A cloud-based computer-aided detection system improves identification of lung nodules on computed tomography scans of patients with extra-thoracic malignancies

Pricing and cost-saving potential for deep-learning computer-aided lung nodule detection software in CT lung cancer screening

OBJECTIVE

An increasing number of commercial deep learning computer-aided detection (DL-CAD) systems are available but their cost-saving potential is largely unknown. This study aimed to gain insight into appropriate pricing for DL-CAD in different reading modes to be cost-saving and to determine the potentially most cost-effective reading mode for lung cancer screening.

METHODS

In three representative settings, DL-CAD was evaluated as a concurrent, pre-screening, and second reader. Scoping review was performed to estimate radiologist reading time with and without DL-CAD. Hourly cost of radiologist time was collected for the USA (€196), UK (€127), and Poland (€45), and monetary equivalence of saved time was calculated. The minimum number of screening CTs to reach break-even was calculated for one-time investment of €51,616 for DL-CAD.

RESULTS

Mean reading time was 162 (95% CI: 111-212) seconds per case without DL-CAD, which decreased by 77 (95% CI: 47-107) and 104 (95% CI: 71-136) seconds for DL-CAD as concurrent and pre-screening reader, respectively, and increased by 33-41 s for DL-CAD as second reader. This translates into €1.0-4.3 per-case cost for concurrent reading and €0.8-5.7 for pre-screening reading in the USA, UK, and Poland. To achieve break-even with a one-time investment, the minimum number of CT scans was 12,300-53,600 for concurrent reader, and 9400-65,000 for pre-screening reader in the three countries.

CONCLUSIONS

Given current pricing, DL-CAD must be priced substantially below €6 in a pay-per-case setting or used in a high-workload environment to reach break-even in lung cancer screening. DL-CAD as pre-screening reader shows the largest potential to be cost-saving.

CRITICAL RELEVANCE STATEMENT

Deep-learning computer-aided lung nodule detection (DL-CAD) software must be priced substantially below 6 euro in a pay-per-case setting or must be used in high-workload environments with one-time investment in order to achieve break-even. DL-CAD as a pre-screening reader has the greatest cost savings potential.

KEY POINTS

• DL-CAD must be substantially below €6 in a pay-per-case setting to reach break-even. • DL-CAD must be used in a high-workload screening environment to achieve break-even. • DL-CAD as a pre-screening reader shows the largest potential to be cost-saving.

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.

The Effects of Artificial Intelligence Assistance on the Radiologists' Assessment of Lung Nodules on CT Scans: A Systematic Review

To reduce the number of missed or misdiagnosed lung nodules on CT scans by radiologists, many Artificial Intelligence (AI) algorithms have been developed. Some algorithms are currently being implemented in clinical practice, but the question is whether radiologists and patients really benefit from the use of these novel tools. This study aimed to review how AI assistance for lung nodule assessment on CT scans affects the performances of radiologists. We searched for studies that evaluated radiologists' performances in the detection or malignancy prediction of lung nodules with and without AI assistance. Concerning detection, radiologists achieved with AI assistance a higher sensitivity and AUC, while the specificity was slightly lower. Concerning malignancy prediction, radiologists achieved with AI assistance generally a higher sensitivity, specificity and AUC. The radiologists' workflows of using the AI assistance were often only described in limited detail in the papers. As recent studies showed improved performances of radiologists with AI assistance, AI assistance for lung nodule assessment holds great promise. To achieve added value of AI tools for lung nodule assessment in clinical practice, more research is required on the clinical validation of AI tools, impact on follow-up recommendations and ways of using AI tools.

Computer-aided detection improves brain metastasis identification on non-enhanced CT in less experienced radiologists

Background

Brain metastases (BMs) are the most common intracranial tumors causing neurological complications associated with significant morbidity and mortality.

Purpose

To evaluate the effect of computer-aided detection (CAD) on the performance of observers in detecting BMs on non-enhanced computed tomography (NECT).

