Utility of the Diffusion Weighted Sequence in Gynecological Imaging: Review Article

Magnetic Resonance Imaging of Endometriosis: The Role of Advanced Techniques

Endometriosis is a chronic inflammatory disease that affects about 10% of women, and it is characterized by the presence of endometrial tissue outside the uterine cavity. Associated symptoms are dyspareunia, chronic pelvic pain, and infertility. The diagnosis of endometriosis can be challenging due to various clinical and imaging presentations. Laparoscopy is the gold standard for the diagnosis, but it is an invasive procedure. The literature has increasingly promoted a switch to less invasive imaging techniques, such as ultrasound and magnetic resonance imaging (MRI). The latter, also in relation to the latest technological advances, allows a comprehensive and accurate assessment of the pelvis and it can also identify sites of endometriosis that escape laparoscopic evaluation. Furthermore, MRI has been found to be more accurate than other imaging techniques in relation to its improved sensitivity and specificity in identifying disease sites, also due to the role of new emerging sequences. This article aims to review the current role of advanced MRI applications in the assessment of endometriosis.

Development and validation of a deep learning-based method for automatic measurement of uterus, fibroid, and ablated volume in MRI after MR-HIFU treatment of uterine fibroids

INTRODUCTION

The non-perfused volume divided by total fibroid load (NPV/TFL) is a predictive outcome parameter for MRI-guided high-intensity focused ultrasound (MR-HIFU) treatments of uterine fibroids, which is related to long-term symptom relief. In current clinical practice, the MR-HIFU outcome parameters are typically determined by visual inspection, so an automated computer-aided method could facilitate objective outcome quantification. The objective of this study was to develop and evaluate a deep learning-based segmentation algorithm for volume measurements of the uterus, uterine fibroids, and NPVs in MRI in order to automatically quantify the NPV/TFL.

MATERIALS AND METHODS

A segmentation pipeline was developed and evaluated using expert manual segmentations of MRI scans of 115 uterine fibroid patients, screened for and/or undergoing MR-HIFU treatment. The pipeline contained three separate neural networks, one per target structure. The first step in the pipeline was uterus segmentation from contrast-enhanced (CE)-T1w scans. This segmentation was subsequently used to remove non-uterus background tissue for NPV and fibroid segmentation. In the following step, NPVs were segmented from uterus-only CE-T1w scans. Finally, fibroids were segmented from uterus-only T2w scans. The segmentations were used to calculate the volume for each structure. Reliability and agreement between manual and automatic segmentations, volumes, and NPV/TFLs were assessed.

RESULTS

For treatment scans, the Dice similarity coefficients (DSC) between the manually and automatically obtained segmentations were 0.90 (uterus), 0.84 (NPV) and 0.74 (fibroid). Intraclass correlation coefficients (ICC) were 1.00 [0.99, 1.00] (uterus), 0.99 [0.98, 1.00] (NPV) and 0.98 [0.95, 0.99] (fibroid) between manually and automatically derived volumes. For manually and automatically derived NPV/TFLs, the mean difference was 5% [-41%, 51%] (ICC: 0.66 [0.32, 0.85]).

CONCLUSION

The algorithm presented in this study automatically calculates uterus volume, fibroid load, and NPVs, which could lead to more objective outcome quantification after MR-HIFU treatments of uterine fibroids in comparison to visual inspection. When robustness has been ascertained in a future study, this tool may eventually be employed in clinical practice to automatically measure the NPV/TFL after MR-HIFU procedures of uterine fibroids.

MR Diffusion Kurtosis Imaging (DKI) of the Normal Human Uterus in Vivo During the Menstrual Cycle

BACKGROUND

The uterus undergoes dynamic changes throughout the menstrual cycle. Diffusion kurtosis imaging (DKI) is based on the non-Gaussian distribution of water molecules and can perhaps represent the changes of uterine microstructure.

