|Table of Contents|

The value of machine learning models based on MRI images in predicting the efficacy of uterine fibroids after HIFU ablation surgery

Journal Of Modern Oncology[ISSN:1672-4992/CN:61-1415/R]

Issue:
2024 06
Page:
1093-1100
Research Field:
Publishing date:

Info

Title:
The value of machine learning models based on MRI images in predicting the efficacy of uterine fibroids after HIFU ablation surgery
Author(s):
HE Yueming1; LUO Jinwen2; QI Yingying1; ZHANG Ting1
1.Department of Gynecology;2.Department of Medical Imaging,the Fifth Affiliated Hospital of Guangzhou Medical University,Guangdong Guangzhou 510700,China.
Keywords:
magnetic resonance; machine learning; uterine fibroids; HIFU
PACS:
R737.33
DOI:
10.3969/j.issn.1672-4992.2024.06.019
Abstract:
Objective:To investigate the value of MRI image-based machine learning models in predicting the efficacy of HIFU ablation for uterine fibroids.Methods:We retrospectively reviewed the MRI images and clinical data of 108 patients with uterine fibroids before and after HIFU ablation in our institution,grouped them by 65% postoperative volume ablation rate (NPVR),performed whole tumor MRI texture machine learning of fibroids,used ITK-SNAP software to delineate the volumetric region of interest (VOI) on preoperative MRI for each fibroid,extracted 1 834 omics features for each VOI using Python,stitched the features with annotated data,to form the dataset and regularize the data,we randomly constructed 76 cases (70%) of the training set and 32 cases (30%) of the test set to screen the features by correlation coefficient Spearman,and LR,SVM,KNN,RandomForest,ExtraTrees,XGBoost,LightGBM,MLP machine to learn the prediction model of ablation efficacy by filtering the nonzero features by Lasso.ROC and DCA curve analysis,confusion matrix plots were drawn and AUC,sensitivity,specificity,positive predictive value (PPV),negative predictive value (NPV),precision rate,recall rate,F1 score of each model were calculated to evaluate the diagnostic performance of the prediction model and its clinical application.Results:After dividing the ablation significant group by ablation effect in 69 patients,ablation non significant group in 39 patients,randomly constructing the training set in 76 patients (70%) and the test set in 32 patients (30%).ROC curve analysis showed that the MRI omics prediction model predicted ablation significant group the top four models in the validation set were LR 0.875（95%CI 0.731~1.000），MLP 0.875（95%CI 0.714~1.000），LightGBM 0.853（95%CI 0.689~1.000），RandomForest 0.848（95%CI 0.676~1.000）,and the difference was statistically significant (P＜0.05).Conclusion:The machine learning model based on texture construction of MRI images predicts the efficacy of HIFU ablation under ultrasound guidance for uterine fibroids with good performance,which may provide a personalized and quantitative reference basis for preoperative selection of HIFU for uterine fibroids and evaluation of postoperative efficacy.

References:

