Automated Evaluation of p16/Ki-67 Dual-Stain Cytology as a Biomarker for Detection of Anal Precancer in Men Who Have Sex With Men and Are Living With Human Immunodeficiency Virus.

BACKGROUND: Human papillomavirus-related biomarkers such as p16/Ki-67 "dual-stain" (DS) cytology have shown promising clinical performance for anal cancer screening. Here, we assessed the performance of automated evaluation of DS cytology (automated DS) to detect anal precancer in men who have sex with men (MSM) and are living with human immunodeficiency virus (HIV). METHODS: We conducted a cross-sectional analysis of 320 MSM with HIV undergoing anal cancer screening and high-resolution anoscopy (HRA) in 2009-2010. We evaluated the performance of automated DS based on a deep-learning classifier compared to manual evaluation of DS cytology (manual DS) to detect anal intraepithelial neoplasia grade 2 or 3 (AIN2+) and grade 3 (AIN3). We evaluated different DS-positive cell thresholds quantified by the automated approach and modeled performance compared with other screening strategies in a hypothetical population of MSM with HIV. RESULTS: Compared with manual DS, automated DS had significantly higher specificity (50.9% vs 42.2%; P < .001) and similar sensitivity (93.2% vs 92.1%) for detection of AIN2+. Human papillomavirus testing with automated DS triage was significantly more specific than automated DS alone (56.5% vs 50.9%; P < .001), with the same sensitivity (93.2%). In a modeled analysis assuming a 20% AIN2+ prevalence, automated DS detected more precancers than manual DS and anal cytology (186, 184, and 162, respectively) and had the lowest HRA referral rate per AIN2+ case detected (3.1, 3.5, and 3.3, respectively). CONCLUSIONS: Compared with manual DS, automated DS detects the same number of precancers, with a lower HRA referral rate.

CNN stability training improves robustness to scanner and IHC-based image variability for epithelium segmentation in cervical histology.

Background: In digital pathology, image properties such as color, brightness, contrast and blurriness may vary based on the scanner and sample preparation. Convolutional Neural Networks (CNNs) are sensitive to these variations and may underperform on images from a different domain than the one used for training. Robustness to these image property variations is required to enable the use of deep learning in clinical practice and large scale clinical research. Aims: CNN Stability Training (CST) is proposed and evaluated as a method to increase CNN robustness to scanner and Immunohistochemistry (IHC)-based image variability. Methods: CST was applied to segment epithelium in immunohistological cervical Whole Slide Images (WSIs). CST randomly distorts input tiles and factors the difference between the CNN prediction for the original and distorted inputs within the loss function. CNNs were trained using 114 p16-stained WSIs from the same scanner, and evaluated on 6 WSI test sets, each with 23 to 24 WSIs of the same tissue but different scanner/IHC combinations. Relative robustness (rAUC) was measured as the difference between the AUC on the training domain test set (i.e., baseline test set) and the remaining test sets. Results: Across all test sets, The AUC of CST models outperformed "No CST" models (AUC: 0.940-0.989 vs. 0.905-0.986, p < 1e - 8), and obtained an improved robustness (rAUC: [-0.038, -0.003] vs. [-0.081, -0.002]). At a WSI level, CST models showed an increase in performance in 124 of the 142 WSIs. CST models also outperformed models trained with random on-the-fly data augmentation (DA) in all test sets ([0.002, 0.021], p < 1e-6). Conclusion: CST offers a path to improve CNN performance without the need for more data and allows customizing distortions to specific use cases. A python implementation of CST is publicly available at https://github.com/TIGACenter/CST_v1.

Accuracy and Efficiency of Deep-Learning-Based Automation of Dual Stain Cytology in Cervical Cancer Screening.

BACKGROUND: With the advent of primary human papillomavirus testing followed by cytology for cervical cancer screening, visual interpretation of cytology slides remains the last subjective analysis step and suffers from low sensitivity and reproducibility. METHODS: We developed a cloud-based whole-slide imaging platform with a deep-learning classifier for p16/Ki-67 dual-stained (DS) slides trained on biopsy-based gold standards. We compared it with conventional Pap and manual DS in 3 epidemiological studies of cervical and anal precancers from Kaiser Permanente Northern California and the University of Oklahoma comprising 4253 patients. All statistical tests were 2-sided. RESULTS: In independent validation at Kaiser Permanente Northern California, artificial intelligence (AI)-based DS had lower positivity than cytology (P < .001) and manual DS (P < .001) with equal sensitivity and substantially higher specificity compared with both Pap (P < .001) and manual DS (P < .001), respectively. Compared with Pap, AI-based DS reduced referral to colposcopy by one-third (41.9% vs 60.1%, P < .001). At a higher cutoff, AI-based DS had similar performance to high-grade squamous intraepithelial lesions cytology, indicating a risk high enough to allow for immediate treatment. The classifier was robust, showing comparable performance in 2 cytology systems and in anal cytology. CONCLUSIONS: Automated DS evaluation removes the remaining subjective component from cervical cancer screening and delivers consistent quality for providers and patients. Moving from Pap to automated DS substantially reduces the number of colposcopies and also achieves excellent performance in a simulated fully vaccinated population. Through cloud-based implementation, this approach is globally accessible. Our results demonstrate that AI not only provides automation and objectivity but also delivers a substantial benefit for women by reduction of unnecessary colposcopies.