PURPOSE

To investigate the temporal changes in DKI-parameters of the normal uterine corpus and cervix during the menstrual cycle.

STUDY TYPE

Prospective.

POPULATION

21 healthy female volunteers (26.64 ± 4.72 years) with regular menstrual cycles (28 ± 7 days).

FIELD STRENGTH/SEQUENCE

Readout segmentation of long variable echo-trains (RESOLVE)-based DKI and fast spin-echo T2-weighted sequences at 3.0T.

ASSESSMENT

Each volunteer was scanned during the menstrual phase, ovulatory phase, and luteal phase. Regions of interest (ROI) were manually delineated in the endometrium, junctional zone, and myometrium of the uterine body, and in the mucosal layer, fibrous stroma layer, and loose stroma layer of the cervix. The mean Kapp (diffusion kurtosis coefficient), Dapp (diffusion coefficient), and ADC (apparent diffusion coefficient) values were measured in the ROI.

STATISTICAL TESTS

ANOVA with Bonferroni or Tamhane correction. Intraclass correlation coefficient (ICC) for assessing agreement. P < 0.05 was considered statistically significant.

RESULTS

During the menstrual cycle, the highest Kapp (0.848 ± 0.184) and lowest Dapp (1.263 ± 0.283 *10-3 mm2/sec) values were found in the endometrium during the menstrual phase. The Dapp values for the myometrium were significantly higher than those of the endometrium and the junctional zone in every phase. Meanwhile, the Dapp values for the three zonal structures of the cervix during ovulation were significantly higher than those during the luteal phase. However, there was no significant difference in the ADC values of the loose stroma between ovulation and the luteal phase (P = 0.568). The reproducibility of DKI parameters was good (ICC, 0.857-0.944).

DATA CONCLUSION

DKI can show dynamic changes of the normal uterus during the menstrual cycle.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY: Stage 2.

Metastatic malignant peripheral nerve sheath tumor of the uterus and cervix: Diagnostic Challenges, prognostic determinants and treatment

•MPNST is an uncommon sarcoma of the nerve sheath that is rarely found in the female reproductive tract.•Preoperative uterine mass imaging should include pelvic MRI and thorough evaluation of imaging by an expert pelvic MRI radiologist.•Metastatic MPNST has a poor prognosis and systemic treatment options lack efficacy.

Evaluation of multiplexed sensitivity encoding diffusion-weighted imaging in detecting uterine lesions: Image quality optimization

PURPOSE

To compare the image quality of multiplexed sensitivity-encoding diffusion-weighted imaging (MUSE-DWI) and single-shot echo-planar imaging (SS-EPI-DWI) techniques in uterine MRI.

METHODS

Eighty-eight eligible patients underwent MUSE-DWI and SS-EPI-DWI examinations simultaneously using a 3.0 T MRI system. Two radiologists independently performed quantitative and qualitative analysis of the two groups of images using a double-blind method. The weighted Kappa test was used to evaluate the interobserver agreement. Wilcoxon's rank sum test was used for qualitative parameters, and paired t-test was used for quantitative parameters. Spearman rank correlation analysis was used to obtained correlation between pathological results and mean apparent diffusion coefficient (ADC) value.

RESULTS

The qualitative and quantitative analysis of the images by the two radiologists were in good or excellent agreement, with weighted kappa value ranging from 0.636 to 0.981. The scores of total subjective image quality (15.4 ± 0.99) and signal-to-noise ratio (158.99 ± 60.71) of MUSE-DWI were significantly higher than those of SS-EPI-DWI (12.93 ± 1.62 P < 0.001; 130.23 ± 48.29 P < 0.05). It effectively reduced image distortion and artifact, and had better lesion conspicuity. There was no significant difference in contrast-to-noise ratio score and average ADC values between the two DWI sequences. The average ADC values of the two DWI sequences were highest in the normal uterus group and lowest in the endometrial cancer group, with statistically significant differences among groups (P < 0.01). In addition, the average ADC values of the two DWI sequences were negatively correlated with the type of lesions, decreasing with the malignancy of the lesions (r = -0.805 P < 0.01, r = -0.815 P < 0.01).

