Abstract
Objectives: This study aimed to combine dynamic magnetic resonance imaging (MRI) with three-dimensional (3D) reconstruction and form a plane based on osseous structures to evaluate the degree of pelvic organ prolapse (POP). The correlation of this novel evaluation approach with the POP-Q system was assessed. Methods: A retrospective analysis was conducted on 71 POP patients with POP-Q stage ≥II. The dynamic MRI images of those patients were reconstructed in three dimensions. A plane was created by using the midpoint of the line between the inferior margins of the two pubic bones and the starting points of the superior margins of the bilateral sacrotuberous ligaments (the pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane). Distances from the lowest point of the anterior vaginal wall, cervix, and rectal ampulla to this evaluation plane were measured, modeled, and categorized. The consistency and correlation of the categorized results with POP-Q scores were verified by performing a kappa analysis and Spearman’s rank correlation analysis, respectively. Results: The highest consistency with POP-Q scores was found in the prolapse of the central pelvic cavity (kappa = 0.713, p < 0.05), followed by the anterior POP (kappa = 0.427, p < 0.05), and posterior POP (kappa = 0.261, p < 0.05), with all showing statistically significant differences. The strongest positive correlation was observed between central POP and POP-Q scores (r = 0.864, p < 0.01), followed by posterior POP and POP-Q scores (r = 0.710, p < 0.01), with both exhibiting a strong positive correlation. Anterior POP and POP-Q scores showed a moderate positive correlation (r = 0.586, p < 0.01). Conclusions: The results of the proposed evaluation method were highly consistent in the anterior and central pelvic cavities and strongly correlated in the central and posterior pelvic cavities. In particular, the assessment of the posterior cavity showed a strong positive correlation with that of the POP-Q system. The evaluation plane demonstrated high consistency and correlation with the POP-Q system.
Introduction
Female pelvic organ prolapse (FPOP) refers to the downward displacement of one or pelvic organs, including the anterior vaginal wall, posterior vaginal wall, uterus, and vaginal apex, due to weakened pelvic floor support function caused by factors such as congenital development, pelvic floor tissue degeneration, and trauma [1]. It is a common pelvic floor dysfunction affecting the quality of life of middle-aged and older women, and its global incidence is increasing with the world’s aging population. The World Health Organization (WHO) found that the prevalence rate of FPOP ranged from 3.8% to 14.2% among women aged 50–79 years [2]. The most specific symptom of prolapse is the sensation or visibility of a vaginal bulge. Prolapse of the bladder, rectum, and uterus may be associated with a range of pelvic floor symptoms, including incomplete emptying of the bladder and bowels, urinary frequency, urinary and fecal incontinence, and sexual dysfunction [3]. FPOP is considered a chronic condition that severely impacts a woman’s health, quality of life, and social activities.
The International Continence Society (ICS) Pelvic Organ Prolapse Quantification (POP-Q) system has served as the gold standard for quantifying anatomical prolapse detected in clinical examinations since 1996 [4]. However, the POP-Q system characterizes the degree of prolapse solely on the distance of prolapsed organs from the hymen along the vertical axis. Clinically, additional diagnostic tools are needed to decide between conservative and surgical treatment options.
Magnetic resonance imaging (MRI) offers high resolution of soft tissues and thus enables a preliminary assessment of pelvic internal structural changes and damage in FPOP patients. Images of pelvic floor muscle groups at each level are used to identify defects and visualize morphological changes in pelvic floor connective tissues [5, 6]. However, international studies have revealed the absence of a grading system that can be used with existing MRI systems and correlates well with clinical POP-Q evaluations [7, 8]. The main reason is that MRI assessments are conducted in two-dimensional (2D) sagittal planes. Inaccurate positioning of anatomical landmarks and discrepancies between reference lines and clinical POP-Q landmarks can lead to inaccurate prolapse grading evaluations. In recent years, the application of dynamic MRI in FPOP assessments has gained extensive attention due to its advantages such as high resolution, repeatability, and exceptional soft tissue contrast [9]. Dynamic MRI of the pelvic floor can comprehensively evaluate the anatomical and functional characteristics of pelvic walls and pelvic organs [10‒12]. Dynamic sequence images facilitate the quantification of prolapse severity [13], and dynamic MRI is regarded as a valuable imaging tool [14]. Studies have demonstrated that it is superior to ultrasound for POP patients [15] and offers more anatomical details due to its higher resolution capability [14].
