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

Li, Li; Liu, Zhou; Huang, Hua; Meng, Lin; Luo, Dehong

doi:10.1111/1759-7714.12931

Cited by 54 publications

(49 citation statements)

References 40 publications

(74 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Five studies [17,19,20,22,25] had results on both classification and detection and tested on local, independently obtained datasets. While all the studies tested a CNN architecture, Tajbakhsh and Suzuki [20] tested both CNN-and MTANN-based algorithms.…”

Section: Both Detection and Classification (7 Studies)mentioning

confidence: 99%

“…While all the studies tested a CNN architecture, Tajbakhsh and Suzuki [20] tested both CNN-and MTANN-based algorithms. Three of the studies [17,19,22] measured detection performance using sensitivity and they reached levels between 86.2-97% (Table 1). Tajbakhsh and Suzuki [20] collected information of false positives when 100% sensitivity was achieved with MTANN and CNN, which resulted in 2.7 and 22.7 false positives per patient, respectively.…”

Section: Both Detection and Classification (7 Studies)mentioning

confidence: 99%

“…Chen et al [22] achieved an overall classification accuracy of 87.5% when classifying adenocarcinomas and benign nodules, and Suzuki [19] achieved 96% sensitivity when classifying malign nodules (Table 2). Li et al [17] tested the performance of characterizing nodules into three pulmonary nodule categories: solid (sensitivity: 90.3%; specificity: 100%), part-solid (sensitivity: 55.5%; specificity: 93%), and ground glass types (sensitivity: 100%; specificity: 96.1%).…”

Section: Both Detection and Classification (7 Studies)mentioning

confidence: 99%

“…On detection, two studies [17,19] tested and trained on different types of datasets and achieved sensitivities of 86.2% and 97.0%, while the studies that trained and tested on the same types of dataset [20][21][22][23]25] had sensitivities between 74.6-97% (Table 1). On classification, the two studies [17,19] that trained and tested on different types of dataset achieved sensitivities of 96% and 100%, and the studies that trained and tested on the same types of dataset [20][21][22][23]25] achieved sensitivities between 76.5-83.7% and accuracies between 81.4-96.3% (Tables 2 and 3).…”

Section: Both Detection and Classification (7 Studies)mentioning

confidence: 99%

“…Most studies have trained and tested their algorithms on the large and publicly available Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) dataset, which makes the studies homogenous [16]. Few studies have tested their algorithms on datasets not from LIDC-IDRI, and only a subgroup of those have trained their algorithms on datasets that were not obtained the same way as the final test data [17,18].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

The Performance of Deep Learning Algorithms on Automatic Pulmonary Nodule Detection and Classification Tested on Different Datasets That Are Not Derived from LIDC-IDRI: A Systematic Review

Vilmun

Carlsen

et al. 2019

Diagnostics

View full text Add to dashboard Cite

The aim of this study was to systematically review the performance of deep learning technology in detecting and classifying pulmonary nodules on computed tomography (CT) scans that were not from the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) database. Furthermore, we explored the difference in performance when the deep learning technology was applied to test datasets different from the training datasets. Only peer-reviewed, original research articles utilizing deep learning technology were included in this study, and only results from testing on datasets other than the LIDC-IDRI were included. We searched a total of six databases: EMBASE, PubMed, Cochrane Library, the Institute of Electrical and Electronics Engineers, Inc. (IEEE), Scopus, and Web of Science. This resulted in 1782 studies after duplicates were removed, and a total of 26 studies were included in this systematic review. Three studies explored the performance of pulmonary nodule detection only, 16 studies explored the performance of pulmonary nodule classification only, and 7 studies had reports of both pulmonary nodule detection and classification. Three different deep learning architectures were mentioned amongst the included studies: convolutional neural network (CNN), massive training artificial neural network (MTANN), and deep stacked denoising autoencoder extreme learning machine (SDAE-ELM). The studies reached a classification accuracy between 68–99.6% and a detection accuracy between 80.6–94%. Performance of deep learning technology in studies using different test and training datasets was comparable to studies using same type of test and training datasets. In conclusion, deep learning was able to achieve high levels of accuracy, sensitivity, and/or specificity in detecting and/or classifying nodules when applied to pulmonary CT scans not from the LIDC-IDRI database.

show abstract