Comprehensive feature selection for classifying the treatment outcome of high-intensity ultrasound therapy in uterine fibroids

Machine learning models for prediction of NPVR ≥80% with HIFU ablation for uterine fibroids

BACKGROUND

Currently high-intensity focused ultrasound (HIFU) is widely used to treat uterine fibroids (UFs). The aim of this study is to develop a machine learning model that can accurately predict the efficacy of HIFU ablation for UFs, assisting the preoperative selection of suitable patients with UFs.

METHODS

This study collected data from 1,000 patients with UFs who underwent ultrasound-guided high-intensity focused ultrasound. The least absolute shrinkage and selection operator (LASSO) regression was used for multidimensional feature screening. Five machine learning algorithms such as logistic regression, random forest, extreme gradient boosting (XGBoost), artificial neural network, and gradient boosting decision tree were utilized to predict ablation efficacy. The efficacy was quantified by the non-perfused volume ratio (NPVR), which was classified into two categories: NPVR <80% and NPVR ≥80%.

RESULTS

The XGBoost model proved to be the most effective, showing the highest AUC of 0.692 (95% CI: 0.622-0.762) in the testing data set. The four key predictors were T2 weighted image, the distance from ventral side of UFs to skin, platelet count, and contrast-enhanced T1 weighted image.

CONCLUSIONS

The machine learning prediction model in this study showed significant potential for accurately predicting the preoperative efficacy of HIFU ablation for UFs. These insights were important for clinicians in the preoperative assessment and selection of patients, which could enhance the precision of treatment planning.

A multi-model machine learning framework for breast cancer risk stratification using clinical and imaging data

PurposeThis study presents a comprehensive machine learning framework for assessing breast cancer malignancy by integrating clinical features with imaging features derived from deep learning.MethodsThe dataset included 1668 patients with documented breast lesions, incorporating clinical data (e.g., age, BI-RADS category, lesion size, margins, and calcifications) alongside mammographic images processed using four CNN architectures: EfficientNet, ResNet, DenseNet, and InceptionNet. Three predictive configurations were developed: an imaging-only model, a hybrid model combining imaging and clinical data, and a stacking-based ensemble model that aggregates both data types to enhance predictive accuracy. Twelve feature selection techniques, including ReliefF and Fisher Score, were applied to identify key predictive features. Model performance was evaluated using accuracy and AUC, with 5-fold cross-valida tion and hyperparameter tuning to ensure robustness.ResultsThe imaging-only models demonstrated strong predictive performance, with EfficientNet achieving an AUC of 0.76. The hybrid model combining imaging and clinical data reached the highest accuracy of 83% and an AUC of 0.87, underscoring the benefits of data integration. The stacking-based ensemble model further optimized accuracy, reaching a peak AUC of 0.94, demonstrating its potential as a reliable tool for malignancy risk assessment.ConclusionThis study highlights the importance of integrating clinical and deep imaging features for breast cancer risk stratification, with the stacking-based model.

Predicting Short-term and Long-term Efficacy of HIFU Treatment for Uterine Fibroids Based on Clinical Information and MRI: A Retrospective Study

RATIONALE AND OBJECTIVES

This study aimed to address the challenge of predicting treatment outcomes for patients with uterine fibroids undergoing high-intensity focused ultrasound (HIFU) ablation. We developed medical-assisted diagnostic models to accurately predict the ablation rates and volume reduction rates, thus assessing both short-term and long-term treatment effects of fibroids.

MATERIALS AND METHODS

For the ablation rate prediction, our study included 348 fibroids, categorized into 181 fully ablated and 167 inadequately ablated fibroids. Using multimodal MRI sequences and clinical characteristics, coupled with data preprocessing steps such as feature extraction, testing, and screening, we constructed an ensemble model for predicting preoperative ablation rates. In the volume reduction rate study, we analyzed 253 fibroids, divided into 142 high-volume responders and 111 low-volume responders. Based on clinical characteristics and T2-weighted image (T2WI) sequences, along with lesion delineation, feature normalization, and other preprocessing steps, we developed an inter-slice information fusion model for predicting preoperative volume reduction rates.

RESULTS

The ensemble model demonstrated an accuracy of 0.800 and an area under the curve (AUC) of 0.830 on the test set, while the inter-slice information fusion model achieved an accuracy of 0.808 and an AUC of 0.891. Both models showed superior predictive performance compared to existing models.

