 20-70 years) who presented at the Affiliated Zhongshan Hospital of Dalian University in 2017 and 2018. All images were imported into a computer-aided detection system. After manually outlining hip bone contours, Tönnis angle, Sharp angle, and CE angle marker lines were automatically extracted, and the angles were measured and recorded. An imaging physician also manually measured all angles and recorded hip development, and Pearson correlation coefficients were used to compare computer-aided system measurements with imaging physician measurements. Accuracy for different angles was calculated, and the area under the receiver operating characteristic (AUROC) curve was used to represent the diagnostic efficiency of the computer-aided system.

Results

For Tönnis angle, Sharp angle, and CE angle, correlation coefficients were 0.902, 0.887, and 0.902, respectively; the accuracies of the computer-aided detection system were 89.1, 93.1, and 82.3%; and the AUROC curve values were 0.940, 0.956, and 0.948.

Conclusion

The measurements of Tönnis angle, Sharp angle, and CE angle using the semiautomatic system were highly correlated with the measurements of the imaging physician and can be used to assess hip joint development with high accuracy and diagnostic efficiency.

Transfer-to-Transfer Learning Approach for Computer Aided Detection of COVID-19 in Chest Radiographs

The coronavirus disease 2019 (COVID-19) global pandemic has severely impacted lives across the globe. Respiratory disorders in COVID-19 patients are caused by lung opacities similar to viral pneumonia. A Computer-Aided Detection (CAD) system for the detection of COVID-19 using chest radiographs would provide a second opinion for radiologists. For this research, we utilize publicly available datasets that have been marked by radiologists into two-classes (COVID-19 and non-COVID-19). We address the class imbalance problem associated with the training dataset by proposing a novel transfer-to-transfer learning approach, where we break a highly imbalanced training dataset into a group of balanced mini-sets and apply transfer learning between these. We demonstrate the efficacy of the method using well-established deep convolutional neural networks. Our proposed training mechanism is more robust to limited training data and class imbalance. We study the performance of our algorithm(s) based on 10-fold cross validation and two hold-out validation experiments to demonstrate its efficacy. We achieved an overall sensitivity of 0.94 for the hold-out validation experiments containing 2265 and 2139 marked as COVID-19 chest radiographs, respectively. For the 10-fold cross validation experiment, we achieve an overall Area under the Receiver Operating Characteristic curve (AUC) value of 0.996 for COVID-19 detection. This paper serves as a proof-of-concept that an automated detection approach can be developed with a limited set of COVID-19 images, and in areas with scarcity of trained radiologists.

Efficiency of a computer-aided diagnosis (CAD) system with deep learning in detection of pulmonary nodules on 1-mm-thick images of computed tomography

PURPOSE

To evaluate the performance of a deep learning-based computer-aided diagnosis (CAD) system at detecting pulmonary nodules on CT by comparing radiologists' readings with and without CAD.

MATERIALS AND METHODS

A total of 120 chest CT images were randomly selected from patients with suspected lung cancer. The gold standard of nodules ≥ 3 mm was established by a panel of three expert radiologists. Two less experienced radiologists read the images without and afterward with CAD system. Their reading times were recorded.

RESULTS

The radiologists' sensitivity increased from 20.9% to 38.0% with the introduction of CAD. The positive predictive value (PPV) decreased from 70.5% to 61.8%, and the F1-score increased from 32.2% to 47.0%. The sensitivity significantly increased from 13.7% to 32.4% for small nodules (3-6 mm) and from 33.3% to 47.6% for medium nodules (6-10 mm). CAD alone showed a sensitivity of 70.3%, a PPV of 57.9%, and an F1-score of 63.5%. Reading time decreased by 11.3% with the use of CAD.

CONCLUSION

CAD improved the less experienced radiologists' sensitivity in detecting pulmonary nodules of all sizes, especially including a significant improvement in the detection of clinically important-sized medium nodules (6-10 mm) as well as small nodules (3-6 mm) and reduced their reading time.

