Deep learning enables automated MRI-based estimation of uterine volume also in patients with uterine fibroids undergoing high-intensity focused ultrasound therapy

Abstract

Background High-intensity focused ultrasound (HIFU) is used for the treatment of symptomatic leiomyomas. We aim to automate uterine volumetry for tracking changes after therapy with a 3D deep learning approach. Methods A 3D nnU-Net model in the default setting and in a modified version including convolutional block attention modules (CBAMs) was developed on 3D T2-weighted MRI scans. Uterine segmentation was performed in 44 patients with routine pelvic MRI (standard group) and 56 patients with uterine fibroids undergoing ultrasound-guided HIFU therapy (HIFU group). Here, preHIFU scans (n = 56), postHIFU imaging maximum one day after HIFU (n = 54), and the last available follow-up examination (n = 53, days after HIFU: 420 ± 377) were included. The training was performed on 80% of the data with fivefold cross-validation. The remaining data were used as a hold-out test set. Ground truth was generated by a board-certified radiologist and a radiology resident. For the assessment of inter-reader agreement, all preHIFU examinations were segmented independently by both. Results High segmentation performance was already observed for the default 3D nnU-Net (mean Dice score = 0.95 ± 0.05) on the validation sets. Since the CBAM nnU-Net showed no significant benefit, the less complex default model was applied to the hold-out test set, which resulted in accurate uterus segmentation (Dice scores: standard group 0.92 ± 0.07; HIFU group 0.96 ± 0.02), which was comparable to the agreement between the two readers. Conclusions This study presents a method for automatic uterus segmentation which allows a fast and consistent assessment of uterine volume. Therefore, this method could be used in the clinical setting for objective assessment of therapeutic response to HIFU therapy.

Efficient feature extraction and hybrid deep learning for early identification of uterine fibroids in ultrasound images

“…Ahmed et al 10 developed a deep neural network, VGG 16, for detecting uterine fibroids in medical images, achieving 98.5% accuracy. Theis et al 11 developed a 3D U‐Net model for automatic uterus segmentation, allowing fast and consistent assessment of uterine volume on 3D T2‐weighted MRI scans. This method is accurate, reproducible, and can be applied post‐interventionally.…”

“…The generalization of these models to broader populations remains a concern. Deep learning models, as employed by Zhang et al, 16 Theis et al, 11 and Yang et al, 20 often lack interpretability, making it challenging for clinicians to understand and trust the decision‐making process. Clear explanations of how the model arrives at its conclusions are essential for clinical acceptance.…”

Multi-omics and machine learning for the prevention and management of female reproductive health

“…To distinguish between a normal and disordered uterus from the TCGA-UCEC dataset, an automated technique based on VGG 16 of deep learning classification models was used. The model's accuracy in predicting the kind of uterine fibrosis from image data is 98.5% (101). Using neural networks, the Fibroid Disease Prediction System (FDPS) was created.…”

Automated Uterine Fibroids Detection in Ultrasound Images Using Deep Convolutional Neural Networks

“…The necessity of fast and precise UF detection in ultrasound images prompted this study [10]. Using DCNN for automated detection may lead to greater precision, reduced burden on healthcare providers, and quicker and cheaper diagnosis [11][12][13][14]. The importance of this study is in the practical applications of DCNN for detecting fibroids.…”

“…It is a major health risk for women and must be diagnosed at an early stage for effective treatment. However, as documented in publications [17][18][19][20], manually identifying UF from ultrasounds is a laborious and time-consuming process that is frequently affected by inter-observer variability [14]. The suggested system uses CNN and other DL techniques to automatically detect UF in ultrasound images.…”