CONCLUSION

The ensemble and inter-slice information fusion models developed in this study exhibit robust predictive capabilities, offering valuable support for clinicians in selecting patients for HIFU treatment. These models hold potential for enhancing patient outcomes through tailored treatment planning.

Application of transabdominal ultrasound- and laparoscopy-guided percutaneous microwave ablation for treating uterine fibroids: 24-month follow-up outcomes

PURPOSE

To determine the ablation efficacy of transabdominal ultrasound- and laparoscopy-guided percutaneous microwave ablation (PMWA), to investigate whether the risk of damage to adjacent organs and endometrium due to this technique can be reduced or even avoided. We also evaluated the clinical efficacy of this technique in the treatment of uterine fibroids of different sizes and at different locations over a 24-month follow-up period.

METHODS

This study included 50 patients with uterine fibroids who underwent transabdominal ultrasound- and laparoscopy-guided PMWA from August 2018 to July 2020. Lesions were confirmed by pathology. The technical efficacy and complications of PMWA were assessed. The lesion diameter, lesion volume, lesion location, and contrast-enhanced ultrasound (CEUS) features before PMWA and within 24 h after PMWA were recorded. Magnetic resonance imaging (MRI) was used for follow-up at 3 and 6 months after PMWA. Transvaginal ultrasound was used for follow-up at 24 months after PMWA.

RESULTS

A total of 50 patients with uterine fibroids received treatment. The median ablation rate of uterine fibroids was 97.21%. The mean lesion volume reduction rates were 32.63%, 57.26%, and 92.64% at 3, 6, and 24 months after treatment, respectively. The size and location of uterine fibroids did not significantly affect the ablation rate and the rate of lesion volume reduction. No major complication was found during and after the procedure.

CONCLUSION

Transabdominal ultrasound- and laparoscopy-guided PMWA can be utilized to safely enhance the ablation rate while minimizing ablation time and avoiding harm to adjacent organs and the endometrium. This technique is applicable for treating uterine fibroids of different sizes and at varying locations.

TRIAL REGISTRATION NUMBER

ChiCTR-IPR-17011910, and date of trial registration: 08/07/2017.

Development and validation of a radiomics model based on T2-weighted imaging for predicting the efficacy of high intensity focused ultrasound ablation in uterine fibroids

Background

The heterogeneity of uterine fibroids in magnetic resonance imaging (MRI) is complex for a subjective visual evaluation, therefore it is difficult for an accurate prediction of the efficacy of high intensity focused ultrasound (HIFU) ablation in fibroids before the treatment. The purpose of this study was to set up a radiomics model based on MRI T2-weighted imaging (T2WI) for predicting the efficacy of HIFU ablation in uterine fibroids, and it would be used in preoperative screening of the fibroids for achieving high non-perfused volume ratio (NPVR).

Methods

A total of 178 patients with uterine fibroids were consecutively enrolled and treated with ultrasound-guided HIFU under conscious sedation between February 2017 and December 2021. Among them, 96 patients with 108 uterine fibroids with high ablation efficacy (NPVR ≥80%, h_NPVR) and 82 patients with 92 fibroids with lower ablation efficacy (NPVR <80%, l_NPVR) were retrospectively analyzed. The transverse T2WI images of fibroids were selected, and the fibroids were delineated slice by slice using ITK-SNAP software. The radiomics analysis was performed to find the imaging biomarker for the construction of a predicting model for the evaluation of the ablation efficacy, including the feature extraction, feature selection and model construction. The prediction model was built by logistic regression and assessed by receiver operating characteristic (ROC) curve, and the prediction efficiency of the two models was compared by Delong test. The ratio of the training set to the testing set was 8:2.

Results

The logistic regression model showed that the mean area under the curve (AUC) of the training set was 0.817 [95% confidence interval (CI): 0.755-0.882], and the testing set was 0.805 (95% CI: 0.670-0.941), respectively, which indicated a strong classification ability. The Delong test showed that there was no significant difference in the area under the ROC curve between the training set and testing set (P>0.05).

Conclusions

The radiomics model based on T2WI is feasible and effective for predicting the efficacy of HIFU ablation in treatment of uterine fibroids.