CONCLUSION

Compared to SS-EPI-DWI, MUSE-DWI can significantly reduce distortion, artifacts, and fuzziness in MRI of uterine lesions, which is more conducive to lesion detection.

MRI combined with clinical features to differentiate ovarian thecoma-fibroma with cystic degeneration from ovary adenofibroma

OBJECTIVE

To explore the value of magnetic resonance imaging (MRI) and clinical features in identifying ovarian thecoma-fibroma (OTF) with cystic degeneration and ovary adenofibroma (OAF).

METHODS

A total of 40 patients with OTF (OTF group) and 28 patients with OAF (OAF group) were included in this retrospective study. Univariable and multivariable analyses were performed on clinical features and MRI between the two groups, and the receiver operating characteristic (ROC) curve was plotted to estimate the optimal threshold and predictive performance.

RESULTS

The OTF group had smaller cyst degeneration degree (P < .001), fewer black sponge sign (20% vs. 53.6%, P = .004), lower minimum apparent diffusion coefficient value (ADCmin) (0.986 (0.152) vs. 1.255 (0.370), P < .001), higher age (57.4 ± 14.2 vs. 44.1 ± 15.9, P = .001) and more postmenopausal women (72.5% vs. 28.6%, P < .001) than OAF. The area under the curve of MRI, clinical features and MRI combined with clinical features was 0.870, 0.841, and 0.954, respectively, and MRI combined with clinical features was significantly higher than the other two (P < .05).

CONCLUSION

The cyst degeneration degree, black sponge sign, ADCmin, age and menopause were independent factors in identifying OTF with cystic degeneration and OAF. The combination of MRI and clinical features has a good effect on the identification of the two.

ADVANCES IN KNOWLEDGE

This is the first time to distinguish OTF with cystic degeneration from OAF by combining MRI and clinical features. It shows the diagnostic performance of MRI, clinical features, and combination of the two. This will facilitate the discriminability and awareness of these two diseases among radiologists and gynaecologists.

Manifestations of Ovarian Cancer in Relation to Other Pelvic Diseases by MRI

Imaging plays a pivotal role in the diagnostic approach of women with suspected ovarian cancer. MRI is widely used for preoperative characterization and risk stratification of adnexal masses. While epithelial ovarian cancer (EOC) has typical findings on MRI; there are several benign and malignant pelvic conditions that may mimic its appearance on imaging. Knowledge of the origin and imaging characteristics of a pelvic mass will help radiologists diagnose ovarian cancer promptly and accurately. Finally, in special subgroups, including adolescents and gravid population, the prevalence of various ovarian tumors differs from that of the general population and there are conditions which uniquely manifest during these periods of life.

Diagnostic Performance Evaluation of Multiparametric Magnetic Resonance Imaging in the Detection of Prostate Cancer with Supervised Machine Learning Methods

Prostate cancer is the second leading cause of cancer-related death in men. Its early and correct diagnosis is of particular importance to controlling and preventing the disease from spreading to other tissues. Artificial intelligence and machine learning have effectively detected and graded several cancers, in particular prostate cancer. The purpose of this review is to show the diagnostic performance (accuracy and area under the curve) of supervised machine learning algorithms in detecting prostate cancer using multiparametric MRI. A comparison was made between the performances of different supervised machine-learning methods. This review study was performed on the recent literature sourced from scientific citation websites such as Google Scholar, PubMed, Scopus, and Web of Science up to the end of January 2023. The findings of this review reveal that supervised machine learning techniques have good performance with high accuracy and area under the curve for prostate cancer diagnosis and prediction using multiparametric MR imaging. Among supervised machine learning methods, deep learning, random forest, and logistic regression algorithms appear to have the best performance.