Research on the diagnostic value of ultrasound combined with MRI features based on deep learning in distinguishing benign and malignant ovarian tumors

Background: This study aimed to assess the diagnostic value of a deep learning (DL)-based multimodal approach that combines ultrasound (US) and magnetic resonance imaging (MRI) features in differentiating benign and malignant ovarian tumors, and to develop an intelligent auxiliary diagnostic tool. Methods: A total of 887 patients (665 benign, 222 malignant) with pathologically confirmed ovarian tumors from 2022 to 2024 were retrospectively enrolled. All patients underwent preoperative US and MRI within one week. A dual-channel DL model was constructed: the US branch (ResNet50) extracted 2D features from grayscale color Doppler flow imaging, while the MRI branch (3D ResNeXt101) extracted 3D features from T2-weighted imaging (T2WI), dynamic contrast-enhanced (DCE)-MRI, and apparent diffusion coefficient (ADC) maps. The extracted features were integrated using an attention-based fusion mechanism. With pathology as the gold standard, the diagnostic performance of the proposed model was compared with US alone, MRI alone, and the Assessment of Different NEoplasias in the adneXa (ADNEX) model. Results: In the test cohort (n = 266), the DL model showed a sensitivity of 92.73%, specificity of 98.58%, accuracy of 97.37%, and an area under the curve (AUC) of 0.957, which were significantly higher than those of US (AUC = 0.792, z = 3.92, p < 0.001), MRI (AUC = 0.844, z = 2.76, p = 0.006), and ADNEX (AUC = 0.885, z = 2.07, p = 0.022). Conclusions: The DL-based US-MRI multimodal fusion model significantly enhances the diagnostic accuracy for differentiating benign and malignant ovarian tumors, providing a promising intelligent auxiliary tool for early and precise diagnosis.

Ovarian tumor; Deep learning; Ultrasonography; Magnetic resonance imaging; Benign and malignant differentiation; Diagnostic value; Multimodal fusion

[1] Tjokroprawiro BA, Novitasari K, Ulhaq RA, Sulistya HA. Clinicopathological analysis of giant ovarian tumors. European Journal of Obstetrics & Gynecology and Reproductive Biology: X. 2024; 22: 100318.

[2] Wang Y, Wang Z, Zhang Z, Wang H, Peng J, Hong L. Burden of ovarian cancer in China from 1990 to 2030: a systematic analysis and comparison with the global level. Frontiers in Public Health. 2023; 11: 1136596.

[3] Timmerman D, Planchamp F, Bourne T, Landolfo C, du Bois A, Chiva L, et al. ESGO/ISUOG/IOTA/ESGE Consensus Statement on pre-operative diagnosis of ovarian tumors. International Journal of Gynecological Cancer. 2021; 31: 961–982.

[4] Zhou Y, Duan Y, Zhu Q, Li S, Liu X, Cheng T, et al. Integrative deep learning and radiomics analysis for ovarian tumor classification and diagnosis: a multicenter large-sample comparative study. Radiologia Medica. 2025; 130: 889–904.

[5] Zhao B, Wen L, Huang Y, Fu Y, Zhou S, Liu J, et al. A deep learning-based automatic recognition model for polycystic ovary ultrasound images. Balkan Medical Journal. 2025; 42: 419–428.

[6] Zeng S, Jia H, Zhang H, Feng X, Dong M, Lin L, et al. Multimodal ultrasound-based radiomics and deep learning for differential diagnosis of O-RADS 4–5 adnexal masses. Cancer Imaging. 2025; 25: 64.

[7] Maniaci A, La Via L, Lavalle S, Lentini M, Pavone P, Iannella G, et al. Presentation, radiologic features, and treatment options of congenital tongue tumors: a comprehensive review. Annali Italiani di Chirurgia. 2024; 95: 481–496.

