Comparative evaluation of clinical characteristics of biopsy-proven conjunctival intraepithelial neoplasia and invasive squamous cell carcinoma using image processing software programs.
2021
PURPOSE To identify morphological parameters aiding clinical differentiation of conjunctival intraepithelial neoplasia (CIN) and invasive squamous cell carcinoma (iSCC) and to demonstrate the utility of image processing software to objectively assess ocular surface squamous neoplasia (OSSN). METHODS This retrospective case series included all biopsy-proven cases of OSSN presenting as an ocular surface nodule. Based on histopathology, lesions were classified as CIN and iSCC. Clinical image analysis utilized 'Contour' and 'ImageJ' software. The effect of predictors demography, seropositivity, lesion dimensions, keratin, pigmentation, corneal involvement, vascularity and feeder vessels on the final histopathologic grade were assessed. RESULTS A total of 108 OSSN lesions (74 CIN and 33 iSCC) were included. Mean age was 46.1 ± 17.2 years in CIN and 47.2 ± 13.9 years in iSCC. By univariate logistic regression analysis, significant predictors of iSCC were HIV seropositivity (p < 0.0001), maximum diameter (p = 0.003), perpendicular to maximum diameter (p = 0.003), height (p = 0.003), nodular morphology (p = 0.006) and feeder vessels (p = 0.03), whereas gelatinous morphology (p = 0.02) was predictor of CIN. By multiple logistic regression, seropositivity was the predictor of iSCC (p < 0.0001, OR 13.33 ± 8.35, 95% CI 3.90-45.53). CONCLUSION HIV seropositivity is an important predictor of iSCC. Large, thick, nodular lesions with feeder vessels may favor the diagnosis of iSCC, whereas gelatinous, small, flatter lesions without feeder vessels may favor CIN. In a first of its kind study, simple and objective analysis of OSSN with image processing software was demonstrated.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
28
References
0
Citations
NaN
KQI