Consistency of radiologists in identifying pulmonary nodules based on low-dose computed tomography

Background

To study the consistency of radiologists in identifying pulmonary nodules based on low-dose computed tomography (LDCT), and to analyze factors that affect the consistency.

Methods

A total of 750 LDCT cases were collected randomly from three medical centers. Three experienced chest radiologists independently evaluated and detected the pulmonary nodules on 625 cases of LDCT images. The detected nodules were classified into 3 groups: group I (detected by all radiologists); group II (detected by two radiologists); group III (detected by only one radiologist). The consistency with respect to the image features of individual nodules was assessed.

Results

A total of 1,206 nodules were identified by the three radiologists. There were 234 (19.4%) nodules in group I, 377 (31.3%) nodules in group II, and 595 (49.3%) nodules in group III. Logistic regression showed that the size, density, and location of the nodules correlated with the detection of nodules. Nodules sized great than or equal to 4 mm were more consistently identified than nodules sized less than 4 mm. Solid and calcified nodules were more consistently identified than sub-solid nodules. Nodules located in the outer zone were more consistently identified than hilar nodules.

Conclusions

There was considerable inter-reader variability with respect to identification of pulmonary nodules in LDCT. Larger nodules, solid or calcified nodules, and nodules located in the outer zone were more consistently identified.

Segmentation and suppression of pulmonary vessels in low-dose chest CT scans

PURPOSE

The suppression of pulmonary vessels in chest computed tomography (CT) images can enhance the conspicuity of lung nodules, thereby improving the detection rate of early lung cancer. This study aimed to develop two key techniques in vessel suppression, that is, segmentation and removal of pulmonary vessels while preserving the nodules.

METHODS

Pulmonary vessel segmentation and removal methods in CT images were developed. The vessel segmentation method used a framework of two cascaded convolutional neural networks (CNNs). A bi-class segmentation network was utilized in the first step to extract high-intensity structures, including both vessels and nonvascular tissues such as nodules. A tri-class segmentation network was employed in the second step to distinguish the vessels from nonvascular tissues (mainly nodules) and the lung parenchyma. In the vessel removal method, the voxels in the segmented vessels were replaced with randomly selected voxels from the surrounding lung parenchyma. The dataset in this study comprised 50 three-dimensional (3D) low-dose chest CT images. The labels for vessel and nodule segmentation were annotated with a semi automatic approach. The two cascaded networks for pulmonary vessel segmentation were trained with CT images of 40 cases and tested with CT images of ten cases. Pulmonary vessels were removed from the ten testing scans based on the predicted segmentation results. In addition to qualitative evaluation to the effects of segmentation and removal, the segmentation results were quantitatively evaluated using Dice coefficient (DICE), Jaccard index (JAC), and volumetric similarity (VS) and the removal results were evaluated using contrast-to-noise ratio (CNR).

RESULTS

In the first step of vessel segmentation, the mean DICE, JAC, and VS for high-intensity tissues, including both vessels and nodules, were 0.943, 0.893, and 0.991, respectively. In the second step, all the nodules were separated from the vessels, and the mean DICE, JAC, and VS for the vessels were 0.941, 0.890, and 0.991, respectively. After vessel removal, the mean CNR for nodules was improved from 4.23 (6.26 dB) to 6.95 (8.42 dB).

CONCLUSIONS

Quantitative and qualitative evaluations demonstrated that the proposed method achieved a high accuracy for pulmonary vessel segmentation and a good effect on pulmonary vessel suppression.

Evaluating the performance of a deep learning-based computer-aided diagnosis (DL-CAD) system for detecting and characterizing lung nodules: Comparison with the performance of double reading by radiologists

Background

The study was conducted to evaluate the performance of a state‐of‐the‐art commercial deep learning‐based computer‐aided diagnosis (DL‐CAD) system for detecting and characterizing pulmonary nodules.

Methods

Pulmonary nodules in 346 healthy subjects (male: female = 221:125, mean age 51 years) from a lung cancer screening program conducted from March to November 2017 were screened using a DL‐CAD system and double reading independently, and their performance in nodule detection and characterization were evaluated. An expert panel combined the results of the DL‐CAD system and double reading as the reference standard.

