Tencent Jarvis Lab, Shenzhen, China.
Diagnosis and Treatment for Cervical Lesions Center, Shenzhen Maternity & Child Healthcare Hospital, Shenzhen, China.
Med Image Anal. 2021 May;70:102006. doi: 10.1016/j.media.2021.102006. Epub 2021 Mar 1.
Cervical cancer causes the fourth most cancer-related deaths of women worldwide. Early detection of cervical intraepithelial neoplasia (CIN) can significantly increase the survival rate of patients. World Health Organization (WHO) divided the CIN into three grades (CIN1, CIN2 and CIN3). In clinical practice, different CIN grades require different treatments. Although existing studies proposed computer aided diagnosis (CAD) systems for cervical cancer diagnosis, most of them are fail to perform accurate separation between CIN1 and CIN2/3, due to the similar appearances under colposcopy. To boost the accuracy of CAD systems, we construct a colposcopic image dataset for GRAding cervical intraepithelial Neoplasia with fine-grained lesion Description (GRAND). The dataset consists of colposcopic images collected from 8,604 patients along with the pathological reports. Additionally, we invite the experienced colposcopist to annotate two main clues, which are usually adopted for clinical diagnosis of CIN grade, i.e., texture of acetowhite epithelium (TAE) and appearance of blood vessel (ABV). A multi-rater model using the annotated clues is benchmarked for our dataset. The proposed framework contains several sub-networks (raters) to exploit the fine-grained lesion features TAE and ABV, respectively, by contrastive learning and a backbone network to extract the global information from colposcopic images. A comprehensive experiment is conducted on our GRAND dataset. The experimental results demonstrate the benefit of using additional lesion descriptions (TAE and ABV), which increases the CIN grading accuracy by over 10%. Furthermore, we conduct a human-machine confrontation to evaluate the potential of the proposed benchmark framework for clinical applications. Particularly, three colposcopists on different professional levels (intern, in-service and professional) are invited to compete with our benchmark framework by investigating a same extra test set-our framework achieves a comparable CIN grading accuracy to that of a professional colposcopist.
宫颈癌是全球导致女性癌症相关死亡的第四大原因。早期发现宫颈上皮内瘤变（CIN）可以显著提高患者的生存率。世界卫生组织（WHO）将 CIN 分为三级（CIN1、CIN2 和 CIN3）。在临床实践中，不同的 CIN 分级需要不同的治疗。尽管现有研究提出了用于宫颈癌诊断的计算机辅助诊断（CAD）系统，但由于阴道镜下外观相似，大多数系统无法准确区分 CIN1 和 CIN2/3。为了提高 CAD 系统的准确性，我们构建了一个用于基于细粒度病变描述的宫颈上皮内瘤变分级的阴道镜图像数据集（GRAND）。该数据集包含从 8604 名患者收集的阴道镜图像以及病理报告。此外，我们邀请经验丰富的阴道镜医生对两个主要线索进行注释，这两个线索通常用于 CIN 分级的临床诊断，即醋酸白色上皮的纹理（TAE）和血管的外观（ABV）。使用标注线索的多评分者模型在我们的数据集上进行基准测试。所提出的框架包含几个子网络（评分者），通过对比学习分别利用细粒度病变特征 TAE 和 ABV，并利用骨干网络从阴道镜图像中提取全局信息。在我们的 GRAND 数据集上进行了全面的实验。实验结果表明，使用额外的病变描述（TAE 和 ABV）的益处，使 CIN 分级准确性提高了 10%以上。此外，我们进行了人机对抗，以评估所提出的基准框架用于临床应用的潜力。特别邀请了三位不同专业水平（实习生、在职和专业）的阴道镜医生通过调查同一个额外的测试集与我们的基准框架竞争-我们的框架达到了与专业阴道镜医生相当的 CIN 分级准确性。
