Machine learning for lung CT texture analysis: Improvement of inter-observer agreement for radiological finding classification in patients with pulmonary diseases. Issue 134 (January 2021)

• Record Type: Journal Article
• Authors: Ohno, Yoshiharu; Aoyagi, Kota; Takenaka, Daisuke; Yoshikawa, Takeshi; Ikezaki, Aina; Fujisawa, Yasuko; Murayama, Kazuhiro; Hattori, Hidekazu; Toyama, Hiroshi
• Is Part Of: European journal of radiology. Issue 134(2021)
• Publication Date: 2021-01
• Subjects: Lung -- CT -- COPD -- Interstitial lung disease -- Connective tissue disease; Medical radiology -- Periodicals; Radiology -- Periodicals; Radiologie médicale -- Périodiques; Medical radiology; Periodicals; 616.075705
• DOI: 10.1016/j.ejrad.2020.109410
• Languages: English
• ISSNs: 0720-048X

Abstract

Highlights: ML-based software can improve agreement with standard reference in not only each reader (p < 0.0001), but also consensus reading (p < 0.0001). Accuracy by consensus with the software was significantly higher than that without the software (p < 0.0001) and the software alone (p < 0.0001). ML-based CT texture analysis software has potential to play as second reader for radiological finding classification in pulmonary diseases. Abstract: Purpose: To evaluate the capability ML-based CT texture analysis for improving interobserver agreement and accuracy of radiological finding assessment in patients with COPD, interstitial lung diseases or infectious diseases. Materials and methods: Training cases (n = 28), validation cases (n = 17) and test cases (n = 89) who underwent thin-section CT at a 320-detector row CT with wide volume scan and two 64-detector row CTs with helical scan were enrolled in this study. From 89 CT data, a total of 350 computationally selected ROI including normal lung, emphysema, nodular lesion, ground-glass opacity, reticulation and honeycomb were evaluated by three radiologists as well as by the software. Inter-observer agreements between consensus reading with and without using the software or software alone and standard references determined by consensus of pulmonologists and chest radiologists were determined using κ statistics. Overall distinguishing accuracies were compared among all methods by McNemar's test. Results: Agreements for consensus readings obtained with and without the software or the software alone with standard references were determined as significant and substantial or excellent (with the software: κ = 0.91, p < 0.0001; without the software: κ = 0.81, p < 0.0001; the software alone: κ = 0.79, p < 0.0001). Overall differentiation accuracy of consensus reading using the software (94.9 [332/350] %) was significantly higher than that of consensus reading without using the software (84.3 [295/350] %, p < 0.0001) and the software alone (82.3 [288/350] %, p < 0.0001). Conclusion: ML-based CT texture analysis software has potential for improving interobserver agreement and accuracy for radiological finding assessments in patients with COPD, interstitial lung diseases or infectious diseases.