Material and Methods

Three less experienced and three experienced radiologists interpreted 30 NECT scans with 89 BMs in 25 cases to detect BMs with and without the assistance of CAD. The observers’ sensitivity, number of false positives (FPs), positive predictive value (PPV), and reading time with and without CAD were compared using paired t-tests. The sensitivity of CAD and the observers were compared using a one-sample t-test

Results

With CAD, less experienced radiologists’ sensitivity significantly increased from 27.7% ± 4.6% to 32.6% ± 4.8% ( P = 0.007), while the experienced radiologists’ sensitivity did not show a significant difference (from 33.3% ± 3.5% to 31.9% ± 3.7%; P = 0.54). There was no significant difference between conditions with CAD and without CAD for FPs (less experienced radiologists: 23.0 ± 10.4 and 25.0 ± 9.3; P = 0.32; experienced radiologists: 18.3 ± 7.4 and 17.3 ± 6.7; P = 0.76) and PPVs (less experienced radiologists: 57.9% ± 8.3% and 50.9% ± 7.0%; P = 0.14; experienced radiologists: 61.8% ± 12.7% and 64.0% ± 12.1%; P = 0.69). There were no significant differences in reading time with and without CAD (85.0 ± 45.6 s and 73.7 ± 36.7 s; P = 0.09). The sensitivity of CAD was 47.2% (with a PPV of 8.9%), which was significantly higher than that of any radiologist ( P < 0.001).

Conclusion

CAD improved BM detection sensitivity on NECT without increasing FPs or reading time among less experienced radiologists, but this was not the case among experienced radiologists.

U"AI" Testing: User Interface and Usability Testing of a Chest X-ray AI Tool in a Simulated Real-World Workflow

Purpose: To observe interactions of practicing radiologists with a chest x-ray AI tool and evaluate its usability and impact on workflow efficiency. Methods: Using a simulated clinical workflow and remote multi-monitor screensharing, we prospectively assessed the interactions of 10 staff radiologists (5-33 years of experience) with a PACS-embedded, regulatory-approved chest x-ray AI tool. Qualitatively, we collected feedback using a think-aloud method and post-testing semi-structured interview; transcript themes were categorized by: (1) AI tool features, (2) deployment considerations, and (3) broad human-AI interactions. Quantitatively, we used time-stamped video recordings to compare reporting and decision-making efficiency with and without AI assistance. Results: For AI tool features, radiologists appreciated the simple binary classification (normal vs abnormal) and found the heatmap essential to understand what the AI considered abnormal; users were uncertain of how to interpret confidence values. Regarding deployment considerations, radiologists thought the tool would be especially helpful for identifying subtle diagnoses; opinions were mixed on whether the tool impacted perceived efficiency, accuracy, and confidence. Considering general human-AI interactions, radiologists shared concerns about automation bias especially when relying on an automated triage function. Regarding decision-making and workflow efficiency, participants began dictating 5 seconds later (42% increase, P = .02) and took 14 seconds longer to complete cases (33% increase, P = .09) with AI assistance. Conclusions: Radiologist usability testing provided insights into effective AI tool features, deployment considerations, and human-AI interactions that can guide successful AI deployment. Early AI adoption may increase radiologists' decision-making and total reporting time but improves with experience.

Higher agreement between readers with deep learning CAD software for reporting pulmonary nodules on CT

Purpose

The aim was to evaluate the impact of CAD software on the pulmonary nodule management recommendations of radiologists in a cohort of patients with incidentally detected nodules on CT.

Methods

For this retrospective study, two radiologists independently assessed 50 chest CT cases for pulmonary nodules to determine the appropriate management recommendation, twice, unaided and aided by CAD with a 6-month washout period. Management recommendations were given in a 4-point grade based on the BTS guidelines. Both reading sessions were recorded to determine the reading times per case. A reduction in reading times per session was tested with a one-tailed paired t-test, and a linear weighted kappa was calculated to assess interobserver agreement.

Results

The mean age of the included patients was 65.0 ± 10.9. Twenty patients were male (40 %). For both readers 1 and 2, a significant reduction of reading time was observed of 33.4 % and 42.6 % (p < 0.001, p < 0.001). The linear weighted kappa between readers unaided was 0.61. Readers showed a better agreement with the aid of CAD, namely by a kappa of 0.84. The mean reading time per case was 226.4 ± 113.2 and 320.8 ± 164.2 s unaided and 150.8 ± 74.2 and 184.2 ± 125.3 s aided by CAD software for readers 1 and 2, respectively.

Conclusion

A dedicated CAD system for aiding in pulmonary nodule reporting may help improve the uniformity of management recommendations in clinical practice.

Impact of Concurrent Use of Artificial Intelligence Tools on Radiologists Reading Time: A Prospective Feasibility Study

RATIONAL AND OBJECTIVES

This study investigated how an AI tool impacted radiologists reading time for non-contrast chest CT exams.

