Automated Detection of Endometrial Polyps from Hysteroscopic Videos Using Deep Learning

Endometrial polyps are common gynecological lesions. The standard treatment for this condition is hysteroscopic polypectomy. However, this procedure may be accompanied by misdetection of endometrial polyps. To improve the diagnostic accuracy and reduce the risk of misdetection, a deep learning model based on YOLOX is proposed to detect endometrial polyps in real time. Group normalization is employed to improve its performance with large hysteroscopic images. In addition, we propose a video adjacent-frame association algorithm to address the problem of unstable polyp detection. Our proposed model was trained on a dataset of 11,839 images from 323 cases provided by a hospital and was tested on two datasets of 431 cases from two hospitals. The results show that the lesion-based sensitivity of the model reached 100% and 92.0% for the two test sets, compared with 95.83% and 77.33%, respectively, for the original YOLOX model. This demonstrates that the improved model may be used effectively as a diagnostic tool during clinical hysteroscopic procedures to reduce the risk of missing endometrial polyps.

[1]  K. Ramamurthy,et al.  A Novel Multi-Feature Fusion Method for Classification of Gastrointestinal Diseases Using Endoscopy Images , 2022, Diagnostics.

[2]  S. Balakrishnan,et al.  Attention Aware Deep Learning Model for Wireless Capsule Endoscopy Lesion Classification and Localization , 2022, Journal of Medical and Biological Engineering.

[3]  Nan Bao,et al.  Automatic evaluation of endometrial receptivity in three-dimensional transvaginal ultrasound images based on 3D U-Net segmentation , 2021, Quantitative Imaging in Medicine and Surgery.

[4]  K. T. Dilna,et al.  Classification of Uterine Fibroids in Ultrasound Images Using Deep Learning Model , 2022, International Conference on Conceptual Structures.

[5]  S. Sumathy,et al.  Transfer learning approach in deep neural networks for uterine fibroid detection , 2022, Int. J. Comput. Sci. Eng..

[6]  Zhiyong Xia,et al.  Deep learning-based hysteroscopic intelligent examination and ultrasound examination for diagnosis of endometrial carcinoma , 2021, The Journal of Supercomputing.

[7]  Xiaomei Yang,et al.  Deep Learning for Intelligent Recognition and Prediction of Endometrial Cancer , 2021, Journal of healthcare engineering.

[8]  Zeming Li,et al.  YOLOX: Exceeding YOLO Series in 2021 , 2021, ArXiv.

[9]  Y. Himoto,et al.  Automatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network , 2021, Scientific Reports.

[10]  Bulent Bayram,et al.  Deep neural network approaches for detecting gastric polyps in endoscopic images , 2021, Medical & Biological Engineering & Computing.

[11]  S. Zhang,et al.  Accuracy of artificial intelligence‐assisted detection of esophageal cancer and neoplasms on endoscopic images: A systematic review and meta‐analysis , 2021, Journal of digestive diseases.

[12]  K. Oda,et al.  Automated system for diagnosing endometrial cancer by adopting deep-learning technology in hysteroscopy , 2021, PloS one.

[13]  Erlend Hodneland,et al.  Automated segmentation of endometrial cancer on MR images using deep learning , 2021, Scientific Reports.

[14]  Yun-zheng Zhang,et al.  Deep learning model for classifying endometrial lesions , 2021, Journal of translational medicine.

[15]  Haavard D. Johansen,et al.  Real-Time Polyp Detection, Localisation and Segmentation in Colonoscopy Using Deep Learning , 2020, ArXiv.

[16]  K. Koike,et al.  Automatic detection of colorectal neoplasia in wireless colon capsule endoscopic images using a deep convolutional neural network , 2020, Endoscopy.

[17]  Cheng-Chang Chang,et al.  Using Deep Learning with Convolutional Neural Network Approach to Identify the Invasion Depth of Endometrial Cancer in Myometrium Using MR Images: A Pilot Study , 2020, International journal of environmental research and public health.

[18]  Tang Chun-ming,et al.  MRI Image Segmentation System of Uterine Fibroids Based on AR-Unet Network , 2020 .

[19]  R. Shlansky-Goldberg,et al.  Deep Learning Based on MR Imaging for Predicting Outcome of Uterine Fibroid Embolization. , 2020, Journal of vascular and interventional radiology : JVIR.

[20]  Quoc V. Le,et al.  EfficientDet: Scalable and Efficient Object Detection , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Kaiming He,et al.  Group Normalization , 2018, International Journal of Computer Vision.

[22]  Balázs Harangi,et al.  Digital Image Analysis with Fully Connected Convolutional Neural Network to Facilitate Hysteroscopic Fibroid Resection , 2018, Gynecologic and Obstetric Investigation.

[23]  Dietrich Paulus,et al.  Simple online and realtime tracking with a deep association metric , 2017, 2017 IEEE International Conference on Image Processing (ICIP).

[24]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.