Through three-dimensional (3D) reconstruction of dynamic MRI sequence images, this study analyzed the correlation and consistency of POP severity assessments made by 3D MRI reconstruction and clinical POP-Q assessments by using a standardized evaluation plane. Furthermore, it explored the value of using MRI dynamic sequence 3D reconstruction combined with a novel quantification methodology for making objective assessments of POP.
Materials and Methods
Study Subjects
Patients who presented with symptoms of FPOP and visited the Department of Obstetrics and Gynecology at Nanfang Hospital, Southern Medical University, from May 2022 to January 2024, were enrolled in this study. After researchers carefully inquired about clinical symptoms, the patients completed a PFDI-20 questionnaire and were assigned POP-Q scores. A total of 71 patients were diagnosed with moderate to severe FPOP. All patients underwent a preoperative dynamic MRI examination. The study exclusion criteria included concurrent diseases affecting pelvic anatomical morphology, such as a giant uterine fibroid or pelvic mass, and contraindications for MRI examination. Informed consent was obtained from all participants. The study protocol was approved by the Ethics Committee of Nanfang Hospital, Southern Medical University (No. 20240220).
POP-Q
POP-Qs were performed and recorded by urogynecology specialists according to the POP-Q system [4].
MRI Equipment
A Philips Achieva TX 3.0T MRI system (The Netherlands) equipped with an 8-channel abdominal coil was used for scanning. All participants performed the Valsalva maneuver. First, a T2-weighted single shot dynamic sequence (T2W ssh-sag-DYN) in the midsagittal plane was scanned with the following parameters: repetition time (TR)/echo time (TE): 705/120 ms; slice thickness: 6.0 mm; slice gap: default; flip angle: 90°; field of view: 360 mm × 300 mm × 6 mm; matrix: 240 × 200. Subsequently, the participants were instructed to perform the Valsalva maneuver 3 times and maintain it while scanning the T2-weighted SSTSE sequence in transverse, sagittal, and coronal planes. The transverse plane extended from the fundus of the uterine to the distal prolapse. The sagittal plane extended bilaterally to the ischial spine. The coronal plane extended from the posterior pubic symphysis to the sacrococcygeal region. Specific parameters for dynamic scanning included repetition time/echo time: 688/80 ms; flip angle: 90°; field of view: 260 mm × 300 mm × 179 mm; matrix: 184 × 188; slice thickness: 4.0 mm; slice gap: 1 mm. The scanning durations in the transverse, sagittal, and coronal planes were approximately 20 s each.
3D Reconstruction, Plane Establishment, Distance Measurement
Image Selection and 3D Reconstruction
MRI acquisition was performed using T2-weighted SSTSE sequences. Raw DICOM 3.0 data were directly imported into Mimics 21.0 software (Materialise, Belgium) for 3D reconstruction. The transverse plane was selected as the primary orientation for editing. Pelvic viscera, including the uterus, vagina, bladder, urethra, and rectum were systematically identified across all MRI sections. Segmentation masks defining regions of interest were digitally created using semiautomated thresholding, after which erasing and drawing tools in the software were used to refine mask boundaries and eliminate irrelevant noises at the edges and redundant data. Next, the contour in the sagittal position was confirmed, ultimately deriving accurate boundaries of pelvic organ structures. All images were processed, converted into the 3D mode, and saved in STL format (Fig. 1a, b).
3D reconstruction of pelvic organs from T2-weighted MRI using Mimics software. a Editing the cross-section of a T2-weighted SSTSE sequence. b Contour confirmation in the sagittal position of the T2-weighted SSTSE sequence. Green represents the bladder, light pink denotes the anterior wall of the vagina, deep pink signifies the posterior wall of the vagina, yellow represents the uterus, and brown denotes the rectum.
