Performance of a convolutional neural network in screening liquid based cervical cytology smears

• Main Authors: Parikshit Sanyal, Sanghita Barui, Prabal Deb, Harish Chander Sharma
• Format: Article
• Language: English
• Published: Wolters Kluwer Medknow Publications 2019-01-01
• Series: Journal of Cytology
• Subjects: Artificial intelligence; cervical cytology; liquid based smears; neural network; screening
• Online Access: http://www.jcytol.org/article.asp?issn=0970-9371;year=2019;volume=36;issue=3;spage=146;epage=151;aulast=Sanyal

Context: Cervical cancer is the second most common cancer in women. The liquid based cervical cytology (LBCC) is a useful tool of choice for screening cervical cancer. Aims: To train a convolutional neural network (CNN) to identify abnormal foci from LBCC smears. Settings and Design: We have chosen retrospective study design from archived smears of patients undergoing screening from cervical cancer by LBCC smears. Materials and Methods: 2816 images, each of 256 × 256 pixels, were prepared from microphotographs of these LBCC smears, which included 816 “abnormal” foci (low grade or high grade squamous intraepithelial lesion) and 2000 'normal' foci (benign epithelial cells and reactive changes). The images were split into three sets, Training, Testing, and Evaluation. A convolutional neural network (CNN) was developed with the python programming language. The CNN was trained with the Training dataset; performance was assayed concurrently with the Testing dataset. Two CNN models were developed, after 20 and 10 epochs of training, respectively. The models were then run on the Evaluation dataset. Statistical Analysis Used: A contingency table was prepared from the original image labels and the labels predicted by the CNN. Results: Combined assessment of both models yielded a sensitivity of 95.63% in detecting abnormal foci, with 79.85% specificity. The negative predictive value was high (99.19%), suggesting potential utility in screening. False positives due to overlapping cells, neutrophils, and debris was the principal difficulty met during evaluation. Conclusions: The CNN shows promise as a screening tool; however, for its use in confirmatory diagnosis, further training with a more diverse dataset will be required.