Prediction of non-perfusion volume ratio for uterine fibroids treated with ultrasound-guided high-intensity focused ultrasound based on MRI radiomics combined with clinical parameters

BACKGROUND

Prediction of non-perfusion volume ratio (NPVR) is critical in selecting patients with uterine fibroids who will potentially benefit from ultrasound-guided high-intensity focused ultrasound (HIFU) treatment, as it reduces the risk of treatment failure. The purpose of this study is to construct an optimal model for predicting NPVR based on T2-weighted magnetic resonance imaging (T2MRI) radiomics features combined with clinical parameters by machine learning.

MATERIALS AND METHODS

This retrospective study was conducted among 223 patients diagnosed with uterine fibroids from two centers. The patients from one center were allocated to a training cohort (n = 122) and an internal test cohort (n = 46), and the data from the other center (n = 55) was used as an external test cohort. The least absolute shrinkage and selection operator (LASSO) algorithm was employed for feature selection in the training cohort. The support vector machine (SVM) was adopted to construct a radiomics model, a clinical model, and a radiomics-clinical model for NPVR prediction, respectively. The area under the curve (AUC) and the decision curve analysis (DCA) were performed to evaluate the predictive validity and the clinical usefulness of the model, respectively.

RESULTS

A total of 851 radiomic features were extracted from T2MRI, of which seven radiomics features were screened for NPVR prediction-related radiomics features. The radiomics-clinical model combining radiomics features and clinical parameters showed the best predictive performance in both the internal (AUC = 0.824, 95% CI 0.693-0.954) and external (AUC = 0.773, 95% CI 0.647-0.902) test cohorts, and the DCA also suggested the radiomics-clinical model had the highest net benefit.

CONCLUSIONS

The radiomics-clinical model could be applied to the NPVR prediction of patients with uterine fibroids treated by HIFU to provide an objective and effective method for selecting potential patients who would benefit from the treatment mostly.

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

Fibroids of the uterus are a common benign tumor affecting women of childbearing age. Uterine fibroids (UF) can be effectively treated with earlier identification and diagnosis. Its automated diagnosis from medical images is an area where deep learning (DL)-based algorithms have demonstrated promising results. In this research, we evaluated state-of-the-art DL architectures VGG16, ResNet50, InceptionV3, and our proposed innovative dual-path deep convolutional neural network (DPCNN) architecture for UF detection tasks. Using preprocessing methods including scaling, normalization, and data augmentation, an ultrasound image dataset from Kaggle is prepared for use. After the images are used to train and validate the DL models, the model performance is evaluated using different measures. When compared to existing DL models, our suggested DPCNN architecture achieved the highest accuracy of 99.8 percent. Findings show that pre-trained deep-learning model performance for UF diagnosis from medical images may significantly improve with the application of fine-tuning strategies. In particular, the InceptionV3 model achieved 90% accuracy, with the ResNet50 model achieving 89% accuracy. It should be noted that the VGG16 model was found to have a lower accuracy level of 85%. Our findings show that DL-based methods can be effectively utilized to facilitate automated UF detection from medical images. Further research in this area holds great potential and could lead to the creation of cutting-edge computer-aided diagnosis systems. To further advance the state-of-the-art in medical imaging analysis, the DL community is invited to investigate these lines of research. Although our proposed innovative DPCNN architecture performed best, fine-tuned versions of pre-trained models like InceptionV3 and ResNet50 also delivered strong results. This work lays the foundation for future studies and has the potential to enhance the precision and suitability with which UF is detected.

Artificial intelligence-aided method to detect uterine fibroids in ultrasound images: a retrospective study

We explored a new artificial intelligence-assisted method to assist junior ultrasonographers in improving the diagnostic performance of uterine fibroids and further compared it with senior ultrasonographers to confirm the effectiveness and feasibility of the artificial intelligence method. In this retrospective study, we collected a total of 3870 ultrasound images from 667 patients with a mean age of 42.45 years ± 6.23 [SD] for those who received a pathologically confirmed diagnosis of uterine fibroids and 570 women with a mean age of 39.24 years ± 5.32 [SD] without uterine lesions from Shunde Hospital of Southern Medical University between 2015 and 2020. The DCNN model was trained and developed on the training dataset (2706 images) and internal validation dataset (676 images). To evaluate the performance of the model on the external validation dataset (488 images), we assessed the diagnostic performance of the DCNN with ultrasonographers possessing different levels of seniority. The DCNN model aided the junior ultrasonographers (Averaged) in diagnosing uterine fibroids with higher accuracy (94.72% vs. 86.63%, P < 0.001), sensitivity (92.82% vs. 83.21%, P = 0.001), specificity (97.05% vs. 90.80%, P = 0.009), positive predictive value (97.45% vs. 91.68%, P = 0.007), and negative predictive value (91.73% vs. 81.61%, P = 0.001) than they achieved alone. Their ability was comparable to that of senior ultrasonographers (Averaged) in terms of accuracy (94.72% vs. 95.24%, P = 0.66), sensitivity (92.82% vs. 93.66%, P = 0.73), specificity (97.05% vs. 97.16%, P = 0.79), positive predictive value (97.45% vs. 97.57%, P = 0.77), and negative predictive value (91.73% vs. 92.63%, P = 0.75). The DCNN-assisted strategy can considerably improve the uterine fibroid diagnosis performance of junior ultrasonographers to make them more comparable to senior ultrasonographers.

