ROC analysis of detection of metastatic pulmonary nodules on digital chest radiographs with temporal subtraction.

RATIONALE AND OBJECTIVES The authors' purpose was to evaluate the effect of temporal subtraction on digital chest radiographs in the detection of metastatic pulmonary nodules. MATERIALS AND METHODS The study included 21 cases with metastatic pulmonary nodule and 21 cases without metastatic nodule. Eleven radiologists, including eight residents and three certified radiologists, provided their confidence levels for the presence or absence of pulmonary nodules without and with temporal subtraction. Their performances without and with temporal subtraction were evaluated by means of receiver operating characteristic analysis with both independent and sequential tests. RESULTS For the independent test, the radiologists' Az (area under the receiver operating characteristic curve) values were 0.871 without and 0.954 with temporal subtraction, compared with 0.882 and 0.955, respectively, for the sequential test. Diagnosis accuracy was significantly improved with the use of temporal subtraction. There was no significant difference in Az values between the independent and sequential tests. CONCLUSION Temporal subtraction is useful in the detection of metastatic pulmonary nodules, and this technique augments the value of digital chest radiography.

[1]  K. Doi,et al.  Iterative image warping technique for temporal subtraction of sequential chest radiographs to detect interval change. , 1999, Medical physics.

[2]  H MacMahon,et al.  Computer-aided diagnosis in chest radiology. , 1990, Journal of thoracic imaging.

[3]  M L Giger,et al.  Computer-aided diagnosis: development of automated schemes for quantitative analysis of radiographic images. , 1992, Seminars in ultrasound, CT, and MR.

[4]  H. Libshitz,et al.  Pulmonary metastatic disease: radiologic-surgical correlation. , 1987, Radiology.

[5]  K Doi,et al.  Digital chest radiography: effect of temporal subtraction images on detection accuracy. , 1997, Radiology.

[6]  N. Petrick,et al.  Improvement of radiologists' characterization of mammographic masses by using computer-aided diagnosis: an ROC study. , 1999, Radiology.

[7]  M. J. Carreira,et al.  Computer-aided diagnoses: automatic detection of lung nodules. , 1998, Medical physics.

[8]  K Doi,et al.  Improvement in radiologists' detection of clustered microcalcifications on mammograms. The potential of computer-aided diagnosis. , 1990, Investigative radiology.

[9]  K. Doi,et al.  Effect of a computer-aided diagnosis scheme on radiologists' performance in detection of lung nodules on radiographs. , 1996, Radiology.

[10]  M. Giger,et al.  Digital image subtraction of temporally sequential chest images for detection of interval change. , 1994, Medical physics.

[11]  M. Giger,et al.  Development of an improved CAD scheme for automated detection of lung nodules in digital chest images. , 1997, Medical physics.

[12]  N. Abecasis,et al.  Surgical treatment of lung metastases: Prognostic factors for long‐term survival , 1999, Journal of surgical oncology.

[13]  M. Giger,et al.  Image feature analysis and computer-aided diagnosis in digital radiography. 3. Automated detection of nodules in peripheral lung fields. , 1988, Medical physics.