Automated Interstitial Lung Abnormality Probability Prediction at CT: A Stepwise Machine Learning Approach in the Boston Lung Cancer Study.

Journal: Radiology
PMID: 39225600

Abstract

Background It is increasingly recognized that interstitial lung abnormalities (ILAs) detected at CT have potential clinical implications, but automated identification of ILAs has not yet been fully established. Purpose To develop and test automated ILA probability prediction models using machine learning techniques on CT images. Materials and Methods This secondary analysis of a retrospective study included CT scans from patients in the Boston Lung Cancer Study collected between February 2004 and June 2017. Visual assessment of ILAs by two radiologists and a pulmonologist served as the ground truth. Automated ILA probability prediction models were developed that used a stepwise approach involving section inference and case inference models. The section inference model produced an ILA probability for each CT section, and the case inference model integrated these probabilities to generate the case-level ILA probability. For indeterminate sections and cases, both two- and three-label methods were evaluated. For the case inference model, we tested three machine learning classifiers (support vector machine [SVM], random forest [RF], and convolutional neural network [CNN]). Receiver operating characteristic analysis was performed to calculate the area under the receiver operating characteristic curve (AUC). Results A total of 1382 CT scans (mean patient age, 67 years ± 11 [SD]; 759 women) were included. Of the 1382 CT scans, 104 (8%) were assessed as having ILA, 492 (36%) as indeterminate for ILA, and 786 (57%) as without ILA according to ground-truth labeling. The cohort was divided into a training set ( = 96; ILA, = 48), a validation set ( = 24; ILA, = 12), and a test set ( = 1262; ILA, = 44). Among the models evaluated (two- and three-label section inference models; two- and three-label SVM, RF, and CNN case inference models), the model using the three-label method in the section inference model and the two-label method and RF in the case inference model achieved the highest AUC, at 0.87. Conclusion The model demonstrated substantial performance in estimating ILA probability, indicating its potential utility in clinical settings. © RSNA, 2024 See also the editorial by Zagurovskaya in this issue.