MRI parameters for predicting the effect of ultrasound-guided high-intensity focused ultrasound in the ablation of uterine fibroids

AIM

To study the value of magnetic resonance imaging (MRI) parameters in predicting the efficacy of ultrasonic ablation of fibroids.

MATERIALS AND METHODS

A total of 91 patients were divided into groups based on non-perfused volume (NPV) ratio and blood supply type. The preoperative MRI parameters were measured and analysed. A correlation analysis between the MRI parameters and the NPV ratio was performed. Receiver operating characteristic (ROC) curves were used to analyse and determine the cut-off value of MRI parameters to predict the ablation rate of fibroids.

RESULTS

The uterine fibroids group with an NPV ratio <80% and the group with an NPV ratio of ≥80% had significant differences in signal intensity (SI) at MRI T2-weighted imaging (WI), fibroid-to-rectus abdominis SI ratio (SIR) at T2WI, and blood supply type (p<0.05). There were no significant differences in fibroid volume, T2WI signal uniformity, and apparent diffusion coefficient (ADC) values. The ADC value and SI and SIR at MRI T2WI in the group with poor blood supply were lower than those in the group with a rich blood supply (p<0.05). SI at MRI T2WI correlated negatively with the NPV ratio. The cut-off values for SI and SIR at MRI T2WI of fibroids whose NPV ratio exceeds 80% were 220.58 and 1.315, respectively.

CONCLUSION

SI at MRI T2WI and blood supply type could be predictors of the efficacy of ablation. Ultrasonic ablation of fibroids with MRI T2WI hyperintensity and a rich blood supply had poor efficacy.

Histological findings in resected leiomyomas following MR-HIFU treatment, single-institution data from seven patients with unfavorable focal therapy

PURPOSE

Magnetic resonance - high-intensity focused ultrasound (MR-HIFU) is a noninvasive treatment option for symptomatic uterine leiomyomas. Currently, pretreatment MRI is used to assess tissue characteristics and predict the most likely therapeutic response for individual patients. However, these predictions still entail significant uncertainties. The impact of tissue properties on therapeutic outcomes remains poorly understood and detailed knowledge of the histological effects of ultrasound ablation is lacking. Investigating these aspects could aid in optimizing patient selection, enhancing treatment effects and improving treatment outcomes.

METHODS AND MATERIALS

We present seven patients who underwent MR-HIFU treatment for leiomyoma followed by second-line surgical treatment. Tissue samples obtained during the surgery were stained with hematoxylin and eosin, Masson's trichrome and Herovici to evaluate general morphology, fibrosis and collagen deposition of leiomyomas. Immunohistochemical CD31, Ki-67 and MMP-2 stainings were performed to study vascularization, proliferation and matrix metalloproteinase-2 protein expression in leiomyomas, respectively.

RESULTS

The clinical characteristics and radiological findings of the leiomyomas prior to treatment as well as qualitative histological findings after the treatment are presented and discussed in the context of current literature. A tentative model for volume reduction is presented.

CONCLUSION

These findings provide insights into potential factors contributing to suboptimal therapeutic outcomes and the variability in histological changes following treatment.

Physician Experience in Technical Success of Achieving NPVR ≥ 80% of High-Intensity Focused Ultrasound Ablation for Uterine Fibroids: A Multicenter Study

Objective

To evaluate the experience of the physician of the technical success in high-intensity focused ultrasound (HIFU) ablation of uterine fibroids with a nonperfused volume ratio (NPVR) of at least 80%.

