Image analysis and machine learning applied to breast cancer diagnosis and prognosis.

Fine needle aspiration (FNA) accuracy is limited by, among other factors, the subjective interpretation of the aspirate. We have increased breast FNA accuracy by coupling digital image analysis methods with machine learning techniques. Additionally, our mathematical approach captures nuclear features ("grade") that are prognostically more accurate than are estimates based on tumor size and lymph node status. An interactive computer system evaluates, diagnoses and determines prognosis based on nuclear features derived directly from a digital scan of FNA slides. A consecutive series of 569 patients provided the data for the diagnostic study. A 166-patient subset provided the data for the prognostic study. An additional 75 consecutive, new patients provided samples to test the diagnostic system. The projected prospective accuracy of the diagnostic system was estimated to be 97% by 10-fold cross-validation, and the actual accuracy on 75 new samples was 100%. The projected prospective accuracy of the prognostic system was estimated to be 86% by leave-one-out testing.

[1]  M. Black,et al.  Survival in breast cancer cases in relation to the structure of the primary tumor and regional lymph nodes. , 1955, Surgery, gynecology & obstetrics.

[2]  Black Mm,et al.  Nuclear structure in cancer tissues. , 1957 .

[3]  E. Kaplan,et al.  Nonparametric Estimation from Incomplete Observations , 1958 .

[4]  E. Parzen.  On Estimation of a Probability Density Function and Mode , 1962 .

[5]  M. R. Mickey,et al.  Estimation of Error Rates in Discriminant Analysis , 1968 .

[6]  Olvi L. Mangasarian,et al.  Multisurface method of pattern separation , 1968, IEEE Trans. Inf. Theory.

[7]  M. Stone.  Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[8]  A. Wallgren,et al.  The prognostic value of the aspiration biopsy smear in mammary carcinoma. , 1976, Acta cytologica.

[9]  J. Vegelius,et al.  Analysis of reproducibility of subjective grading systems for breast carcinoma. , 1979, Journal of clinical pathology.

[10]  M. Boon,et al.  Prognostic indicators in breast cancer‐morphometric methods , 1982, Histopathology.

[11]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[12]  J. Baak,et al.  The value of morphometry to classic prognosticators in breast cancer , 1985, Cancer.

[13]  C Wittekind,et al.  Computerized morphometric image analysis of cytologic nuclear parameters in breast cancer. , 1987, Analytical and quantitative cytology and histology.

[14]  B. V. Pedersen,et al.  Histologic malignancy grading of invasive ductal breast carcinoma. A regression analysis of prognostic factors in low‐risk carcinomas from a multicenter trial , 1987, Cancer.

[15]  I. O. Ellis,et al.  Confirmation of a prognostic index in primary breast cancer. , 1987, British Journal of Cancer.

[16]  Donald E. Henson,et al.  Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases , 1989 .

[17]  Mubarak Shah,et al.  A fast algorithm for active contours , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[18]  D Komitowski,et al.  Quantitative features of chromatin structure in the prognosis of breast cancer , 1990, Cancer.

[19]  C. Redmond,et al.  Pathologic findings from the national surgical adjuvant breast and bowel projects (NSABP) prognostic discriminants for 8‐year survival for node‐negative invasive breast cancer patients , 1990, Cancer.

[20]  Sholom M. Weiss,et al.  Computer Systems That Learn , 1990 .

[21]  L S Freedman,et al.  Relationship among outcome, stage of disease, and histologic grade for 22,616 cases of breast cancer. The basis for a prognostic index , 1991, Cancer.

[22]  K. Pienta,et al.  Correlation of nuclear morphometry with progression of breast cancer , 1991, Cancer.

[23]  I. Mittra,et al.  A meta-analysis of reported correlations between prognostic factors in breast cancer: does axillary lymph node metastasis represent biology or chronology? , 1991, European journal of cancer.

[24]  J. Hermans,et al.  The value of aspiration cytologic examination of the breast a statistical review of the medical literature , 1992, Cancer.

[25]  Kristin P. Bennett,et al.  Decision Tree Construction Via Linear Programming , 1992 .

[26]  Olvi L. Mangasarian,et al.  Mathematical Programming in Neural Networks , 1993, INFORMS J. Comput..

[27]  Olvi L. Mangasarian,et al.  Nuclear feature extraction for breast tumor diagnosis , 1993, Electronic Imaging.

[28]  W. N. Street,et al.  Breast cytology diagnosis with digital image analysis. , 1993, Analytical and quantitative cytology and histology.