Results

The DL‐CAD system showed a higher detection rate than double reading, regardless of nodule size (86.2% vs. 79.2%; P < 0.001): nodules ≥ 5 mm (96.5% vs. 88.0%; P = 0.008); nodules < 5 mm (84.3% vs. 77.5%; P < 0.001). However, the false positive rate (per computed tomography scan) of the DL‐CAD system (1.53, 529/346) was considerably higher than that of double reading (0.13, 44/346; P < 0.001). Regarding nodule characterization, the sensitivity and specificity of the DL‐CAD system for distinguishing solid nodules > 5 mm (90.3% and 100.0%, respectively) and ground‐glass nodules (100.0% and 96.1%, respectively) were close to that of double reading, but dropped to 55.5% and 93%, respectively, when discriminating part solid nodules.

Conclusion

Our DL‐CAD system detected significantly more nodules than double reading. In the future, false positive findings should be further reduced and characterization accuracy improved.

Vessel suppressed chest Computed Tomography for semi-automated volumetric measurements of solid pulmonary nodules

OBJECTIVE

To evaluate whether vessel-suppressed computed tomography (VSCT) can be reliably used for semi-automated volumetric measurements of solid pulmonary nodules, as compared to standard CT (SCT) MATERIAL AND METHODS: Ninety-three SCT were elaborated by dedicated software (ClearRead CT, Riverain Technologies, Miamisburg, OH, USA), that allows subtracting vessels from lung parenchyma. Semi-automated volumetric measurements of 65 solid nodules were compared between SCT and VSCT. The measurements were repeated by two readers. For each solid nodule, volume measured on SCT by Reader 1 and Reader 2 was averaged and the average volume between readers acted as standard of reference value. Concordance between measurements was assessed using Lin's Concordance Correlation Coefficient (CCC). Limits of agreement (LoA) between readers and CT datasets were evaluated.

RESULTS

Standard of reference nodule volume ranged from 13 to 366 mm³. The mean overestimation between readers was 3 mm³ and 2.9 mm³ on SCT and VSCT, respectively. Semi-automated volumetric measurements on VSCT showed substantial agreement with the standard of reference (Lin's CCC = 0.990 for Reader 1; 0.985 for Reader 2). The upper and lower LoA between readers' measurements were (16.3, -22.4 mm³) and (15.5, -21.4 mm³) for SCT and VSCT, respectively.

CONCLUSIONS

VSCT datasets are feasible for the measurements of solid nodules, showing an almost perfect concordance between readers and with measurements on SCT.

Performance analysis of a computer-aided detection system for lung nodules in CT at different slice thicknesses

We study the performance of a computer-aided detection (CAD) system for lung nodules in computed tomography (CT) as a function of slice thickness. In addition, we propose and compare three different training methodologies for utilizing nonhomogeneous thickness training data (i.e., composed of cases with different slice thicknesses). These methods are (1) aggregate training using the entire suite of data at their native thickness, (2) homogeneous subset training that uses only the subset of training data that matches each testing case, and (3) resampling all training and testing cases to a common thickness. We believe this study has important implications for how CT is acquired, processed, and stored. We make use of 192 CT cases acquired at a thickness of 1.25 mm and 283 cases at 2.5 mm. These data are from the publicly available Lung Nodule Analysis 2016 dataset. In our study, CAD performance at 2.5 mm is comparable with that at 1.25 mm and is much better than at higher thicknesses. Also, resampling all training and testing cases to 2.5 mm provides the best performance among the three training methods compared in terms of accuracy, memory consumption, and computational time.

Missed lung cancer: when, where, and why?