3D reconstruction of pelvic organs from T2-weighted MRI using Mimics software. a Editing the cross-section of a T2-weighted SSTSE sequence. b Contour confirmation in the sagittal position of the T2-weighted SSTSE sequence. Green represents the bladder, light pink denotes the anterior wall of the vagina, deep pink signifies the posterior wall of the vagina, yellow represents the uterus, and brown denotes the rectum.
Establishment of Evaluation Plane
The theoretical bases for the evaluation plane included the following:
- (1)
The anatomical position of the pubic symphysis remained fixed and did not change during the Valsalva maneuver. This aided in locating pelvic organ positions and measuring mobility.
- (2)
The plane established based on osseous structures possessed improved stability.
- (3)
Use of the plane in prolapse assessment provided for more stereo and intuitive outcomes.
The reference line used for POP-Q scoring was closely aligned with the hymenal edge, which was nearly at the same level as the lower edge of the pubic symphysis. However, it is important to note that the POP-Q score is highly subjective, and the hymenal margin is not distinctly visible on MRI. In contrast, the bony structures are relatively fixed, providing a clearer delineation than the hymen. In the 3D reconstructed 2D sagittal image, we drew a parallel line adjacent to the lower edge of the pubic symphysis, which was then translated horizontally into a plane. This line was subsequently matched with the 3D structure of the pelvis following reconstruction, allowing us to identify the corresponding anatomical landmark, which landmark corresponded to the upper edge of the sacrotuberous ligament. By utilizing the built-in program of Mimics software, the surface was defined by three points: the vertex, which is the midpoint of the lower edges of the two pubic symphyses (set using the “Midpoint” function in Mimics software), and two other points, which were the upper edges of the bilateral sacrotuberous ligaments (Fig. 2a–c).
Geometric shapes of visualized bone structures and selection of reference point positions lateral view (a); frontal view (b, c).
Geometric shapes of visualized bone structures and selection of reference point positions lateral view (a); frontal view (b, c).
Measurement of Distances from Points to the Plane
By establishing 3D images of pelvic and prolapsed pelvic organs using the Image Selection and 3D Reconstruction method, a plane was formed employing the midpoint of the line between the inferior margins of the two pubic bones and the starting points of the superior margins of the bilateral sacrotuberous ligaments, designated as the “pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane” (PIM-SSL plane) (as shown in Fig. 3a–c). Next, Mimics 21.0 software was used to gauge distances according to the following steps: (1) the measurement points included the lowest point of the anterior vaginal wall (corresponding to point Ba), the external os of the cervix (corresponding to point C), and the rectal ampulla (corresponding to point Bp); (2) perpendicular lines from the measurement points to the plane were depicted using system tools; (3) the intersection points where those perpendicular lines met the evaluation plane in step 2 were considered to be the projection points of the measurement points, and (4) the distances between the measurement points and their respective projection points were determined (Fig. 3d–g take the uterus as an example, as shown below).
3D measurement of pelvic organ projection distances relative to the PIM-SSL plane. a A frontal view of the “pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane.” b A dorsal view of the “pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane.” c A left view of the “pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane.” d The vertical line from the measurement point (external os of the cervix) to the plane. e The intersection point between the measurement point (external os of the cervix) and the planar perpendicular line. f The distance between the measurement point (external os of the cervix) and the intersection point of the planar perpendicular line. g Automatic distance measurement.
3D measurement of pelvic organ projection distances relative to the PIM-SSL plane. a A frontal view of the “pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane.” b A dorsal view of the “pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane.” c A left view of the “pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane.” d The vertical line from the measurement point (external os of the cervix) to the plane. e The intersection point between the measurement point (external os of the cervix) and the planar perpendicular line. f The distance between the measurement point (external os of the cervix) and the intersection point of the planar perpendicular line. g Automatic distance measurement.
Spearman’s Rank Correlation Analysis
A Spearman’s rank correlation analysis was performed on the following values:
- (1)
The distance from the Ba point to the edge of the hymen during the Valsalva maneuver in clinical POP-Q and the vertical distance from the lowest point of the anterior vaginal wall to the evaluation plane in the 3D reconstruction model.
- (2)
The distance from the C point to the edge of the hymen during the Valsalva maneuver in clinical POP-Q and the vertical distance from the external os of the cervix to the evaluation plane in the 3D reconstruction model.