Authors

  • Akinori Hata
    Department of Radiology, Osaka University Graduate School of Medicine.
  • Kota Aoyagi
    From the Department of Radiology, Fujita Health University School of Medicine, 1-98 Dengakugakubo, Kutsukake-cho, Toyoake 470-1192, Japan (Y.O.); Joint Research Laboratory of Advanced Medical Imaging, Fujita Health University School of Medicine, Toyoake, Japan (Y.O.); Division of Functional and Diagnostic Imaging Research, Department of Radiology, Kobe University Graduate School of Medicine, Kobe, Japan (Y.O., S.S., T.Y.); Canon Medical Systems, Otawara, Japan (K.A.); Corporate Research and Development Center, Toshiba, Kawasaki, Japan (A.Y.); Division of Radiology, Department of Radiology, Kobe University Graduate School of Medicine, Kobe, Japan (Y.U., Y.K.); Department of Radiology, Kohnan Hospital, Kobe, Japan (Y.K.); and Department of Radiology, Hyogo Cancer Center, Akashi, Japan (D.T.).
  • Takuya Hino
    From the Center for Pulmonary Functional Imaging, Department of Radiology (A.H., T.H., N.W., V.I.V., M. Nishino, H.H.), and Pulmonary and Critical Care Division (G.M.H.), Brigham and Women's Hospital and Harvard Medical School, 75 Francis St, Boston, MA 02115; Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Osaka, Japan (A.H., N.T.); Canon Medical Systems, Tochigi, Japan (K.A., Y.M., M. Nakatsugawa, A.K., N.S., M.O.); Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan (T.H., N.W.); R&D Headquarters, Canon, Tokyo, Japan (M.K.); Department of Biostatistics, University of Michigan, Ann Arbor, Mich (J.S., Y.L.); Departments of Biostatistics (X.W., D.C.C.) and Environmental Health (D.C.C.), Harvard T.H. Chan School of Public Health, Boston, Mass; and Department of Imaging, Dana Farber Cancer Institute, Boston, Mass (M. Nishino).
  • Masami Kawagishi
    From the Center for Pulmonary Functional Imaging, Department of Radiology (A.H., T.H., N.W., V.I.V., M. Nishino, H.H.), and Pulmonary and Critical Care Division (G.M.H.), Brigham and Women's Hospital and Harvard Medical School, 75 Francis St, Boston, MA 02115; Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Osaka, Japan (A.H., N.T.); Canon Medical Systems, Tochigi, Japan (K.A., Y.M., M. Nakatsugawa, A.K., N.S., M.O.); Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan (T.H., N.W.); R&D Headquarters, Canon, Tokyo, Japan (M.K.); Department of Biostatistics, University of Michigan, Ann Arbor, Mich (J.S., Y.L.); Departments of Biostatistics (X.W., D.C.C.) and Environmental Health (D.C.C.), Harvard T.H. Chan School of Public Health, Boston, Mass; and Department of Imaging, Dana Farber Cancer Institute, Boston, Mass (M. Nishino).
  • Noriaki Wada
    Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan.
  • Jiyeon Song
    From the Center for Pulmonary Functional Imaging, Department of Radiology (A.H., T.H., N.W., V.I.V., M. Nishino, H.H.), and Pulmonary and Critical Care Division (G.M.H.), Brigham and Women's Hospital and Harvard Medical School, 75 Francis St, Boston, MA 02115; Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Osaka, Japan (A.H., N.T.); Canon Medical Systems, Tochigi, Japan (K.A., Y.M., M. Nakatsugawa, A.K., N.S., M.O.); Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan (T.H., N.W.); R&D Headquarters, Canon, Tokyo, Japan (M.K.); Department of Biostatistics, University of Michigan, Ann Arbor, Mich (J.S., Y.L.); Departments of Biostatistics (X.W., D.C.C.) and Environmental Health (D.C.C.), Harvard T.H. Chan School of Public Health, Boston, Mass; and Department of Imaging, Dana Farber Cancer Institute, Boston, Mass (M. Nishino).
  • Xinan Wang
    The Key Laboratory of Integrated Microsystems, Peking University Shenzhen Graduate School, Shenzhen 518055, China.
  • Vladimir I Valtchinov
    Center for Evidence-Based Imaging, Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Brookline, Massachusetts; Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts; Department of Biomedical Informatics, Harvard Medical School, Boston, Massachusetts. Electronic address: vvaltchinov@bwh.harvard.edu.
  • Mizuki Nishino
    From the Department of Radiology, Faculty of Medicine, Saga University, 5-1-1 Nabeshima, Saga 849-8501, Japan (R.E.); and Department of Radiology, Brigham and Women's Hospital and Dana-Farber Cancer Institute, Boston, Mass (M.N.).
  • Yohei Muraguchi
    From the Center for Pulmonary Functional Imaging, Department of Radiology (A.H., T.H., N.W., V.I.V., M. Nishino, H.H.), and Pulmonary and Critical Care Division (G.M.H.), Brigham and Women's Hospital and Harvard Medical School, 75 Francis St, Boston, MA 02115; Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Osaka, Japan (A.H., N.T.); Canon Medical Systems, Tochigi, Japan (K.A., Y.M., M. Nakatsugawa, A.K., N.S., M.O.); Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan (T.H., N.W.); R&D Headquarters, Canon, Tokyo, Japan (M.K.); Department of Biostatistics, University of Michigan, Ann Arbor, Mich (J.S., Y.L.); Departments of Biostatistics (X.W., D.C.C.) and Environmental Health (D.C.C.), Harvard T.H. Chan School of Public Health, Boston, Mass; and Department of Imaging, Dana Farber Cancer Institute, Boston, Mass (M. Nishino).
  • Minoru Nakatsugawa
    Canon Medical Systems Corporation.
  • Akihiro Koga
    From the Center for Pulmonary Functional Imaging, Department of Radiology (A.H., T.H., N.W., V.I.V., M. Nishino, H.H.), and Pulmonary and Critical Care Division (G.M.H.), Brigham and Women's Hospital and Harvard Medical School, 75 Francis St, Boston, MA 02115; Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Osaka, Japan (A.H., N.T.); Canon Medical Systems, Tochigi, Japan (K.A., Y.M., M. Nakatsugawa, A.K., N.S., M.O.); Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan (T.H., N.W.); R&D Headquarters, Canon, Tokyo, Japan (M.K.); Department of Biostatistics, University of Michigan, Ann Arbor, Mich (J.S., Y.L.); Departments of Biostatistics (X.W., D.C.C.) and Environmental Health (D.C.C.), Harvard T.H. Chan School of Public Health, Boston, Mass; and Department of Imaging, Dana Farber Cancer Institute, Boston, Mass (M. Nishino).
  • Naoki Sugihara
    From the Center for Pulmonary Functional Imaging, Department of Radiology (A.H., T.H., N.W., V.I.V., M. Nishino, H.H.), and Pulmonary and Critical Care Division (G.M.H.), Brigham and Women's Hospital and Harvard Medical School, 75 Francis St, Boston, MA 02115; Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Osaka, Japan (A.H., N.T.); Canon Medical Systems, Tochigi, Japan (K.A., Y.M., M. Nakatsugawa, A.K., N.S., M.O.); Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan (T.H., N.W.); R&D Headquarters, Canon, Tokyo, Japan (M.K.); Department of Biostatistics, University of Michigan, Ann Arbor, Mich (J.S., Y.L.); Departments of Biostatistics (X.W., D.C.C.) and Environmental Health (D.C.C.), Harvard T.H. Chan School of Public Health, Boston, Mass; and Department of Imaging, Dana Farber Cancer Institute, Boston, Mass (M. Nishino).
  • Masahiro Ozaki
    From the Center for Pulmonary Functional Imaging, Department of Radiology (A.H., T.H., N.W., V.I.V., M. Nishino, H.H.), and Pulmonary and Critical Care Division (G.M.H.), Brigham and Women's Hospital and Harvard Medical School, 75 Francis St, Boston, MA 02115; Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Osaka, Japan (A.H., N.T.); Canon Medical Systems, Tochigi, Japan (K.A., Y.M., M. Nakatsugawa, A.K., N.S., M.O.); Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan (T.H., N.W.); R&D Headquarters, Canon, Tokyo, Japan (M.K.); Department of Biostatistics, University of Michigan, Ann Arbor, Mich (J.S., Y.L.); Departments of Biostatistics (X.W., D.C.C.) and Environmental Health (D.C.C.), Harvard T.H. Chan School of Public Health, Boston, Mass; and Department of Imaging, Dana Farber Cancer Institute, Boston, Mass (M. Nishino).
  • Gary M Hunninghake
    Division of Pulmonary and Critical Care Medicine, Brigham and Women's Hospital, Boston, MA.
  • Noriyuki Tomiyama
    Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Suita, Osaka, Japan.
  • Yi Li
    Wuhan Zoncare Bio-Medical Electronics Co., Ltd, Wuhan, China.
  • David C Christiani
  • Hiroto Hatabu
    Department of Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, MA.

Keywords

External Resources

Popular Topics
Recent Journals