［1］ WANG Y，WANG Z，XU Y．Efficacy，efficiency，and safety of magnetic ｒesonance-guided high-intensity focused ultrasound for ablation of uterine fibroids:comparison with ultrasound-guided method［J］．Korean Journal of Ｒadiology，2018，19:724-732．
［2］ GIULIANI E，AS-SANIE S，MARSH EE．Epidemiology and management ofuterine fibroids［J］．Int J Gynaecol Obstet，2020，149:3-9．
［3］ CHEUNG VY.High-intensity focused ultrasound therapy［J］.Best Pract Res Clin Obstet Gynaecol,2018,46:74-83.
［4］ CHEN J,LI Y,WANG Z,et al.Evaluation of high-intensity focused ultrasound ablation for uterine fibroids:an IDEAL prospective exploration study［J］.BJOG:Int J Obstet Gy,2018,125(3):354-364.
［5］ JI Y,HU K,ZHANG Y,et al.High-intensity focused ultrasound (HIFU) treatment for uterine fibroids:a meta-analysis［J］.Arch Gynecol Obstet,2017,296(6):1181-1188.
［6］ MINDJUK I,TRUMM CG,HERZOG P,et al.MRI predictors of clinical success in MR-guided focused ultrasound (MRgFUS) treatments of uterine fibroids:results from a single centre［J］.Eur Radiol,2015,25(5):1317-1328.
［7］ KIM YS,LIM HK,PARK MJ,et al.Screening magnetic resonance imaging-based prediction model for assessing immediate therapeutic response to magnetic resonance imaging-guided highintensity focused ultrasound ablation of uterine fibroids［J］.Invest Radiol,2016,51(1):15-24.
［8］范宏杰,寸江平,黄建强,等.HIFU消融治疗不同MRI特征子宫肌瘤的效果［J］.中国介入影像与治疗学,2018,15(7):397-400. FAN HJ,CUN JP,HUANG JQ,et al.Effect of HIFU ablation on uterine fibroid with different MRI features ［J］.Chinese Journal of Interventional Imaging and Therapy,2018,15(7):397-400.
［9］罗锦文,李新春,刘美玲,等.基于CT增强图像的影像组学模型对鉴别结直肠癌旁肿瘤沉积及转移淋巴结的诊断价值［J］.放射学实践,2020,35(12):1553-1559. LUO JW,LI XC,LIU ML,et al.The diagnostic value of an imaging omics model based on CT enhanced images in distinguishing tumor deposition and metastatic lymph nodes adjacent to colorectal cancer ［J］.Radiology Practice,2020,35(12):1553-1559.
［10］罗锦文.基于螺旋CT图像在鉴别结直肠癌旁肿瘤沉积与转移淋巴结的价值初步分析［D］.广州：广州医科大学,2019. LUO JW.Preliminary analysis of the value of spiral CT images in differentiating tumor deposits and metastatic lymph nodes adjacent to colorectal cancer ［D］.Guangzhou：Guangzhou Medical University,2019.
［11］ 林子淮,李欣明,全显跃.基于机器学习CT影像组学特征联合血清学特征模型预测高出血风险食管静脉曲张［J］.中国CT和MRI杂志,2023,21(06):93-96. LIN ZH,LI XM,QUAN XY.Prediction of esophageal varices with high bleeding risk based on machine learning CT radiomics features combined with serological feature model［J］.Chinese Journal of CT and MRI,2023,21(06):93-96.
［12］赵鹏飞.基于磁共振栖息地影像和影像组学的肿瘤内异质性量化指标预测局部晚期宫颈鳞癌同步放化疗的早期疗效及长期预后的实验和临床研究［D］.沈阳：中国医科大学,2022. ZHAO PF.Experimental and clinical studies on the prediction of early efficacy and long-term prognosis of concurrent chemoradiotherapy in locally advanced cervical squamous cell carcinoma based on quantitative indicators of intratumor heterogeneity based on magnetic resonance habitat imaging and radiomics［D］.Shenyang:China Medical University,2022.
［13］陈小霞,蓝颖,吴能娴,等.基于~(18)F-FDG PET/CT影像组学模型预测局部晚期宫颈癌的放疗敏感性［J］.中国癌症防治杂志,2023,15(04):417-423. CHEN XX,LAN Y,WU NX,et al.Prediction of radiotherapy sensitivity of locally advanced cervical cancer based on ~(18)F-FDG PET/CT radiomics model［J］.Chinese Journal of Cancer Prevention and Treatment,2023,15(04):417-423.
［14］陈思瑶,李成海,周野,等.血流灌注对高强度聚焦超声消融子宫肌瘤的影响［J］.中国医学物理学杂志,2021,38(11):1412-1416. CHEN SY,LI CH,ZHOU Y,et al.The effect of blood perfusion on high intensity focused ultrasound ablation of uterine fibroids ［J］.Chinese Journal of Medical Physics,2021,38(11):1412-1416.
［15］周野,李成海,陈思瑶,等.基于MRI影像预测高强度聚焦超声治疗子宫肌瘤疗效的研究进展［J］.医学研究杂志,2021,50(08):16-19. ZHOU Y,LI CH,CHEN SY,et al.Research progress on predicting the efficacy of high-intensity focused ultrasound in the treatment of uterine fibroids based on MRI images［J］.Journal of Medical Research,2021,50(08):16-19.
［16］顾颖歆.基于随机森林的变压器油温预测方法［J］.运筹与模糊学,2021,11(4):462-469. GU YX.A method for predicting transformer oil temperature based on random forest ［J］.Operations Research and Fuzziness,2021,11(4):462-469.
［17］刘畅.基于深度学习的网上虚假交易识别研究［D］.武汉:武汉大学,2017. LIU C.Research on online false transaction identification based on deep learning ［D］.Wuhan：Wuhan University,2017.
［18］石胜源,朱磊,叶琳,等.基于随机森林算法的心血管疾病预测研究［J］.智能计算机与应用,2021,11(04):176-178，181. SHI SY,ZHU L,YE L,et al.Research on cardiovascular disease prediction based on Random forest algorithm ［J］.Intelligent Computer and Application,2021,11(04):176-178，181.
［19］连建文,马剑坤,王仕莉，等.顺北断控碳酸盐岩油藏油柱高度的计算方法研究［J］.重庆科技学院学报(自然科学版),2020,22(03):36-40. LIAN JW,MA JK,WANG SL，et al.Study on the calculation method of oil column height in Shunbei fault controlled carbonate reservoirs ［J］.Journal of Chongqing University of Science and Technology (Natural Science Edition),2020,22(03):36-40.
［20］吴琼,李荣琳,洪海生，等.基于混合重抽样和LightGBM算法的配变低压跳闸预测［J］.电力系统保护与控制,2021,49(12):71-78. WU Q,LI RL,HONG HS,et al.Low voltage trip prediction of distribution transformers based on mixed resampling and LightGBM algorithm ［J］.Power System Protection and Control,2021,49(12):71-78.
［21］尹瑞文.数据驱动的监狱自杀风险模型及监管策略研究［D］.南京：南京航空航天大学,2021. YIN RW.Data driven prison suicide risk model and regulatory strategy research ［D］.Nanjing：Nanjing University of Aeronautics and Astronautics,2021.
［22］张月飞,王伟,代伟.重介分选过程产品指标在线预测方法研究［J］.煤炭工程,2021,53(S1):108-111. ZHANG YF,WANG W,DAI W.Research on online prediction method for product indicators in heavy media sorting process ［J］.Coal Engineering,2021,53(S1):108-111.
［23］万松峰.基于MPL分类器的零件识别［J］.工业控制计算机,2022,35(11):126-127，130. WAN SF.Part identification based on MPL classifier［J］.Industrial Control Computer,2022,35（11）:126-127，130.

Memo

Memo:
广东省广州市基础研究计划基础与应用基础研究项目（编号:202102080250）
Last Update: 1900-01-01