Missed lung cancer is a source of concern among radiologists and an important medicolegal challenge. In 90% of the cases, errors in diagnosis of lung cancer occur on chest radiographs. It may be challenging for radiologists to distinguish a lung lesion from bones, pulmonary vessels, mediastinal structures, and other complex anatomical structures on chest radiographs. Nevertheless, lung cancer can also be overlooked on computed tomography (CT) scans, regardless of the context, either if a clinical or radiologic suspect exists or for other reasons. Awareness of the possible causes of overlooking a pulmonary lesion can give radiologists a chance to reduce the occurrence of this eventuality. Various factors contribute to a misdiagnosis of lung cancer on chest radiographs and on CT, often very similar in nature to each other. Observer error is the most significant one and comprises scanning error, recognition error, decision-making error, and satisfaction of search. Tumor characteristics such as lesion size, conspicuity, and location are also crucial in this context. Even technical aspects can contribute to the probability of skipping lung cancer, including image quality and patient positioning and movement. Albeit it is hard to remove missed lung cancer completely, strategies to reduce observer error and methods to improve technique and automated detection may be valuable in reducing its likelihood.

The impact of trained radiographers as concurrent readers on performance and reading time of experienced radiologists in the UK Lung Cancer Screening (UKLS) trial

Objectives

To compare radiologists’ performance reading CTs independently with their performance using radiographers as concurrent readers in lung cancer screening.

Methods

369 consecutive baseline CTs performed for the UK Lung Cancer Screening (UKLS) trial were double-read by radiologists reading either independently or concurrently with a radiographer. In concurrent reading, the radiologist reviewed radiographer-identified nodules and then detected any additional nodules. Radiologists recorded their independent and concurrent reading times. For each radiologist, sensitivity, average false-positive detections (FPs) per case and mean reading times for each method were calculated.

Results

694 nodules in 246/369 (66.7%) studies comprised the reference standard. Radiologists’ mean sensitivity and average FPs per case both increased with concurrent reading compared to independent reading (90.8 ± 5.6% vs. 77.5 ± 11.2%, and 0.60 ± 0.53 vs. 0.33 ± 0.20, respectively; p < 0.05 for 3/4 and 2/4 radiologists, respectively). The mean reading times per case decreased from 9.1 ± 2.3 min with independent reading to 7.2 ± 1.0 min with concurrent reading, decreasing significantly for 3/4 radiologists (p < 0.05).

Conclusions

The majority of radiologists demonstrated improved sensitivity, a small increase in FP detections and a statistically significantly reduced reading time using radiographers as concurrent readers.

Key Points

• Radiographers as concurrent readers could improve radiologists’ sensitivity in lung nodule detection.

• An increase in false-positive detections with radiographer-assisted concurrent reading occurred.

• The false-positive detection rate was still lower than reported for computer-aided detection.

• Concurrent reading with radiographers was also faster than single reading.

• The time saved per case using concurrently reading radiographers was relatively modest.

Electronic supplementary material

The online version of this article (doi:10.1007/s00330-017-4903-z) contains supplementary material, which is available to authorized users.

Role of Computer Aided Diagnosis (CAD) in the detection of pulmonary nodules on 64 row multi detector computed tomography

AIMS AND OBJECTIVES

To determine the overall performance of an existing CAD algorithm with thin-section computed tomography (CT) in the detection of pulmonary nodules and to evaluate detection sensitivity at a varying range of nodule density, size, and location.

MATERIALS AND METHODS

A cross-sectional prospective study was conducted on 20 patients with 322 suspected nodules who underwent diagnostic chest imaging using 64-row multi-detector CT. The examinations were evaluated on reconstructed images of 1.4 mm thickness and 0.7 mm interval. Detection of pulmonary nodules, initially by a radiologist of 2 years experience (RAD) and later by CAD lung nodule software was assessed. Then, CAD nodule candidates were accepted or rejected accordingly. Detected nodules were classified based on their size, density, and location. The performance of the RAD and CAD system was compared with the gold standard that is true nodules confirmed by consensus of senior RAD and CAD together. The overall sensitivity and false-positive (FP) rate of CAD software was calculated.

