Email this article 
Assessment of ovarian follicular reserve according to ultrasound data based on machine learning methods
- Authors: Laputin F.A.1, Sidorov I.V.1, Moshkin A.S.2
-
Affiliations:
- Higher School of Economics
- Orel State University
- Issue: Vol 5, No 1S (2024)
- Pages: 40-42
- Section: Articles by YOUNG SCIENTISTS
- Submitted: 28.01.2024
- Accepted: 07.03.2024
- Published: 03.07.2024
- URL: https://jdigitaldiagnostics.com/DD/article/view/626171
- DOI: https://doi.org/10.17816/DD626171
- ID: 626171
Cite item
Full Text
Abstract
BACKGROUND: Ovarian reserve reflects a woman's ability to successfully realize reproductive function. The assessment of ovarian reserve is an urgent task for clinical practice [1] and is important in scientific research. The use of computerized diagnostic image processing methods can accelerate and facilitate the performance of routine tasks in clinical practice. Their use in retrospective data analysis for scientific purposes allows to increase the objectivity of the study and supplement it with auxiliary information [2].
The issue of ovarian localization and follicle segmentation on ultrasound images has been previously investigated in other works. For instance, Z. Chen et al. [3] employed the U-net model to identify follicles on ultrasound images. Similarly, V.K. Singh et al. [4] addressed a related problem using a variant of U-net, namely UNet++ [5], which has gained considerable traction in the field of medical image analysis [6].
AIM: The study aimed to develop machine learning models for analyzing ovarian images obtained from an ultrasound machine.
MATERIALS AND METHODS: An open dataset with a labeled ovary region was used for pre-training ovarian segmentation and follicle detection models. Subsequently, the dataset, which contains marked-up ovarian and follicle regions, was employed for training and testing. It encompasses a total of approximately 800 examples from 50 unique patients.
The localization of follicles in an ultrasound image is a challenging task. To address this, the designed detector system was divided into two parts: ovary segmentation and follicle detection within the selected region. This approach allows the model to focus on a region where there are no other organs and various ultrasound artifacts that can be falsely perceived as the object under investigation. For the purpose of ovarian segmentation, the UNet++ architecture [5] was employed in conjunction with the ResNeSt encoder [8], which incorporates the SE-Net [9] and SK-Net [10] attention mechanisms.
The object detection model is employed to identify the location of follicles within the ovary, as it enables precise enumeration of the number of follicles, even in the presence of overlapping structures, a capability that the segmentation model lacks. In our study, we used the YOLOv8 model [11].
Furthermore, data preprocessing has been employed to enhance the quality of model predictions. This has involved the identification and removal of regions with auxiliary information, the reduction of noise, and the augmentation of data.
RESULTS: Two ovarian localization models are presented based on the results of this study. The first model is a segmentation model with an IoU quality of at least 50%. The second model is a detection model with a mAP quality of at least 65%. A third model is a model for follicle detection with subsequent follicle counting. This model has an MAPE error not exceeding 35%.
CONCLUSIONS: The study resulted in the proposal of a method for applying machine learning techniques to the task of analyzing ultrasound images. The developed segmentation and detection models reduce the time and errors in analyzing ovaries and follicles in the images. The use of an attention mechanism and data preprocessing improves the quality of the models. The neural network for follicle detection provides follicle counting, even when follicles overlap.
Full Text
BACKGROUND: Ovarian reserve reflects a woman's ability to successfully realize reproductive function. The assessment of ovarian reserve is an urgent task for clinical practice [1] and is important in scientific research. The use of computerized diagnostic image processing methods can accelerate and facilitate the performance of routine tasks in clinical practice. Their use in retrospective data analysis for scientific purposes allows to increase the objectivity of the study and supplement it with auxiliary information [2].
The issue of ovarian localization and follicle segmentation on ultrasound images has been previously investigated in other works. For instance, Z. Chen et al. [3] employed the U-net model to identify follicles on ultrasound images. Similarly, V.K. Singh et al. [4] addressed a related problem using a variant of U-net, namely UNet++ [5], which has gained considerable traction in the field of medical image analysis [6].
AIM: The study aimed to develop machine learning models for analyzing ovarian images obtained from an ultrasound machine.
MATERIALS AND METHODS: An open dataset with a labeled ovary region was used for pre-training ovarian segmentation and follicle detection models. Subsequently, the dataset, which contains marked-up ovarian and follicle regions, was employed for training and testing. It encompasses a total of approximately 800 examples from 50 unique patients.