Methods

Patients from a 20-center prospective study were enrolled in this study. In this study, among the 20 clinical centers, five centers had physician with >3 years of HIFU experience, and the other 15 centers initiated HIFU therapy <3 years, were defined as the experienced group and the inexperienced group, respectively. Technical success was defined as achieving NPVR ≥ 80% of uterine fibroids with no major complications and it was defined as the successful group; otherwise, it was defined as the unsuccessful group.

Results

A total of 1,352 patients were included at the age of 41.32 ± 5.08 years. The mean NPVR (87.48 ± 14.91%) was obtained in the inexperienced group (86.50 ± 15.76%) and in the experienced group (89.21 ± 13.12%), respectively. The multivariate analysis showed that the volume of uterus, location of fibroids, and physician experience were significantly correlated with technical success (p < 0.05). In the experienced group, 82.20% of uterine fibroids obtained NPVR ≥ 80%, compared with 75.32% in the inexperienced group, and the difference was significant (p = 0.003). The technical success rate of the experienced group was 82.00% which was higher than 75.20% of the inexperienced group (p = 0.004).

Conclusion

In technical success of achieving NPVR ≥ 80%, experience of the physician was positively correlated with technical success; NPVR and major complications for the inexperienced group were comparable to those of the experienced group from a clinical perspective; inexperienced physicians could reach NPVR ≥ 80% of sufficient ablation and were trustworthy in efficacy. Smaller uterus and fibroids of anterior wall were correlated with better technical success; experienced physicians still have better technical success when choosing patients with larger uterus, contributing to clinical decision-making and patient referral.

Should Ultrasound-Guided High Frequency Focused Ultrasound Be Considered as an Alternative Non-Surgical Treatment of Uterine Fibroids in Non-Asiatic Countries? An Opinion Paper

Minimally invasive interventions for myomata treatment have gained acceptance due to the possibility of preserving fertility with reduced trauma induced by laparotomy as way of entrance. There are insufficient data regarding outcomes of high intensity focused ultrasound (HIFU) in non-Asiatic women. Therefore, we revised the available evidence to present an expert opinion that could support physicians, patients and policy-makers for considering this approach in other populations. We revisited systematic reviews, randomized controlled trials and cohort studies from January 2018 to August 2021 using PubMed and Google scholar, regarding short and long term outcomes after ablation with focused ultrasound waves. In total, 33 studies, including 114,810 adult patients showed that outcomes of this approach depend on several parameters directly related with resistance to thermal ablation, especially fibroid size and vascularization. Two studies report satisfactory outcomes in Afro-American women. In accordance to the technique used, fibroid volume reduction showed to be higher in fibroids <300 cm3 after ultrasound guided HIFU than after MRI guided. Compared to myomectomy and uterine artery embolization, HIFU seems to have shorter hospital stay, higher pregnancy rates and similar adverse events rates, with skin burn being the most reported. Symptoms and quality of life improvement is similar to myomectomy but lower than embolization, however reintervention rate is higher after HIFU. Lacks evidence about long-term sarcoma risk after ablation. Available evidence shows that HIFU can be considered as a uterine sparing treatment for women of different ethnicities suffering of uterine myomatosis, especially for those wishing to preserve their fertility.

A feasibility analysis of the ArcBlate MR-guided high-intensity focused ultrasound system for the ablation of uterine fibroids

Purpose

Uterine fibroids are benign gynecologic tumors and commonly occur in women by the age of 50. Women with symptomatic uterine fibroids generally receive surgical intervention, while they do not favor the invasive therapies. To evaluate the feasibility and safety of a novel magnetic resonance-guided high-intensity focused ultrasound (MRgHIFU) modality, ArcBlate, in the treatment of uterine fibroids.

Methods

Nine patients with uterine fibroids and one patient with adenomyosis were treated with ArcBlate MRgHIFU. Tumor size and quality of life were evaluated postoperatively at 1 and 3 months by magnetic resonance imaging (MRI) and the 36-Item Short Form Survey (SF-36), respectively.

Results

All patients completed the ArcBlate MRgHIFU procedure and there were no treatment-related adverse effects either during the procedure or during the 3 months of follow-up. Despite limiting the ablation volume to under 50% of the treated fibroid volume as a safety precaution, tumor volumes were markedly reduced in four patients by 15.78–58.87% at 3-month post-treatment. Moreover, SF-36 scale scores had improved at 3 months from baseline by 2–8 points in six patients, indicating relief of symptoms and improved quality of life.

Conclusion

This study evidence demonstrates the safety and feasibility of ArcBlate MRgHIFU and suggests its potential for treating uterine fibroids.