- (3)
The distance from the Bp point to the edge of the hymen during the Valsalva maneuver in clinical POP-Q and the vertical distance from the rectal ampulla to the evaluation plane in the 3D reconstruction model.
According to Spearman’s rank correlation analysis [16], a p value <0.05 was considered to be statistically significant.
Basic Data Processing
In this study, clinical physical examination results were quantitatively graded using the POP-Q system. Upon 3D reconstruction of MRI examination results, a reference plane was set and defined as the “pubic inferior midpoint – sacrotuberous ligament starting point superior edge plane.” This reference plane facilitated the comparative analysis between clinical POP-Q results and MRI 3D reconstruction results. With regard to the POP-Q scoring system, the area above the reference plane was denoted as negative, and the area below was marked as positive.
Statistical Method
The data were analyzed using SPSS 26.0 software.
Modeling
The grading categories used for the POP-Q system the proposed classifications were designated as normal, mild, moderate, and severe. Datasets for 15 normal control subjects and 71 POP cases from the experimental patient group were input, and the number of categories was set to four. A K-means clustering analysis was used to derive approximate judgment levels. The rationality of the original data classification was assessed through the category containing the outliers. If inclusion of the outliers led to an unreasonable classification, they were removed, and the K-means clustering analysis was repeated to obtain the optimal data classification.
Finally, scatter plots were created to illustrate the grading results of patients using the POP-Q and proposed classification methods. The boundaries of grades were demarcated by dashed lines in corresponding colors. Based on those results, the differences and rationality of the two classification methods were analyzed.
Scatter plots were created to show the grades of patients when using the POP-Q and proposed classification methods. Dashed lines in corresponding colors represented grade boundaries. In Figure 4a–c, the dashed lines intersect, indicating that POP-Q grading did not strictly comply with grading criteria. In contrast, the dashed lines in Figure 4d and e exhibit clear boundaries without intersections, demonstrating that the proposed method can enhance classification accuracy based on imaging information.
Comparative evaluation of POP-Q grading vs. imaging-enhanced classification via K-means clustering. a The prolapse degree of the anterior vaginal wall and the grading criteria of the POP-Q system. b The prolapse degree of the posterior vaginal wall and the grading criteria of the POP-Q system. c The prolapse degree of the uterus and the grading criteria of the POP-Q system. d The prolapse degree of the anterior vaginal wall and the grading criteria of the proposed planar evaluation system. e The prolapse degree of the posterior vaginal wall and the grading criteria of the proposed planar evaluation system. f The prolapse degree of the uterus and the grading criteria of the proposed plane evaluation system.
Comparative evaluation of POP-Q grading vs. imaging-enhanced classification via K-means clustering. a The prolapse degree of the anterior vaginal wall and the grading criteria of the POP-Q system. b The prolapse degree of the posterior vaginal wall and the grading criteria of the POP-Q system. c The prolapse degree of the uterus and the grading criteria of the POP-Q system. d The prolapse degree of the anterior vaginal wall and the grading criteria of the proposed planar evaluation system. e The prolapse degree of the posterior vaginal wall and the grading criteria of the proposed planar evaluation system. f The prolapse degree of the uterus and the grading criteria of the proposed plane evaluation system.
Result Analysis
Scatter Plot
As depicted in Figures 4a–c, the scatter plots of the POP-Q method displayed intersecting dashed lines, indicating that the POP-Q grading did not completely follow grading requirements. By comparison, in Figures 4d–f, the dashed lines of the proposed classification method show clear boundaries without intersection, suggesting that this approach can further accurately classify patients based on imaging information.
Kappa Consistency Test
In this study, we completed 3D reconstruction of the pelvis and prolapsed organs in 71 POP patients. Moreover, the distances from the respective points to the evaluation plane were gauged. Next, after modeling and grouping, the measured values were compared with the POP-Q scores, and the following conclusions were reached. When using the evaluation plane as the standard and based on the above modeling criteria, detailed POP level in Table 1.