MATERIALS AND METHODS

An AI tool was implemented into the PACS reading workflow of non-contrast chest CT exams between April and May 2020. The reading time was recorded for one CONSULTANT RADIOLOGIST and one RADIOLOGY RESIDENT by an external observer. After each case radiologists answered questions regarding additional findings and perceived case overview. Reading times were recorded for 25 cases without and 20 cases with AI tool assistance for each reader. Differences in reading time with and without the AI tool were assessed using Welch's t-test for non-inferiority with non-inferiority limits defined as 100 seconds for the consultant and 200 seconds for the resident.

RESULTS

The mean reading time for the radiology resident was not significantly affected by the AI tool (without AI 370s vs with AI 437s; +67s 95% CI -28s to +163s, p = 0.16). The reading time for the radiology consultant was also not significantly affected by the AI tool (without AI 366s vs with AI 380s; +13s (95% CI - -57s to 84s, p = 0.70). The AI tool led to additional actionable findings in 5/40 (12.5%) studies and better overview in 18/20 (90%) of studies for the resident.

CONCLUSION

A PACS based implementation of an AI tool for concurrent reading of chest CT exams did not increase reading time with additional actionable findings made as well as a perceived better case overview for the radiology resident.

Assessing the predictive accuracy of lung cancer, metastases, and benign lesions using an artificial intelligence-driven computer aided diagnosis system

Background

Artificial intelligence (AI) products have been widely used for the clinical detection of primary lung tumors. However, their performance and accuracy in risk prediction for metastases or benign lesions remain underexplored. This study evaluated the accuracy of an AI-driven commercial computer-aided detection (CAD) product (InferRead CT Lung Research, ICLR) in malignancy risk prediction using a real-world database.

Methods

This retrospective study assessed 486 consecutive resected lung lesions, including 320 adenocarcinomas, 40 other malignancies, 55 metastases, and 71 benign lesions, from September 2015 to November 2018. The malignancy risk probability of each lesion was obtained using the ICLR software based on a 3D convolutional neural network (CNN) with DenseNet architecture as a backbone (without clinical data). Two resident doctors independently graded each lesion using patient clinical history. One doctor (R1) has 3 years of chest radiology experience, and the other doctor (R2) has 3 years of general radiology experience. Cochran's Q test was used to assess the performances of the AI compared to the radiologists.

Results

The accuracy of malignancy-risk prediction using the ICLR for adenocarcinomas, other malignancies, metastases, and benign lesions was 93.4% (299/320), 95.0% (38/40), 50.9% (28/55), and 40.8% (29/71), respectively. The accuracy was significantly higher in adenocarcinomas and other malignancies compared to metastases and benign lesions (all P<0.05). The overall accuracy of risk prediction for R1 was 93.6% (455/486) and 87.4% for R2 (425/486), both of which were higher than the 81.1% accuracy obtained with the ICLR (394/486) (R1 vs. ICLR: P<0.001; R2 vs. ICLR: P=0.001), especially in assessing the risk of metastases (P<0.05). R1 performed better than R2 at risk prediction (P=0.001).

Conclusions

The accuracy of the ICLR for risk prediction is very high for primary lung cancers but poor for metastases and benign lesions.

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.

Association of Clinician Diagnostic Performance With Machine Learning-Based Decision Support Systems: A Systematic Review

IMPORTANCE

An increasing number of machine learning (ML)-based clinical decision support systems (CDSSs) are described in the medical literature, but this research focuses almost entirely on comparing CDSS directly with clinicians (human vs computer). Little is known about the outcomes of these systems when used as adjuncts to human decision-making (human vs human with computer).

OBJECTIVES

To conduct a systematic review to investigate the association between the interactive use of ML-based diagnostic CDSSs and clinician performance and to examine the extent of the CDSSs' human factors evaluation.

EVIDENCE REVIEW

A search of MEDLINE, Embase, PsycINFO, and grey literature was conducted for the period between January 1, 2010, and May 31, 2019. Peer-reviewed studies published in English comparing human clinician performance with and without interactive use of an ML-based diagnostic CDSSs were included. All metrics used to assess human performance were considered as outcomes. The risk of bias was assessed using Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) and Risk of Bias in Non-Randomised Studies-Intervention (ROBINS-I). Narrative summaries were produced for the main outcomes. Given the heterogeneity of medical conditions, outcomes of interest, and evaluation metrics, no meta-analysis was performed.