Feasibility of T2 relaxation time in predicting the technical outcome of MR-guided high-intensity focused ultrasound treatment of uterine fibroids

PURPOSE

The aim of this study was to assess the feasibility of T2 relaxation time in predicting the immediate technical outcome i.e., nonperfused volume ratio (NPVr) of magnetic resonance-guided high-intensity focused ultrasound (MRgHIFU) treatment of symptomatic uterine fibroids and to compare it with existing T2-weighted imaging methods (Funaki classification and scaled signal intensity, SSI).

MATERIALS AND METHODS

30 patients with 32 uterine fibroids underwent an MRI study including a quantitative T2 relaxation time measurement prior to MRgHIFU treatment. T2 relaxation times were measured with a multi-echo fast imaging-based technique with 16 echoes. The correlation between pretreatment values of the uterine fibroids and treatment outcomes, that is nonperfused volume ratios (NPVr), was assessed with nonparametric statistical measures. T2 relaxation time-based method was compared to existing T2-weighted imaging-based methods using receiver-operating-characteristics (ROC) curve analysis and Chi-square test.

RESULTS

Nonparametric measures of association revealed a statistically significant negative correlation between T2 relaxation time values and NPVr. The T2 relaxation time classification (T2 I, T2 II, and T2 III) resulted in the whole model p-value of 0.0019, whereas the Funaki classification resulted in a p-value of 0.56. The T2 relaxation time classification (T2 I and T2 II) achieved a whole model of a p-value of 0.0024, whereas the SSI classification had a p-value of 0.0749.

CONCLUSIONS

A longer T2 relaxation time of the fibroid prior to treatment correlated with a lower NPVr. Based on our results, the T2 relaxation time classifications seem to outperform the Funaki classification and the SSI method.

The efficacy and safety of MR-HIFU and US-HIFU in treating uterine fibroids with the volume <300 cm3: a meta-analysis

BACKGROUND

High-intensity focused ultrasound (HIFU) is a promising and non-invasive therapy for symptomatic uterine fibroids. Currently, the main image-guided methods for HIFU include magnetic resonance-guided (MR-HIFU) and ultrasound-guided (US-HIFU). However, there are few comparative studies on the therapeutic efficacy and safety of MR-HIFU and US-HIFU in treating symptomatic uterine fibroids with a volume <300 cm³.

OBJECTIVE

We performed this meta-analysis to evaluate the efficacy and safety of MR-HIFU and US-HIFU in treating symptomatic uterine fibroids with a volume <300 cm³.

METHODS

We searched relevant literature in PubMed, EMBASE, Cochrane Library CNKI from inception until 2021. The mean value, the proportion, and their 95% confidence intervals (CIs) were measured by random-effects models. Publication bias was assessed using funnel plots.

RESULTS

48 studies met our inclusion criteria-28 describing MR-HIFU and 20 describing US-HIFU. The mean non-perfused volume rate (NPVR) was 81.07% in the US-HIFU group and 58.92% in the MR-HIFU group, respectively. The mean volume reduction rates at month-3, month-6, and month-12 were 42.42, 58.72, and 65.55% in the US-HIFU group, while 34.79, 37.39, and 36.44% in the MR-HIFU group. The incidence of post-operative abdominal pain and abnormal vaginal discharge in the US-HIFU group was lower than that of MRI-HIFU. However, post-operative skin burn and sciatic nerve pain were more common in the US-HIFU group compared with MRI-HIFU. The one-year reintervention rate after MR-HIFU was 13.4%, which was higher than 5.2% in the US-HIFU group.

CONCLUSION

US-HIFU may show better efficiency and safety than MR-HIFU in treating symptomatic fibroids with a volume <300 cm³.

Magnetic resonance imaging parameters in predicting the ablative efficiency of high-intensity focused ultrasound for uterine fibroids

OBJECTIVE

To evaluate the role of quantitative MRI parameters in predicting HIFU ablation results for uterine fibroids.

MATERIAL AND METHODS

A total of 245 patients with uterine fibroids who underwent HIFU treatment in Chongqing Haifu Hospital were reviewed retrospectively. The patients were divided into two groups according to the non-perfused volume (NPV) ratio which was either higher or lower than 80%. The MRI parameters were measured, and a logistical regression analysis was performed to investigate the potential predictors associated with the NPV ratio. Receiver operating characteristics (ROC) analysis was used to determine the cut off value for MRI parameters in predicting a high NPV ratio.