The localization of follicles in an ultrasound image is a challenging task. To address this, the designed detector system was divided into two parts: ovary segmentation and follicle detection within the selected region. This approach allows the model to focus on a region where there are no other organs and various ultrasound artifacts that can be falsely perceived as the object under investigation. For the purpose of ovarian segmentation, the UNet++ architecture [5] was employed in conjunction with the ResNeSt encoder [8], which incorporates the SE-Net [9] and SK-Net [10] attention mechanisms.
The object detection model is employed to identify the location of follicles within the ovary, as it enables precise enumeration of the number of follicles, even in the presence of overlapping structures, a capability that the segmentation model lacks. In our study, we used the YOLOv8 model [11].
Furthermore, data preprocessing has been employed to enhance the quality of model predictions. This has involved the identification and removal of regions with auxiliary information, the reduction of noise, and the augmentation of data.
RESULTS: Two ovarian localization models are presented based on the results of this study. The first model is a segmentation model with an IoU quality of at least 50%. The second model is a detection model with a mAP quality of at least 65%. A third model is a model for follicle detection with subsequent follicle counting. This model has an MAPE error not exceeding 35%.
CONCLUSIONS: The study resulted in the proposal of a method for applying machine learning techniques to the task of analyzing ultrasound images. The developed segmentation and detection models reduce the time and errors in analyzing ovaries and follicles in the images. The use of an attention mechanism and data preprocessing improves the quality of the models. The neural network for follicle detection provides follicle counting, even when follicles overlap.
About the authors
Fedor A. Laputin
Higher School of Economics
Author for correspondence.
Email: falaputin@edu.hse.ru
ORCID iD: 0009-0009-4037-799X
Russian Federation, Moscow
Ivan V. Sidorov
Higher School of Economics
Email: ivsidorov@edu.hse.ru
ORCID iD: 0009-0004-5150-2737
Russian Federation, Moscow
Andrey S. Moshkin
Orel State University
Email: as.moshkin@internet.ru
ORCID iD: 0000-0003-2085-0718
SPIN-code: 9718-2516
Russian Federation, Orel
References
- Nikolenko VN, Gevorgyan MM, Moshkin AS, Hunanyan AL, Hovhannisyan MV. Comparative characteristics of the volume of ovaries and the number of follicles according to MRI studies in the aspect of assessing the ovarian reserve in different age periods of women. Proceedings of International Scientific and Practical Conference "Modern medicine: new approaches and topical research"; 2020 Nov 22; Grozny. (In Russ). EDN: RWSZNS doi: 10.36684/33-2020-1-584-593
- Certificate of state registration of computer program No. 2022666924/ 12.09.2022. Moshkin AS. Program for analysis of video results of ultrasound Dopplerography (PAVUD). (In Russ). EDN: LPERUS
- Chen Z, Zhang C, Li Z, Yang J, Deng H. Automatic segmentation of ovarian follicles using deep neural network combined with edge information. Frontiers in Reproductive Health. 2022;4. doi: 10.3389/frph.2022.877216
- Singh VK, Yousef Kalafi E, Cheah E, et al. HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy. Diagnostics. 2022;12(12):3213. doi: 10.3390/diagnostics12123213
- Zhou Z, Rahman Siddiquee MM, Tajbakhsh N, Liang J. UNet++: A Nested U-Net Architecture for Medical Image Segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 4th International Workshop, DLMIA 2018 and 8th International Workshop, ML-CDS 2018 Held in Conjunction with MICCAI 2018; 2018 Sept 20; Granada. doi: 10.1007/978-3-030-00889-5_1
- Zaev RI, Romanov AY, Solovyev RA. Segmentation of Prostate Cancer on TRUS Images Using ML. Proceedings of International Russian Smart Industry Conference, SmartIndustryCon; 2023. doi: 10.1109/SmartIndustryCon57312.2023.10110727
- Zhao Q, Lyu S, Bai W, et al. MMOTU: A Multi-Modality Ovarian Tumor Ultrasound Image Dataset for Unsupervised Cross-Domain Semantic Segmentation. arXiv. Preprint. 2022.
- Zhang H, Wu C, Zhang Z, et al. ResNeSt: Split-Attention Networks. IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW); 2022; New Orleans. doi: 10.1109/CVPRW56347.2022.00309
- Hu J, Shen L, Sun G. Squeeze-and-Excitation Networks. IEEE/CVF Conference on Computer Vision and Pattern Recognition; 2018, Salt Lake City. doi: 10.1109/CVPR.2018.00745
- Li X, Wang W, Hu X, Yang J. Selective Kernel Networks. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR); June, 2019. doi: 10.1109/CVPR.2019.00060
- Ultralytics YOLOv8 Docs [Internet]. Ultralytics Inc.; c2024. [cited 2024 Feb 9]. Available from: https://docs.ultralytics.com/ru
Supplementary files
