Abstract
Technical advancements significantly improve earlier diagnosis of cervical cancer, but accurate diagnosis is still difficult due to various factors. We develop an artificial intelligence assistive diagnostic solution, AIATBS, to improve cervical liquid-based thin-layer cell smear diagnosis according to clinical TBS criteria. We train AIATBS with >81,000 retrospective samples. It integrates YOLOv3 for target detection, Xception and Patch-based models to boost target classification, and U-net for nucleus segmentation. We integrate XGBoost and a logical decision tree with these models to optimize the parameters given by the learning process, and we develop a complete cervical liquid-based cytology smear TBS diagnostic system which also includes a quality control solution. We validate the optimized system with >34,000 multicenter prospective samples and achieve better sensitivity compared to senior cytologists, yet retain high specificity while achieving a speed of <180s/slide. Our system is adaptive to sample preparation using different standards, staining protocols and scanners.
Highlights
Technical advancements significantly improve earlier diagnosis of cervical cancer, but accurate diagnosis is still difficult due to various factors
It is especially worth noting that deep learning (DL) has been extensively studied in pathology, and convolutional neural networks (CNNs) have become the preferred technology for general image classification[10]
We have developed an Artificial Intelligence (AI) assistive diagnostic system that can help cytologists perform data interpretation and improve the efficiency and quality of screening based on TBS standards
Summary
Technical advancements significantly improve earlier diagnosis of cervical cancer, but accurate diagnosis is still difficult due to various factors. We develop an artificial intelligence assistive diagnostic solution, AIATBS, to improve cervical liquid-based thin-layer cell smear diagnosis according to clinical TBS criteria. The accurate diagnosis of cervical liquid-based cytology smears is still a challenge due to the following two reasons: (I) technically speaking, a cytologist must find only a few abnormal and malignant cells in a sample composed of tens of thousands of cells, (II) in addition, due to the lack of experienced and qualified cytologists or cytotechnologists, and the influence of factors such as their diagnostic experiences, moods, fatigue, etc. It is especially worth noting that deep learning (DL) has been extensively studied in pathology, and CNNs have become the preferred technology for general image classification[10] It has been used for image-based detection tasks[11,12] and applied to identify and quantify cellular[13] and histological features[14,15]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