[8] Yang R, Zou Y, Li L, Liu WV, Liu C, Wen Z, et al. Enhancing repeatability of follicle counting with deep learning reconstruction high-resolution MRI in PCOS patients. Scientific Reports. 2025; 15: 1241.

[9] Wang X, Quan T, Chu X, Gao M, Zhang Y, Chen Y, et al. Deep learning radiomics nomogram based on MRI for differentiating between borderline ovarian tumors and stage I ovarian cancer: a multicenter study. Academic Radiology. 2025; 32: 3485–3497.

[10] Hsu WC, Wang Y, Wu YF, Chen R, Afyouni S, Liu J, et al. MRI-based ovarian lesion classification via a foundation segmentation model and multimodal analysis: a multicenter study. Radiology. 2025; 316: e243412.

[11] Altinsoy E, Bakirarar B, Culcu S. Machine learning in predicting gastric cancer survival Presenting a novel decision support system model. Annali Italiani di Chirurgia. 2023; 94: 631–638.

[12] Yin R, Guo Y, Wang Y, Zhang Q, Dou Z, Wang Y, et al. Predicting neoadjuvant chemotherapy response and high-grade serous ovarian cancer from CT images in ovarian cancer with multitask deep learning: a multicenter study. Academic Radiology. 2023; 30: S192–S201.

[13] Yin R, Dou Z, Wang Y, Zhang Q, Guo Y, Wang Y, et al. Preoperative CECT-based multitask model predicts peritoneal recurrence and disease-free survival in advanced ovarian cancer: a multicenter study. Academic Radiology. 2024; 31: 4488–4498.

[14] Akazawa M, Hashimoto K. Artificial intelligence in gynecologic cancers: current status and future challenges—a systematic review. Artificial Intelligence in Medicine. 2021; 120: 102164.

[15] Xu HL, Gong TT, Liu FH, Chen HY, Xiao Q, Hou Y, et al. Artificial intelligence performance in image-based ovarian cancer identification: a systematic review and meta-analysis. eClinicalMedicine. 2022; 53: 101662.

[16] Du Y, Wang T, Qu L, Li H, Guo Q, Wang H, et al. Preoperative molecular subtype classification prediction of ovarian cancer based on multi-parametric magnetic resonance imaging multi-sequence feature fusion network. Bioengineering. 2024; 11: 472.

[17] Elguoshy A, Zedan H, Saito S. Machine learning-driven insights in cancer metabolomics: from subtyping to biomarker discovery and prognostic modeling. Metabolites. 2025; 15: 514.

[18] Geysels A, Garofalo G, Timmerman S, Barrenada L, De Moor B, Timmerman D, et al. Artificial intelligence applied to ultrasound diagnosis of pelvic gynecological tumors: a systematic review and meta-analysis. Gynecologic and Obstetric Investigation. 2025. PMID: 40340944; PMCID: PMC12180770.

[19] Du Y, Xiao Y, Guo W, Yao J, Lan T, Li S, et al. Development and validation of an ultrasound-based deep learning radiomics nomogram for predicting the malignant risk of ovarian tumours. BioMedical Engineering OnLine. 2024; 23: 41.

[20] Bogaerts JM, Steenbeek MP, Bokhorst JM, van Bommel MH, Abete L, Addante F, et al. Assessing the impact of deep-learning assistance on the histopathological diagnosis of serous tubal intraepithelial carcinoma (STIC) in fallopian tubes. The Journal of Pathology: Clinical Research. 2024; 10: e70006.

[21] Dai WL, Wu YN, Ling YT, Zhao J, Zhang S, Gu ZW, et al. Development and validation of a deep learning pipeline to diagnose ovarian masses using ultrasound screening: a retrospective multicenter study. eClinicalMedicine. 2024; 78: 102923.

[22] Du Y, Guo W, Xiao Y, Chen H, Yao J, Wu J. Ultrasound-based deep learning radiomics model for differentiating benign, borderline, and malignant ovarian tumours: a multi-class classification exploratory study. BMC Medical Imaging. 2024; 24: 89.