RESULTS

The subcutaneous fat thickness in the group of patients with an NPV ratio over 80% was significantly thinner than that in the group of patients with an NPV ratio less than 80% (15 mm versus 21 mm). The signal intensity ratio of fibroids to skeletal muscle on T2WI was significantly lower in the group of patients with an NPV ratio over 80% compared with the group with an NPV ratio lower than 80% (2.46 versus 3.23). The signal intensity ratio of fibroid to skeletal muscle correlated negatively with the NPV ratio and positively with the energy efficiency factor (EEF). The cut off value of signal intensity ratio of fibroid to muscle for predicting the NPV ratio over 80% is 3.045.

CONCLUSION

The signal intensity ratio of fibroid to skeletal muscle on T2WI can be used as a factor for predicting the effectiveness of HIFU ablation of uterine fibroids.

Feasibility of apparent diffusion coefficient in predicting the technical outcome of MR-guided high-intensity focused ultrasound treatment of uterine fibroids - a comparison with the Funaki classification

PURPOSE

To investigate the feasibility of using an apparent diffusion coefficient (ADC) classification in predicting the technical outcome of magnetic resonance imaging-guided high-intensity focused ultrasound (MRgHIFU) treatment of symptomatic uterine fibroids and to compare it to the Funaki classification.

MATERIALS AND METHODS

Forty-two patients with forty-eight uterine fibroids underwent diffusion-weighted imaging (DWI) before MRgHIFU treatment. The DW images were acquired with five different b-values. Correlations between ADC values and treatment parameters were assessed. Optimal ADC cutoff values were determined to predict technical outcomes, that is, nonperfused volume ratios (NPVr) such that three classification groups were created (NPVr of <30%, 30-80%, or >80%). Results were compared to the Funaki classification using receiver-operating-characteristic (ROC) curve analysis, with statistical significance being tested with the Chi-square test.

RESULTS

A statistically significant negative correlation (Spearman's ρ = -0.31, p-value < 0.05) was detected between ADC values and NPV ratios. ROC curve analysis indicated that optimal ADC cutoff values of 980 × 10^-6mm²/s (NPVr > 80%) and 1800 × 10^-6mm²/s (NPVr < 30%) made it possible to classify fibroids into three groups: ADC I (NPVr > 80%), ADC II (NPVr 30-80%) and ADC III (NPVr < 30%). Analysis of the whole model area under the curve resulted in values of 0.79 for the ADC classification (p-value = 0.0007) and 0.62 for the Funaki classification (p-value = 0.0527).

CONCLUSIONS

Lower ADC values prior to treatment correlate with higher NPV ratios. The ADC classification seems to be able to predict the NPV ratio and may even outperform the Funaki classification. Based on these results DWI and ADC maps should be included in the MRI screening protocol.

Magnetic Resonance-Guided High-Intensity Focused Ultrasound Ablation of Uterine Fibroids-Efficiency Assessment with the Use of Dynamic Contrast-Enhanced Magnetic Resonance Imaging and the Potential Role of the Administration of Uterotonic Drugs

OBJECTIVE

The assessment of the usefulness of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) when qualifying patients with uterine fibroids (UFs) for magnetic resonance-guided high-intensity ultrasound (MR-HIFU).

MATERIAL AND METHODS

This retrospective, single center study included 283 women who underwent DCE-MRI and were treated with MR-HIFU. The patients were divided according to non-perfused volume (NPV) as well as by the type of curve for patients with a washout curve in the DCE-MRI study and patients without a washout curve. The studied women were assessed in three groups according to the type of uterotonics administered. Group A (57 patients) received one dose of misoprostol/diclofenac transvaginally and group B (71 patients) received oxytocin intravenously prior to the MR-HIFU procedure. The remaining 155 women (group C) were treated with the traditional non-drug enhanced MR-HIFU procedure.

RESULTS

The average NPV value was higher in no washout group, and depended on the uterotonics used.

CONCLUSIONS

We demonstrated a correlation between dynamic contrast enhancement curve types and the therapeutic efficacy of MR-HIFU. Our results suggest that DCE-MRI has the potential to assess treatment outcomes among patients with UFs, and patients with UFs that present with a washout curve may benefit from the use of uterotonic drugs. More studies are required to draw final conclusions.

