Image analysis and neural networks discriminating endometrial lesions

What is it about?

This study aims investigates the efficacy of Artificial Neural Networks (ANN) to discriminate between benign and malignant endometrial nuclei and lesions in cytological specimens. 416 histologically confirmed cytological smears were used. Morphometric characteristics of 90 nuclei per case were analyzed using image analysis; half of them were used to train the ANN and the remaining 50% were used to evaluate the ANN. The accuracy of the ANN model for the classification of endometrial nuclei was 81.33%. For the case classification the overall accuracy was 90.87%. Conclusion: Computerized systems based on ANNs can aid the cytological classification of endometrial nuclei and lesions with sufficient sensitivity and specificity.

Featured Image

Why is it important?

It is a two step system classifying via artificial intelligence and morphometry, initially nuclei, and subsequently patients. This paper describes a system that may help computerized automation of endometrium cytological diagnoses.