[23] El-Latif EIA, El-Dosuky M, Darwish A, Hassanien AE. A deep learning approach for ovarian cancer detection and classification based on fuzzy deep learning. Scientific Reports. 2024; 14: 26463.

[24] He D, Jin L, Geng H, Cao L. Deep learning-based analysis of gross features for ovarian epithelial tumors classification: a tool to assist pathologists for frozen section sampling. Human Pathology. 2025; 157: 105762.

[25] Hou B, Lee S, Lee JM, Koh C, Xiao J, Pickhardt PJ, et al. Deep learning segmentation of ascites on abdominal CT scans for automatic volume quantification. Radiology: Artificial Intelligence. 2024; 6: e230601.

[26] Bhuvaneshwari KV, Lahza H, Sreenivasa BR, Lahza HFM, Shawly T, Poornima B. Optimising ovarian tumor classification using a novel CT sequence selection algorithm. Scientific Reports. 2024; 14: 25010.

[27] Giourga M, Petropoulos I, Stavros S, Potiris A, Gerede A, Sapantzoglou I, et al. Enhancing ovarian tumor diagnosis: performance of convolutional neural networks in classifying ovarian masses using ultrasound images. Journal of Clinical Medicine. 2024; 13: 4123.

[28] Ma L, Gao W, Hu X, Zhou D, Wang C, Yu J, et al. An improved cancer diagnosis algorithm for protein mass spectrometry based on PCA and a one-dimensional neural network combining ResNet and SENet. Analyst. 2024; 149: 5675–5683.

[29] He X, Bai XH, Chen H, Feng WW. Machine learning models in evaluating the malignancy risk of ovarian tumors: a comparative study. Journal of Ovarian Research. 2024; 17: 219.

[30] Jung Y, Kim T, Han MR, Kim S, Kim G, Lee S, et al. Ovarian tumor diagnosis using deep convolutional neural networks and a denoising convolutional autoencoder. Scientific Reports. 2022; 12: 17024.

[31] Saida T, Mori K, Hoshiai S, Sakai M, Urushibara A, Ishiguro T, et al. Diagnosing ovarian cancer on MRI: a preliminary study comparing deep learning and radiologist assessments. Cancers. 2022; 14: 987.

[32] Yang Q, Zhang H, Ma PQ, Peng B, Yin GT, Zhang NN, et al. Value of ultrasound and magnetic resonance imaging combined with tumor markers in the diagnosis of ovarian tumors. World Journal of Clinical Cases. 2023; 11: 7553–7561.

[33] Mitchell S, Gleeson J, Tiwari M, Bailey F, Gaughran J, Mehra G, et al. Accuracy of ultrasound, magnetic resonance imaging and intraoperative frozen section in the diagnosis of ovarian tumours: data from a London tertiary centre. BJC Reports. 2024; 2: 50.

[34] Wang WH, Zheng CB, Gao JN, Ren SS, Nie GY, Li ZQ. Systematic review and meta-analysis of imaging differential diagnosis of benign and malignant ovarian tumors. Gland Surgery. 2022; 11: 330–340.

[35] Cui L, Xu H, Zhang Y. Diagnostic accuracies of the ultrasound and magnetic resonance imaging ADNEX scoring systems for ovarian adnexal mass: systematic review and meta-analysis. Academic Radiology. 2022; 29: 897–908.

[36] Hu Y, Chen B, Dong H, Sheng B, Xiao Z, Li J, et al. Comparison of ultrasound-based ADNEX model with magnetic resonance imaging for discriminating adnexal masses: a multi-center study. Frontiers in Oncology. 2023; 13: 1101297.

[37] Adusumilli P, Ravikumar N, Hall G, Scarsbrook AF. A methodological framework for AI-assisted diagnosis of ovarian masses using CT and MR imaging. Journal of Personalized Medicine. 2025; 15: 76.