PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 8168060

Computerized contextual analysis in diagnosis.

D J Zahniser·L M Isenstein·M L Hutchinson
View original source ↗Export metadata ↓

Abstract

Cervical cytology specimens prepared as monolayers were analyzed with a combination of high-resolution single-cell image analysis and lower-resolution contextual image analysis. For the latter, quantitative features of cluster architecture, object distribution and background were used to mimic the pathologists global assessment of a specimen. Contextual analysis correctly classified 81% of the cases for diagnosis, about on a par with the companion single-cell cytometric analysis. Contextual and single cell analysis complemented each other, yielding greater accuracy when the two methods were combined. Additional applications of contextual analysis are discussed.

Explore related subjects

Discriminant AnalysisFemaleHumansImage Processing, Computer-AssistedUterine Cervical NeoplasmsVaginal SmearsJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

D J Zahniser, L M Isenstein, M L Hutchinson. 1994-03-15. Computerized contextual analysis in diagnosis.. https://doi.org/10.1016/0304-3835(94)90096-5

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed1996

Multivariate statistical analysis for pathologist. Part I, The logistic model.

This paper reviews concepts of multivariate statistical modeling via the logistic regression model, which has become very popular for modeling the relationship between a positive clinical outcome and a variety of predictor variables. The process is illustrated using a composite of data from three large prostate specific antigen based screening studies of prostate cancer.

Discriminant Analysis↗
PubMed1995

Sex determination from the pubis by discriminant function analysis.

A total of 122 adult human pubes (66 males and 56 females) are used in this study of a discriminant function analysis for determining sex from four pubic measurements: the angle formed by the middle line of the superior ramus and inferior ramus of pubis (in degrees, X1), subpubic angle (in degrees, X2), the minimum distance from the symphyseal surface to the obturator (in millimeters, X3), and the minimum thickness of the ischiopubic ramus (in millimeters, X4). The method introduced by Fisher is used. Using the formula Y = 0.270 mean1-0.985mean2-0.761mean3 + 1.806mean4,F = 200 F = 200 (F > F0.01(4,117), P < 0.01), section point -32.47, an accuracy of 100% could be achieved for 230 pubes (130 males, 100 females).

Discriminant Analysis↗
PubMed1995

Clinical value of sperm morphology for in-vivo fertility: comparison between World Health Organization criteria of 1987 and 1992.

The World Health Organization (WHO, 1992) has suggested new criteria for scoring sperm morphology. This study compares the clinical value of the new criteria, i.e. classification of a man as fertile or infertile, to those previously established by the WHO (1987). Papanicolaou-stained semen smears from 166 men attending our infertility clinic, whose fertility status was known, were scored using both methods. Using logistic discriminant analysis for compositional data, no difference between these two sets of criteria with respect to predicting pregnancy outcome was observed. The categorization of the abnormalities (head, midpiece, tail) provides no extra clarification. The WHO (1992) cut-off point of 30% for normal forms is not appropriate, as approximately half of the men in the fertile group had a normal sperm morphology below this limit. In conclusion, the present WHO (1992) classification of sperm morphology is of no additional clinical value. Studies on sperm morphology should concentrate on obtaining biological data on, and measurements of, spermatozoa which are functionally active. Only then can the definition of normal be achieved and clinically useful criteria be adopted.

Discriminant Analysis↗