FINDINGS

A total of 8112 studies were initially retrieved and 5154 abstracts were screened; of these, 37 studies met the inclusion criteria. The median number of participating clinicians was 4 (interquartile range, 3-8). Of the 107 results that reported statistical significance, 54 (50%) were increased by the use of CDSSs, 4 (4%) were decreased, and 49 (46%) showed no change or an unclear change. In the subgroup of studies carried out in representative clinical settings, no association between the use of ML-based diagnostic CDSSs and improved clinician performance could be observed. Interobserver agreement was the commonly reported outcome whose change was the most strongly associated with CDSS use. Four studies (11%) reported on user feedback, and, in all but 1 case, clinicians decided to override at least some of the algorithms' recommendations. Twenty-eight studies (76%) were rated as having a high risk of bias in at least 1 of the 4 QUADAS-2 core domains, and 6 studies (16%) were considered to be at serious or critical risk of bias using ROBINS-I.

CONCLUSIONS AND RELEVANCE

This systematic review found only sparse evidence that the use of ML-based CDSSs is associated with improved clinician diagnostic performance. Most studies had a low number of participants, were at high or unclear risk of bias, and showed little or no consideration for human factors. Caution should be exercised when estimating the current potential of ML to improve human diagnostic performance, and more comprehensive evaluation should be conducted before deploying ML-based CDSSs in clinical settings. The results highlight the importance of considering supported human decisions as end points rather than merely the stand-alone CDSSs outputs.

COVIDiag: a clinical CAD system to diagnose COVID-19 pneumonia based on CT findings

Objectives

CT findings of COVID-19 look similar to other atypical and viral (non-COVID-19) pneumonia diseases. This study proposes a clinical computer-aided diagnosis (CAD) system using CT features to automatically discriminate COVID-19 from non-COVID-19 pneumonia patients.

Methods

Overall, 612 patients (306 COVID-19 and 306 non-COVID-19 pneumonia) were recruited. Twenty radiological features were extracted from CT images to evaluate the pattern, location, and distribution of lesions of patients in both groups. All significant CT features were fed in five classifiers namely decision tree, K-nearest neighbor, naïve Bayes, support vector machine, and ensemble to evaluate the best performing CAD system in classifying COVID-19 and non-COVID-19 cases.

Results

Location and distribution pattern of involvement, number of the lesion, ground-glass opacity (GGO) and crazy-paving, consolidation, reticular, bronchial wall thickening, nodule, air bronchogram, cavity, pleural effusion, pleural thickening, and lymphadenopathy are the significant features to classify COVID-19 from non-COVID-19 groups. Our proposed CAD system obtained the sensitivity, specificity, and accuracy of 0.965, 93.54%, 90.32%, and 91.94%, respectively, using ensemble (COVIDiag) classifier.

Conclusions

This study proposed a COVIDiag model obtained promising results using CT radiological routine features. It can be considered an adjunct tool by the radiologists during the current COVID-19 pandemic to make an accurate diagnosis.

Key Points

• Location and distribution of involvement, number of lesions, GGO and crazy-paving, consolidation, reticular, bronchial wall thickening, nodule, air bronchogram, cavity, pleural effusion, pleural thickening, and lymphadenopathy are the significant features between COVID-19 from non-COVID-19 groups.

• The proposed CAD system, COVIDiag, could diagnose COVID-19 pneumonia cases with an AUC of 0.965 (sensitivity = 93.54%; specificity = 90.32%; and accuracy = 91.94%).

• The AUC, sensitivity, specificity, and accuracy obtained by radiologist diagnosis are 0.879, 87.10%, 88.71%, and 87.90%, respectively.

Electronic supplementary material

The online version of this article (10.1007/s00330-020-07087-y) contains supplementary material, which is available to authorized users.

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.

Novel platform for development, training, and validation of computer-assisted detection/diagnosis software

PURPOSE

To build a novel, open-source, purely web-based platform system to address problems in the development and clinical use of computer-assisted detection/diagnosis (CAD) software. The new platform system will replace the existing system for the development and validation of CAD software, Clinical Infrastructure for Radiologic Computation of United Solutions (CIRCUS).

METHODS

In our new system, the two top-level applications visible to users are the web-based image database (CIRCUS DB; database) and the Docker plug-in-based CAD execution platform (CIRCUS CS; clinical server). These applications are built on top of a shared application programming interface server, a three-dimensional image viewer component, and an image repository.

RESULTS

We successfully installed our new system into a Linux server at two clinical sites. A total of 1954 cases were registered in CIRCUS DB. We have been utilizing CIRCUS CS with four Docker-based CAD plug-ins.

CONCLUSIONS

We have successfully built a new version of the CIRCUS system. Our platform was successfully implemented at two clinical sites, and we plan to publish it as an open-source software project.