Spearman’s rank correlation analysis of the POP-Q system and the evaluation plane
. | Ⅰ . | Ⅱ . | Ⅲ . | Ⅳ . | rs . | p value . |
---|---|---|---|---|---|---|
POP-Q anterior pelvic cavity | 1 (1.4%) | 16 (22.5%) | 47 (66.2%) | 7 (9.9%) | 0.586 | <0.001 |
The evaluation plane – anterior pelvic cavity | - | 28 (39.4%) | 28 (39.4%) | 14 (19.7%) | ||
POP-Q posterior pelvic cavity | 10 (14.1%) | 45 (63.4%) | 9 (12.7%) | 7 (9.9%) | 0.71 | <0.001 |
The evaluation plane – posterior pelvic cavity | - | 32 (45.1%) | 31 (43.7%) | 8 (11.3%) | ||
POP-Q central pelvic cavity | 3 (4.2%) | 19 (26.8%) | 36 (50.7%) | 13 (18.3%) | 0.864 | <0.001 |
The evaluation plane – central pelvic cavity | 2 (2.8%) | 17 (23.9%) | 30 (42.3%) | 20 (28.2%) |
. | Ⅰ . | Ⅱ . | Ⅲ . | Ⅳ . | rs . | p value . |
---|---|---|---|---|---|---|
POP-Q anterior pelvic cavity | 1 (1.4%) | 16 (22.5%) | 47 (66.2%) | 7 (9.9%) | 0.586 | <0.001 |
The evaluation plane – anterior pelvic cavity | - | 28 (39.4%) | 28 (39.4%) | 14 (19.7%) | ||
POP-Q posterior pelvic cavity | 10 (14.1%) | 45 (63.4%) | 9 (12.7%) | 7 (9.9%) | 0.71 | <0.001 |
The evaluation plane – posterior pelvic cavity | - | 32 (45.1%) | 31 (43.7%) | 8 (11.3%) | ||
POP-Q central pelvic cavity | 3 (4.2%) | 19 (26.8%) | 36 (50.7%) | 13 (18.3%) | 0.864 | <0.001 |
The evaluation plane – central pelvic cavity | 2 (2.8%) | 17 (23.9%) | 30 (42.3%) | 20 (28.2%) |
The correlation is significant at a confidence level (two-tailed) of 0.01. All values above were measured during the Valsalva maneuver.
The diagnosis results of the evaluation plane were compared with the POP-Q results using the Kappa consistency test. The consistency of uterine prolapse was the highest across the anterior, central, and posterior pelvic cavities (Kappa = 0.713, p < 0.01, as shown in Table 2). This was followed by anterior vaginal wall prolapse (Kappa = 0.427, p < 0.01, as shown in Table 3) and posterior vaginal wall prolapse (Kappa = 0.261, p < 0.05, as shown in Table 4).
Consistency between the evaluation plane and POP-Q for uterine prolapse
. | . | POP – uterus . | . | . | . | Total . | Kappa . | p value . |
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 0.713 | 0 | |||
The evaluation plane – uterus | 1 | 2 | 0 | 0 | 0 | 2 | ||
2 | 1 | 16 | 0 | 0 | 17 | |||
3 | 0 | 2 | 27 | 1 | 30 | |||
4 | 0 | 0 | 9 | 11 | 20 | |||
Total | 3 | 18 | 36 | 12 | 69 |
. | . | POP – uterus . | . | . | . | Total . | Kappa . | p value . |
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 0.713 | 0 | |||
The evaluation plane – uterus | 1 | 2 | 0 | 0 | 0 | 2 | ||
2 | 1 | 16 | 0 | 0 | 17 | |||
3 | 0 | 2 | 27 | 1 | 30 | |||
4 | 0 | 0 | 9 | 11 | 20 | |||
Total | 3 | 18 | 36 | 12 | 69 |
Consistency between the evaluation plane and POP-Q for anterior vaginal wall prolapse
. | . | POP – anterior vaginal wall . | . | . | . | Total . | Kappa . | p value . |
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 0.427 | 0 | |||
The evaluation plane – anterior vaginal wall | 2 | 1 | 14 | 13 | 0 | 28 | ||
3 | 0 | 1 | 26 | 1 | 28 | |||
4 | 0 | 1 | 8 | 5 | 14 | |||
Total | 1 | 16 | 47 | 6 | 70 |
. | . | POP – anterior vaginal wall . | . | . | . | Total . | Kappa . | p value . |
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 0.427 | 0 | |||
The evaluation plane – anterior vaginal wall | 2 | 1 | 14 | 13 | 0 | 28 | ||
3 | 0 | 1 | 26 | 1 | 28 | |||
4 | 0 | 1 | 8 | 5 | 14 | |||
Total | 1 | 16 | 47 | 6 | 70 |
Consistency between the evaluation plane and POP-Q for posterior vaginal wall prolapse
. | . | POP – posterior vaginal wall . | . | . | . | Total . | Kappa . | p value . |
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 0.261 | 0 | |||
The evaluation plane – posterior vaginal wall | 2 | 10 | 22 | 0 | 0 | 32 | ||
3 | 0 | 23 | 8 | 0 | 31 | |||
