We consider here the problem of classifying a macro-level object based on measurements of embedded (micro-level) observations within each object, for example, classifying a patient based on measurements on a collection of a random number of their cells. Classification problems with this hierarchical, nested structure have not received the same statistical understanding as the general classification problem. Some heuristic approaches have been developed and a few authors have proposed formal statistical models. We focus on the problem where heterogeneity exists between the macro-level objects within a class. We propose a model-based statistical methodology that models the log-odds of the macro-level object belonging to a class using a latent-class variable model to account for this heterogeneity. The latent classes are estimated by clustering the macro-level object density estimates. We apply this method to the detection of patients with cervical neoplasia based on quantitative cytology measurements on cells in a Papanicolaou smear. Quantitative cytology is much cheaper and potentially can take less time than the current standard of care. The results show that the automated quantitative cytology using the proposed method is roughly equivalent to clinical cytopathology and shows significant improvement over a statistical model that does not account for the heterogeneity of the data.

Testing emerging technologies involves the evaluation of biologic plausibility, technical efficacy, clinical effectiveness, patient satisfaction, and cost-effectiveness. The objective of this study was to select an effective classification algorithm for optical spectroscopy as an adjunct to colposcopy and obtain preliminary estimates of its accuracy for the detection of CIN 2 or worse. We recruited 1000 patients from screening and prevention clinics and 850 patients from colposcopy clinics at two comprehensive cancer centers and a community hospital. Optical spectroscopy was performed and 4864 biopsies were obtained from the sites measured, including abnormal and normal colposcopic areas. The gold standard was the histologic report of biopsies, read 2–3 times by histopathologists blinded to the cytologic, histopathologic, and spectroscopic results. We calculated sensitivities, specificities, receiver operating characteristic (ROC) curves, and areas under the ROC curves. We identified a cutpoint for an algorithm based on optical spectroscopy that yielded an estimated sensitivity of 1.00 [95% confidence interval (CI) = 0.92 – 1.00] and an estimated specificity of 0.71 [95% CI = 0.62 – 0.79] in a combined screening and diagnostic population. The positive and negative predictive values were 0.58 and 1.00, respectively. The area under the ROC curve was 0.85 (95% CI 0.81 – 0.89). The per-patient and per-site performance were similar in the diagnostic and poorer in the screening settings. Like colposcopy, the device performs best in a diagnostic population. Alternative statistical approaches demonstrate that the analysis is robust and that spectroscopy works as well as or slightly better than colposcopy for the detection of CIN 2 to cancer.

The purpose of this work was to investigate the potential of the normalized axial-shear strain area (NASSA) feature, derived from axial-shear strain elastograms (ASSE), for breast lesion classification of fibroadenoma and cancer. This study consisted of previously-acquired in vivo digital RF-data of breast lesions. A total of 33 biopsy-proven malignant tumors and 30 fibroadenoma cases were included in the study that involved 3 observers blinded to the original BIRADS®-ultrasound scores. The observers outlined the lesions on the sonograms. The ASSEs were segmented and color-overlaid on the sonograms, and the NASSA feature from the ASSE was computed semi-automatically. Receiver operating characteristic (ROC) curves were then generated and the area under the curve (AUC) was calculated for each observer performance. A logistic regression classifier was built to compare the improvement in the AUC when using BIRADS scores plus NASSA values as opposed to BIRADS scores alone. BIRADS score ROC had an AUC of 0.89 (95% CI = 0.81 – 0.97). In comparison, the average of the AUC for all the three observers using ASSE feature alone was 0.84. However, the AUC increased to 0.94 (average of 3 observers) when BIRADS score and ASSE feature were combined. The results demonstrate that the NASSA feature derived from ASSE has the potential to improve BIRADS breast lesion classification of fibroadenoma and malignant tumors.

OBJECTIVE

To estimate whether women who deliver small babies due to preterm birth or growth restriction have excess risk for cardiovascular disease and diabetes later in life.

METHODS

Eight years after pregnancy, we estimated the prevalence of metabolic syndrome and its components in a cohort study of women with prior preterm (preterm birth before 37 weeks, n=181) or small for gestational age ([SGA], less than the tenth percentile, n=192) births, compared with women with term births (37 or more weeks, n=306). Women delivered at Magee-Womens Hospital in Pittsburgh, Pennsylvania, and those with preeclampsia or prepregnancy diabetes or hypertension were excluded. Women underwent a structured interview and fasting blood sampling.

RESULTS

Women were, on average, 8 years postpartum and 39 years old at evaluation. Women with a prior preterm birth had higher blood pressure, triglycerides, and LDL-cholesterol compared with those in a term control group. Women with prior SGA births were leaner and more likely to smoke compared with those with term births. Women with prior preterm birth had elevated risk of metabolic syndrome, adjusted for demographic, smoking and body size factors (23% preterm compared with 17% control group; odds ratio [OR] 1.76 [1.06, 2.80]). In women with a prior preterm birth, low HDL (11% preterm compared with 5% control group; OR 2.6 [1.2, 5.2]), hypertriglyceridemia (22% compared with 14%; OR 1.9 [1.2, 2.9]), and elevated glucose (24% compared with 19%; OR 1.5 [1.0, 2.3]) accounted for this excess metabolic syndrome. In women with SGA, the only element of metabolic syndrome that was aberrant was glucose metabolism.

CONCLUSION

Eight years after pregnancy, women with prior preterm or SGA births had evidence of metabolic syndrome compared with women with term births. Screening and intervention in these women after pregnancy may delay or prevent disease.