The goal of the proposed research is to develop computer-aided diagnosis (CAD) schemes in order to improve the diagnostic accuracy of breast cancer in mammography.
Three specific aims are included: (1) development of a computer program for the detection and characterization of microcalcifications, (2) development of a computer program for characterization of masses, and (3) investigation of digitization requirements for CAD schemes and digital mammographic systems. The proposed CAD schemes will aid radiologists in screening mammograms for suspicious lesions, thereby reducing the miss rate due to human errors. The study of digitization requirements will provide information for practical implementation of the CAD schemes and for development of digital mammographic systems. Initially, a data base of clinical mammograms which include malignant and benign microcalcifications and masses will be established. Physical measures which characterize the significant image features of the lesions will be developed. Based on these measures, a statistical classifier will be designed for estimation of the likelihood of malignancy for each type of lesions. For automated detection and classification of microcalcifications, effective spatial filters will be developed for enhancement of the signal- to-noise ratio (SNR) of the microcalcifications. Signal-extraction techniques will then be designed to isolate the microcalcifications from the background. Physical characteristics such as size, shape, frequency spectrum, spatial distribution, and clustering properties will be determined and analyzed with the classifier. An observer performance study using receiver operating characteristic (ROC) methodology will be conducted to evaluate the effects of the CAD scheme on radiologists' performance. For the classification of mass, automated analysis will be performed in user-selected regions that include suspected masses. Data compression, noise smoothing, edge enhancement, and structured background correction methods will be developed for detection of the mass boundary. Physical characteristics such as size, density, edge sharpness, calcifications, shape, lobulation, and spiculation will be extracted from the mass and analyzed with the classifier. An ROC study will be conducted to evaluate the effect of CAD on radiologists' performance in differentiation of malignant or benign masses. The effects of spatial resolution and grey-level resolution of digitization on detection of subtle microcalcifications by computer or human observer will be investigated using a high-resolution film digitizer. The SNR enhancement and signal-extraction strategies for computer detection will be optimized for each digitization condition and the detection accuracy will be compared. Observer performance study will be conducted to evaluate the visual detectability of subtle microcalcifications digitized at various resolutions. The detectability will be correlated with the resolution and noise properties of the digitizer.

Agency
National Institute of Health (NIH)
Institute
National Cancer Institute (NCI)
Type
Research Project (R01)
Project #
2R01CA048129-06
Application #
3192121
Study Section
Diagnostic Radiology Study Section (RNM)
Project Start
1989-05-01
Project End
1996-04-30
Budget Start
1992-06-01
Budget End
1993-04-30
Support Year
6
Fiscal Year
1992
Total Cost
Indirect Cost
Name
University of Michigan Ann Arbor
Department
Type
Schools of Medicine
DUNS #
791277940
City
Ann Arbor
State
MI
Country
United States
Zip Code
48109
Helvie, Mark A; Hadjiiski, Lubomir; Makariou, Erini et al. (2004) Sensitivity of noncommercial computer-aided detection system for mammographic breast cancer detection: pilot clinical trial. Radiology 231:208-14
Gurcan, Metin N; Chan, Heang-Ping; Sahiner, Berkman et al. (2002) Optimal neural network architecture selection: improvement in computerized detection of microcalcifications. Acad Radiol 9:420-9
Paquerault, Sophie; Petrick, Nicholas; Chan, Heang-Ping et al. (2002) Improvement of computerized mass detection on mammograms: fusion of two-view information. Med Phys 29:238-47
Petrick, Nicholas; Sahiner, Berkman; Chan, Heang-Ping et al. (2002) Breast cancer detection: evaluation of a mass-detection algorithm for computer-aided diagnosis -- experience in 263 patients. Radiology 224:217-24
Sahiner, B; Petrick, N; Chan, H P et al. (2001) Computer-aided characterization of mammographic masses: accuracy of mass segmentation and its effects on characterization. IEEE Trans Med Imaging 20:1275-84
Sahiner, B; Chan, H P; Petrick, N et al. (2001) Improvement of mammographic mass characterization using spiculation meausures and morphological features. Med Phys 28:1455-65
Hadjiiski, L; Sahiner, B; Chan, H P et al. (2001) Analysis of temporal changes of mammographic features: computer-aided classification of malignant and benign breast masses. Med Phys 28:2309-17
Gurcan, M N; Sahiner, B; Chan, H P et al. (2001) Selection of an optimal neural network architecture for computer-aided detection of microcalcifications--comparison of automated optimization techniques. Med Phys 28:1937-48
Hadjiiski, L; Chan, H P; Sahiner, B et al. (2001) Automated registration of breast lesions in temporal pairs of mammograms for interval change analysis--local affine transformation for improved localization. Med Phys 28:1070-9
Chan, H P; Helvie, M A; Petrick, N et al. (2001) Digital mammography: observer performance study of the effects of pixel size on the characterization of malignant and benign microcalcifications. Acad Radiol 8:454-66

Showing the most recent 10 out of 36 publications