4 | 0 | 0 | 1 | 7 | 8 | |||
Total | 10 | 45 | 9 | 7 | 71 |
. | . | POP – posterior vaginal wall . | . | . | . | Total . | Kappa . | p value . |
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 0.261 | 0 | |||
The evaluation plane – posterior vaginal wall | 2 | 10 | 22 | 0 | 0 | 32 | ||
3 | 0 | 23 | 8 | 0 | 31 | |||
4 | 0 | 0 | 1 | 7 | 8 | |||
Total | 10 | 45 | 9 | 7 | 71 |
Spearman’s Correlation Analysis
A Spearman’s correlation analysis was conducted to compare clinical POP-Q results and the diagnostic results of the assessment plane. The central pelvic cavity showed a strong positive correlation, which was the most significant (r = 0.864, p < 0.01). It was followed by the posterior pelvic cavity with a strong positive correlation (r = 0.71, p < 0.01) and the anterior pelvic cavity with a moderate positive correlation (r = 0.586, p < 0.01). Details of the analysis are presented in Table 1.
Imaging Analysis
3D reconstruction can intuitively and comprehensively display prolapsed pelvic organs from multiple orientations and angles (Fig. 5a, b). This study established an assessment plane based on osseous structures. Through modeling, classification, and correlation comparison, it was confirmed that using the assessment plane after 3D reconstruction could efficiently produce results that corresponded with POP-Q scores. The advantage of imaging lies in its intuitive nature, and it is less influenced by the subjective factors of examining physicians. This allows it to provide a further objective conclusion regarding the severity of prolapse. After accumulating a specific number of prolapse patients, imaging-based classification can be performed using the data gathered from 3D reconstruction and the assessment plane. Combining imaging-based classification and POP-Q scoring can promote the objectivity of POP diagnoses, thereby favoring the formulation of personalized treatment plans.
3D pelvic prolapse reconstruction and objective assessment via osseous landmarks vs. POP-Q. a A stereoscopic image of 3D reconstruction (showing the pelvis). b A stereoscopic image of 3D reconstruction (hiding the pelvis).
3D pelvic prolapse reconstruction and objective assessment via osseous landmarks vs. POP-Q. a A stereoscopic image of 3D reconstruction (showing the pelvis). b A stereoscopic image of 3D reconstruction (hiding the pelvis).
Discussion
Surgical treatment remains the most effective approach for patients with moderate to severe POP. However, these procedures carry a high recurrence rate of nearly 40% after autologous tissue repair. Moreover, the anterior pelvic cavity is the most common recurrence site (approximately 13%) [17]. Scholars worldwide generally attribute this to two factors: (1) the pelvic floor functions as a balanced entity, and optimal surgical treatment should repair damage while maintaining a holistic mechanical balance of the pelvic floor [18], and (2) FPOP evaluation methods are imperfect. Although POP-Q scores are widely used in clinical practice due to their convenience and cost-effectiveness, the POP-Q system has its limitations. Moreover, the poor consistency of results obtained from different 2D MRI assessment systems may result from the following factors: (1) difficulty in identifying anatomical landmarks, leading to inaccurate positioning, and (2) assessment in MRI 2D sagittal planes limits the selection of points that can represent actual organ positions. Some research performed outside of China [19‒21] has suggested that merely using dynamic MRI results and existing sagittal plane assessment lines to assess the degree of prolapse produced results that poorly correlated with POP-Q scores and significantly differed from the clinical manifestations of patients.