OBSERVATIONS AND RESULTS

Of the 322 suspected nodules, 221 were classified as true nodules on the consensus of senior RAD and CAD together. Of the true nodules, the RAD detected 206 (93.2%) and 202 (91.4%) by the CAD. CAD and RAD together picked up more number of nodules than either CAD or RAD alone. Overall sensitivity for nodule detection with the CAD program was 91.4%, and FP detection per patient was 5.5%. The CAD showed comparatively higher sensitivity for nodules of size 4-10 mm (93.4%) and nodules in hilar (100%) and central (96.5%) location when compared to RAD's performance.

CONCLUSION

CAD performance was high in detecting pulmonary nodules including the small size and low-density nodules. CAD even with relatively high FP rate, assists and improves RAD's performance as a second reader, especially for nodules located in the central and hilar region and for small nodules by saving RADs time.

Maximum-Intensity-Projection and Computer-Aided-Detection Algorithms as Stand-Alone Reader Devices in Lung Cancer Screening Using Different Dose Levels and Reconstruction Kernels

OBJECTIVE

The objective of our study was to evaluate lung nodule detection rates on standard and microdose chest CT with two different computer-aided detection systems (SyngoCT-CAD, VA 20, Siemens Healthcare [CAD1]; Lung CAD, IntelliSpace Portal DX Server, Philips Healthcare [CAD2]) as well as maximum-intensity-projection (MIP) images. We also assessed the impact of different reconstruction kernels.

MATERIALS AND METHODS

Standard and microdose CT using three reconstruction kernels (i30, i50, i70) was performed with an anthropomorphic chest phantom. We placed 133 ground-glass and 133 solid nodules (diameters of 5 mm, 8 mm, 10 mm, and 12 mm) in 55 phantoms. Four blinded readers evaluated the MIP images; one recorded the results of CAD1 and CAD2. Sensitivities for CAD and MIP nodule detection on standard dose and microdose CT were calculated for each reconstruction kernel.

RESULTS

Dose for microdose CT was significantly less than that for standard-dose CT (0.1323 mSv vs 1.65 mSv; p < 0.0001). CAD1 delivered superior results compared with CAD2 for standard-dose and microdose CT (p < 0.0001). At microdose level, the best stand-alone sensitivity (97.6%) was comparable with CAD1 sensitivity (96.0%; p = 0.36; both with i30 reconstruction kernel). Pooled sensitivities for all nodules, doses, and reconstruction kernels on CAD1 ranged from 88.9% to 97.3% versus 49.6% to 73.9% for CAD2. The best sensitivity was achieved with standard-dose CT, i50 kernel, and CAD1 (97.3%) versus 96% with microdose CT, i30 or i50 kernel, and CAD1. MIP images and CAD1 had similar performance at both dose levels (p = 0.1313 and p = 0.48).

CONCLUSION

Submillisievert CT is feasible for detecting solid and ground-glass nodules that require soft-tissue kernels for MIP and CAD systems to achieve acceptable sensitivities. MIP reconstructions remain a valuable adjunct to the interpretation of chest CT for increasing sensitivity and have the advantage of significantly lower false-positive rates.

Low-Dose CT Screening for Lung Cancer: Computer-aided Detection of Missed Lung Cancers

Purpose To update information regarding the usefulness of computer-aided detection (CAD) systems with a focus on the most critical category, that of missed cancers at earlier imaging, for cancers that manifest as a solid nodule. Materials and Methods By using a HIPAA-compliant institutional review board-approved protocol where informed consent was obtained, 50 lung cancers that manifested as a solid nodule on computed tomographic (CT) scans in annual rounds of screening (time 1) were retrospectively identified that could, in retrospect, be identified on the previous CT scans (time 0). Four CAD systems were compared, which were referred to as CAD 1, CAD 2, CAD 3, and CAD 4. The total number of accepted CAD-system-detected nodules at time 0 was determined by consensus of two radiologists and the number of CAD-system-detected nodules that were rejected by the radiologists was also documented. Results At time 0 when all the cancers had been missed, CAD system detection rates for the cancers were 56%, 70%, 68%, and 60% (κ = 0.45) for CAD systems 1, 2, 3, and 4, respectively. At time 1, the rates were 74%, 82%, 82%, and 78% (κ = 0.32), respectively. The average diameter of the 50 cancers at time 0 and time 1 was 4.8 mm and 11.4 mm, respectively. The number of CAD-system-detected nodules that were rejected per CT scan for CAD systems 1-4 at time 0 was 7.4, 1.7, 0.6, and 4.5 respectively. Conclusion CAD systems detected up to 70% of lung cancers that were not detected by the radiologist but failed to detect about 20% of the lung cancers when they were identified by the radiologist, which suggests that CAD may be useful in the role of second reader. (©) RSNA, 2016.

