IILS: Intelligent imaging layout system for automatic imaging report standardization and intra-interdisciplinary clinical workflow optimization

Abstract

Background: To achieve imaging report standardization and improve the quality and efficiency of the intrainterdisciplinary clinical workflow, we proposed an intelligent imaging layout system (IILS) for a clinical decision support system-based ubiquitous healthcare service, which is a lung nodule management system using medical images. Methods: We created a lung IILS based on deep learning for imaging report standardization and workflow optimization for the identification of nodules. Our IILS utilized a deep learning plus adaptive auto layout tool, which trained and tested a neural network with imaging data from all the main CT manufacturers from 11,205 patients. Model performance was evaluated by the receiver operating characteristic curve (ROC) and calculating the corresponding area under the curve (AUC). The clinical application value for our IILS was assessed by a comprehensive comparison of multiple aspects. Findings: Our IILS is clinically applicable due to the consistency with nodules detected by IILS, with its highest consistency of 0•94 and an AUC of 90•6% for malignant pulmonary nodules versus benign nodules with a sensitivity of 76•5% and specificity of 89•1%. Applying this IILS to a dataset of chest CT images, we demonstrate performance comparable to that of human experts in providing a better layout and aiding in diagnosis in 100% valid images and nodule display. The IILS was superior to the traditional manual system in performance, such as reducing the number of clicks from 14•45 ± 0•38 to 2, time consumed from 16•87 ± 0•38 s to 6•92 ± 0•10 s, number of invalid images from 7•06 ± 0•24 to 0, and missing lung nodules from 46•8% to 0%. Interpretation: This IILS might achieve imaging report standardization, and improve the clinical workflow therefore opening a new window for clinical application of artificial intelligence. Fund: The National Natural Science Foundation of China.

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

“…[22] 2019 87.5 Ciompi, Francesco et al [29] 2017 79.5 * Jakimovski, Goran et al [30] 2019 99.6 Lakshmanaprabu, S.K. et al [31] 2018 94.5 Liao, Fangzhou et al [23] 2019 81.4 Liu, Xinglong et al [33] 2017 90.3 * Masood, Anum et al [21] 2018 96.3 Nishio, Mizuho et al [34] 2018 68 Onishi, Yuya et al [35] 2018 81.7 Polat, Huseyin et al [36] 2019 91.8 Qiang, Yan et al [37] 2017 82.8 Rangaswamy et al [38] 2019 96 Sori, Worku Jifara et al [39] 2018 87.8 Wang, Shengping et al [40] 2018 84 Wang, Yang et al [25] 2019 87.3 Yuan, Jingjing et al [41] 2017 93.9 * Zhang, Chao et al [42] 2019 92 * (c)…”

“…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.…”

“…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. Detection performance was measured by Wang et al [25] using the kappa consistency coefficient and reached 0.94 when compared to human experts. On classification, four of the above-mentioned studies [19,20,22,25] tested on dichotomous categories.…”

“…Detection performance was measured by Wang et al [25] using the kappa consistency coefficient and reached 0.94 when compared to human experts. On classification, four of the above-mentioned studies [19,20,22,25] tested on dichotomous categories. Two of the studies [20,25] reported AUC values of 77.6% and 90.6%.…”

“…On classification, four of the above-mentioned studies [19,20,22,25] tested on dichotomous categories. Two of the studies [20,25] reported AUC values of 77.6% and 90.6%. 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).…”