3D reconstruction technology converts MRI 2D image datasets into 3D mode, vividly depicting the holistic anatomical structure of the pelvic cavity. Dynamic MRI examination provides real-time and visualized observations of the entire process of pelvic organ descent, displacement, initiation, and reaching the lowest point of prolapse. In 3D reconstructed stereo images, distances can be measured with greater accuracy using defined planes. Currently, this approach remains in the experimental evaluation stage, but our team has observed strong consistency and correlation with POP-Q scores. This plane will eventually be utilized to develop automated assessment software that can be integrated with an MRI examination system.
MRI provides a comprehensive 3D representation of anterior, middle, and posterior pelvic prolapse. In the sagittal plane of an MRI image, Mimics software was utilized to delineate these structures and facilitate 3D reconstruction and planar assessment, thereby enhancing the clinician’s ability to simulate a POP-Q examination from an imaging perspective. Our evaluation plane was established based on a relatively fixed bony structure; however, it is unique to each patient, reflecting individual anatomical variations.
By using a plane based on osseous structures (the detailed description and theoretical bases are mentioned above) to assess the degree of POP, the stability was superior to that of assessment lines in MRI 2D sagittal planes. This is because the plane chosen in this study anatomically aligned closest to the baseline (hymen edge) of the POP-Q system. Evaluating the degree of POP by using the plane after dynamic MRI 3D reconstruction, combined with the POP-Q system, can help clinicians gain a more intuitive understanding of pelvic organ positions and prolapse severity, enabling highly accurate assessments of POP and the creation of personalized surgical approaches.
Limitations
This study has the following limitations; however, the methods used can be further optimized and standardized. (1) The impact of heterogeneity in MRI equipment on results needs to be considered. If an MRI examination does not support dynamic sequences, the clinician should instruct the patient to exert force to the maximum extent of pelvic organ prolapse prior to the MRI examination. If dynamic MRI sequences are available, the evaluation should be conducted according to the “MRI Examination Instruments and Equipment” section of this article. (2) The 3D reconstruction of MRI images presents significant challenges, typically necessitating a team of urogynecologists who are skilled in operating Mimics software and possess a thorough understanding of pelvic floor anatomy.
Conclusion
In this study, a plane evaluation after a 3D reconstruction of dynamic MRI was compared with the POP-Q system. The outcomes demonstrated that results for the anterior and central pelvic cavities were highly consistent with those provided by the POP-Q system. The correlations of the anterior, central, and posterior pelvic cavities with the POP-Q evaluation system were strong, particularly for the posterior cavity, which was strongly positively correlated with the POP-Q evaluation. This suggests that the imaging evaluation of POP may be associated with a better replication of actual organ structures by the evaluation plane based on the osseous anatomy and the 3D reconstruction. Particularly, the evaluation of the posterior pelvic cavity showed a higher correlation with that of the POP-Q system. The novel integration of dynamic MRI with 3D reconstruction and the evaluation plane provides for potential enhancements in the diagnostic accuracy of FPOP. It also provides a basis for refined and personalized diagnoses and can assist clinicians in formulating tailored surgical strategies, which ultimately improve surgical outcomes and enhance the quality of life for patients.
Statement of Ethics
This study was performed in accordance with the Declaration of Helsinki. This human study was approved by Nanfang Hospital, Southern Medical University – approval No. 20240220. All adult participants provided written informed consent to participate in this study.
Conflict of Interest Statement
The author(s) declare no conflicts of interest.
Funding Sources
This project was supported by National Health Commission Medical and Health Science and Technology Development Fund (No. HDSL202003000).
Author Contributions
L.S.: conceptualization, data curation, and formal analysis. H.B. and X.S.: investigation and methodology. P.L.: conceptualization and writing – review and editing. C.C.: visualization and writing – original draft. All authors read and approved the final manuscript.
Data Availability Statement
Due to the fact that the data contain information that may compromise the privacy of research participants, the data supporting the results of this study are not publicly available but can be obtained from P.L. and C.C. upon reasonable request.