Effect of radiation dose reduction and iterative reconstruction on computer-aided detection of pulmonary nodules: Intra-individual comparison

OBJECTIVE

To evaluate the effect of radiation dose reduction and iterative reconstruction (IR) on the performance of computer-aided detection (CAD) for pulmonary nodules.

METHODS

In this prospective study twenty-five patients were included who were scanned for pulmonary nodule follow-up. Image acquisition was performed at routine dose and three reduced dose levels in a single session by decreasing mAs-values with 45%, 60% and 75%. Tube voltage was fixed at 120 kVp for patients ≥ 80 kg and 100 kVp for patients < 80 kg. Data were reconstructed with filtered back projection (FBP), iDose(4) (levels 1,4,6) and IMR (levels 1-3). All noncalcified solid pulmonary nodules ≥ 4 mm identified by two radiologists in consensus served as the reference standard. Subsequently, nodule volume was measured with CAD software and compared to the reference consensus. The numbers of true-positives, false-positives and missed pulmonary nodules were evaluated as well as the sensitivity.

RESULTS

Median effective radiation dose was 2.2 mSv at routine dose and 1.2, 0.9 and 0.6 mSv at respectively 45%, 60% and 75% reduced dose. A total of 28 pulmonary nodules were included. With FBP at routine dose, 89% (25/28) of the nodules were correctly identified by CAD. This was similar at reduced dose levels with FBP, iDose(4) and IMR. CAD resulted in a median number of false-positives findings of 11 per scan with FBP at routine dose (93% of the CAD marks) increasing to 15 per scan with iDose(4) (95% of the CAD marks) and 26 per scan (96% of the CAD marks) with IMR at the lowest dose level.

CONCLUSION

CAD can identify pulmonary nodules at submillisievert dose levels with FBP, hybrid and model-based IR. However, the number of false-positive findings increased using hybrid and especially model-based IR at submillisievert dose while dose reduction did not affect the number of false-positives with FBP.

The Solitary Pulmonary Nodule

Due to the high etiological diversity and the potential for malignancy, pulmonary nodules represent a clinical challenge, becoming increasingly frequent as the number of CT examinations rises. The topic gains even more importance as clear evidence for the effectiveness of CT screening was provided by the National Lung Screening Trial (NLST). Yet, the results were tempered by the high false-positive rate and the requirement of performing further diagnostic procedures. The management of those detected solitary pulmonary nodules is currently based on the individuals' risk of developing lung cancer, the pulmonary nodule characteristics and the capability of diagnostic and therapeutic approaches.

Screening for lung cancer with low-dose computed tomography: a review of current status

Screening using low-dose computed tomography (CT) represents an exciting new development in the struggle to improve outcomes for people with lung cancer. Randomised controlled evidence demonstrating a 20% relative lung cancer mortality benefit has led to endorsement of screening by several expert bodies in the US and funding by healthcare providers. Despite this pivotal result, many questions remain regarding technical and logistical aspects of screening, cost-effectiveness and generalizability to other settings. This review discusses the rationale behind screening, the results of on-going trials, potential harms of screening and current knowledge gaps.

Lung cancer screening: state of the art

Results from the National Lung Screening Trial have confirmed that lung cancer mortality is reduced using low-dose CT screening. Opening a lung cancer screening program requires a multidisciplinary approach. While the fundamental aspects of a screening program are similar, such as scheduling, performing, and managing follow-up, there are aspects of a lung cancer screening program that are unique. This article will discuss factors important in establishing a state of the art lung cancer screening program.

