Infrared spectroscopy of exfoliated cervical cell specimens - Proceed withcaution

OBJECTIVE: To determine whether diagnostic information may be recovered fro m the infrared spectra of exfoliated cell specimens by using a novel spectr al feature extraction method, in conjunction with linear and quadratic disc riminant analysis, for spectral classification. STUDY DESIGN: Over 800 infrared spectra were included in the study, with co rresponding clinical diagnoses based upon cytology and, when available, his tology reports. Three sets of classification trials were carried out with t he aim of distinguishing the spectra corresponding to normal specimens from CIN 1, 2 and 3. For each of these three cases, the procedure was to: (1) d evelop a set of provisional classification models using only a "training" s ubset of the spectra, and (2) test each provisional model by its ability to correctly predict the diagnoses on the basis of the remaining spectra. RESULTS: For optimal classification trials, training set classification acc uracies were 68% for normal/CIN 1, 73% for normal/CIN 2 and 81% for normal/ CIN 3; for the corresponding test sets the classification accuracies were 6 0%, 60% and 67%, respectively. CONCLUSION: The infrared spectra of exfoliated cervical cells carry informa tion regarding the presence or absence of dysplasia, and that information i s recoverable-albeit imperfectly at this stage-from the spectra of of "real life" cell preparations.