Research progress of computer aided diagnosis system for pulmonary nodules in CT images

BACKGROUND AND OBJECTIVE

Since CAD (Computer Aided Diagnosis) system can make it easier and more efficient to interpret CT (Computer Tomography) images, it has gained much attention and developed rapidly in recent years. This article reviews recent CAD techniques for pulmonary nodule detection and diagnosis in CT Images.

METHODS

CAD systems can be classified into computer-aided detection (CADe) and computer-aided diagnosis (CADx) systems. This review reports recent researches of both systems, including the database, technique, innovation and experimental results of each work. Multi-task CAD systems, which can handle segmentation, false positive reduction, malignancy prediction and other tasks at the same time. The commercial CAD systems are also briefly introduced.

RESULTS

We have found that deep learning based CAD is the mainstream of current research. The reported sensitivity of deep learning based CADe systems ranged between 80.06% and 94.1% with an average 4.3 false-positive (FP) per scan when using LIDC-IDRI dataset, and between 94.4% and 97.9% with an average 4 FP/scan when using LUNA16 dataset, respectively. The overall accuracy of deep learning based CADx systems ranged between 86.84% and 92.3% with an average AUC of 0.956 reported when using LIDC-IDRI dataset.

CONCLUSIONS

We summarized the current tendency and limitations as well as future challenges in this field. The development of CAD needs to meet the rigid clinical requirements, such as high accuracy, strong robustness, high efficiency, fine-grained analysis and classification, and to provide practical clinical functions. This review provides helpful information for both engineering researchers and radiologists to learn the latest development of CAD systems.

Artificial intelligence applications for thoracic imaging

Artificial intelligence is a hot topic in medical imaging. The development of deep learning methods and in particular the use of convolutional neural networks (CNNs), have led to substantial performance gain over the classic machine learning techniques. Multiple usages are currently being evaluated, especially for thoracic imaging, such as such as lung nodule evaluation, tuberculosis or pneumonia detection or quantification of diffuse lung diseases. Chest radiography is a near perfect domain for the development of deep learning algorithms for automatic interpretation, requiring large annotated datasets, in view of the high number of procedures and increasing data availability. Current algorithms are able to detect up to 14 common anomalies, when present as isolated findings. Chest computed tomography is another major field of application for artificial intelligence, especially in the perspective of large scale lung cancer screening. It is important for radiologists to apprehend, contribute actively and lead this new era of radiology powered by artificial intelligence. Such a perspective requires understanding new terms and concepts associated with machine learning. The objective of this paper is to provide useful definitions for understanding the methods used and their possibilities, and report current and future developments for thoracic imaging. Prospective validation of AI tools will be required before reaching routine clinical implementation.

Evaluation of an AI-Powered Lung Nodule Algorithm for Detection and 3D Segmentation of Primary Lung Tumors

Automated detection and segmentation is a prerequisite for the deployment of image-based secondary analyses, especially for lung tumors. However, currently only applications for lung nodules ≤3 cm exist. Therefore, we tested the performance of a fully automated AI-based lung nodule algorithm for detection and 3D segmentation of primary lung tumors in the context of tumor staging using the CT component of FDG-PET/CT and including all T-categories (T1-T4). FDG-PET/CTs of 320 patients with histologically confirmed lung cancer performed between 01/2010 and 06/2016 were selected. First, the main primary lung tumor within each scan was manually segmented using the CT component of the PET/CTs as reference. Second, the CT series were transferred to a platform with AI-based algorithms trained on chest CTs for detection and segmentation of lung nodules. Detection and segmentation performance were analyzed. Factors influencing detection rates were explored with binominal logistic regression and radiomic analysis. We also processed 94 PET/CTs negative for pulmonary nodules to investigate frequency and reasons of false-positive findings. The ratio of detected tumors was best in the T1-category (90.4%) and decreased continuously: T2 (70.8%), T3 (29.4%), and T4 (8.8%). Tumor contact with the pleura was a strong predictor of misdetection. Segmentation performance was excellent for T1 tumors (r = 0.908, p < 0.001) and tumors without pleural contact (r = 0.971, p < 0.001). Volumes of larger tumors were systematically underestimated. There were 0.41 false-positive findings per exam. The algorithm tested facilitates a reliable detection and 3D segmentation of T1/T2 lung tumors on FDG-PET/CTs. The detection and segmentation of more advanced lung tumors is currently imprecise due to the conception of the algorithm for lung nodules <3 cm. Future efforts should therefore focus on this collective to facilitate segmentation of all tumor types and sizes to bridge the gap between CAD applications for screening and staging of lung cancer.