Comparative Effectiveness and Safety of High-Intensity Focused Ultrasound for Uterine Fibroids: A Systematic Review and Meta-Analysis

Background: Uterine fibroids are common benign tumors among premenopausal women. High- intensity focused ultrasound (HIFU) is an emerging non-invasive intervention which uses the high-intensity ultrasound waves from ultrasound probes to focus on the targeted fibroids. However, the efficacy of HIFU in comparison with that of other common treatment types in clinical procedure remains unclear.

Objective: To investigate the comparative effectiveness and safety of HIFU with other techniques which have been widely used in clinical settings.

Methods: We searched the Cochrane Central Register of Controlled Trials, PubMed, EMBASE, Cumulative Index to Nursing & Allied Health Literature, Web of Science, ProQuest Nursing & Allied Health Database, and three Chinese academic databases, including randomized controlled trials (RCTs), non-RCTs, and cohort studies. The primary outcome was the rate of re-intervention, and the GRADE approach was used to interpret the findings.

Results: About 18 studies met the inclusion criteria. HIFU was associated with an increased risk of re-intervention rate in comparison with myomectomy (MYO) [pooled odds ratio (OR): 4.05, 95% confidence interval (CI): 1.82–8.9]. The results favored HIFU in comparison with hysterectomy (HYS) on the change of follicle-stimulating hormone [pooled mean difference (MD): −7.95, 95% CI: −8.92–6.98), luteinizing hormone (MD: −4.38, 95% CI: −5.17−3.59), and estradiol (pooled MD: 43.82, 95% CI: 36.92–50.72)]. HIFU had a shorter duration of hospital stay in comparison with MYO (pooled MD: −4.70, 95% CI: −7.46−1.94, p < 0.01). It had a lower incidence of fever (pooled OR: 0.15, 95% CI: 0.06–0.39, p < 0.01) and a lower incidence of major adverse events (pooled OR: 0.04, 95% CI: 0.00–0.30, p < 0.01) in comparison with HYS.

Conclusions: High-intensity focused ultrasound may help maintain feminity and shorten the duration of hospital stay. High-quality clinical studies with a large sample size, a long-term follow-up, and the newest HIFU treatment protocol for evaluating the re-intervention rate are suggested to be carried out. Clinical decision should be based on the specific situation of the patients and individual values.

Predictive Modeling for Frailty Conditions in Elderly People: Machine Learning Approaches

Background

Frailty is one of the most critical age-related conditions in older adults. It is often recognized as a syndrome of physiological decline in late life, characterized by a marked vulnerability to adverse health outcomes. A clear operational definition of frailty, however, has not been agreed so far. There is a wide range of studies on the detection of frailty and their association with mortality. Several of these studies have focused on the possible risk factors associated with frailty in the elderly population while predicting who will be at increased risk of frailty is still overlooked in clinical settings.

Objective

The objective of our study was to develop predictive models for frailty conditions in older people using different machine learning methods based on a database of clinical characteristics and socioeconomic factors.

Methods

An administrative health database containing 1,095,612 elderly people aged 65 or older with 58 input variables and 6 output variables was used. We first identify and define six problems/outputs as surrogates of frailty. We then resolve the imbalanced nature of the data through resampling process and a comparative study between the different machine learning (ML) algorithms – Artificial neural network (ANN), Genetic programming (GP), Support vector machines (SVM), Random Forest (RF), Logistic regression (LR) and Decision tree (DT) – was carried out. The performance of each model was evaluated using a separate unseen dataset.

Results

Predicting mortality outcome has shown higher performance with ANN (TPR 0.81, TNR 0.76, accuracy 0.78, F1-score 0.79) and SVM (TPR 0.77, TNR 0.80, accuracy 0.79, F1-score 0.78) than predicting the other outcomes. On average, over the six problems, the DT classifier has shown the lowest accuracy, while other models (GP, LR, RF, ANN, and SVM) performed better. All models have shown lower accuracy in predicting an event of an emergency admission with red code than predicting fracture and disability. In predicting urgent hospitalization, only SVM achieved better performance (TPR 0.75, TNR 0.77, accuracy 0.73, F1-score 0.76) with the 10-fold cross validation compared with other models in all evaluation metrics.

Conclusions

We developed machine learning models for predicting frailty conditions (mortality, urgent hospitalization, disability, fracture, and emergency admission). The results show that the prediction performance of machine learning models significantly varies from problem to problem in terms of different evaluation metrics. Through further improvement, the model that performs better can be used as a base for developing decision-support tools to improve early identification and prediction of frail older adults.