Machine Learning in Computer-aided Diagnosis of the Thorax and Colon in CT: A Survey

Computer-aided detection (CADe) and diagnosis (CAD) has been a rapidly growing, active area of research in medical imaging. Machine leaning (ML) plays an essential role in CAD, because objects such as lesions and organs may not be represented accurately by a simple equation; thus, medical pattern recognition essentially require "learning from examples." One of the most popular uses of ML is the classification of objects such as lesion candidates into certain classes (e.g., abnormal or normal, and lesions or non-lesions) based on input features (e.g., contrast and area) obtained from segmented lesion candidates. The task of ML is to determine "optimal" boundaries for separating classes in the multidimensional feature space which is formed by the input features. ML algorithms for classification include linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), multilayer perceptrons, and support vector machines (SVM). Recently, pixel/voxel-based ML (PML) emerged in medical image processing/analysis, which uses pixel/voxel values in images directly, instead of features calculated from segmented lesions, as input information; thus, feature calculation or segmentation is not required. In this paper, ML techniques used in CAD schemes for detection and diagnosis of lung nodules in thoracic CT and for detection of polyps in CT colonography (CTC) are surveyed and reviewed.

Computer-aided diagnosis systems for lung cancer: challenges and methodologies

This paper overviews one of the most important, interesting, and challenging problems in oncology, the problem of lung cancer diagnosis. Developing an effective computer-aided diagnosis (CAD) system for lung cancer is of great clinical importance and can increase the patient's chance of survival. For this reason, CAD systems for lung cancer have been investigated in a huge number of research studies. A typical CAD system for lung cancer diagnosis is composed of four main processing steps: segmentation of the lung fields, detection of nodules inside the lung fields, segmentation of the detected nodules, and diagnosis of the nodules as benign or malignant. This paper overviews the current state-of-the-art techniques that have been developed to implement each of these CAD processing steps. For each technique, various aspects of technical issues, implemented methodologies, training and testing databases, and validation methods, as well as achieved performances, are described. In addition, the paper addresses several challenges that researchers face in each implementation step and outlines the strengths and drawbacks of the existing approaches for lung cancer CAD systems.

A review of computer-aided diagnosis in thoracic and colonic imaging

Medical imaging has been indispensable in medicine since the discovery of x-rays. Medical imaging offers useful information on patients' medical conditions and on the causes of their symptoms and diseases. As imaging technologies advance, a large number of medical images are produced which physicians/radiologists must interpret. Thus, computer aids are demanded and become indispensable in physicians' decision making based on medical images. Consequently, computer-aided detection and diagnosis (CAD) has been investigated and has been an active research area in medical imaging. CAD is defined as detection and/or diagnosis made by a radiologist/physician who takes into account the computer output as a "second opinion". In CAD research, detection and diagnosis of lung and colorectal cancer in thoracic and colonic imaging constitute major areas, because lung and colorectal cancers are the leading and second leading causes, respectively, of cancer deaths in the U.S. and also in other countries. In this review, CAD of the thorax and colon, including CAD for detection and diagnosis of lung nodules in thoracic CT, and that for detection of polyps in CT colonography, are reviewed.

Technical parameters and interpretive issues in screening computed tomography scans for lung cancer

Lung cancer screening computed tomographies (CTs) differ from traditional chest CT scans in that they are performed at very low radiation doses, which allow the detection of small nodules but which have a much higher noise content than would be acceptable in a diagnostic chest CT. The technical parameters require a great deal of attention on the part of the user, because inappropriate settings could result in either excess radiation dose to the large population of screened individuals or in low-quality images with impaired nodule detectability. Both situations undermine the main goal of the screening program, which is to detect lung nodules using as low a radiation dose as can reasonably be achieved. Once an image has been obtained, there are unique interpretive issues that must be addressed mainly because of the very high noise content of the images and the high prevalence of incidental findings in the chest unrelated to the sought-after pulmonary nodules.