Deep learning-assisted diagnosis of placenta accreta spectrum using the DenseNet-121 model: a multicenter, retrospective study
Open accessAbstract
To explore the diagnostic value of deep learning (DL) imaging based on MRI in predicting placenta accreta spectrum (PAS) in high-risk pregnant women. A total of 263 patients with suspected placenta accreta from Institution I and Institution II were retrospectively analyzed and divided into training (n = 170) and external verification sets (n = 93). Through imaging acquisition, feature extraction, and radiomic data processing, 15 radiomic features were used to train support vector machine (SVM), K-nearest neighbor (KNN), random forest (RF), light gradient boosting machine (LGBM), and DL models to predict PAS. The diagnostic performances of the models were evaluated in the training set using the area under the curve (AUC) and accuracy and further validated in the external verification set. Univariate and multivariate logistic regression analysis revealed that a history of cesarean section, placental thickness, and placenta previa were independent clinical risk factors for predicting PAS. Among machine learning (ML) models, SVM demonstrated the highest diagnostic power (AUC = 0.944), with an accuracy of 0.876. The diagnostic efficiency of the DL model was significantly better than that of other models, with an AUC of 0.956 (95% CI 0.931-0.981) in the training set and 0.863 (95% CI 0.816-0.910) in the external verification set. In terms of specificity, the DL model outperformed the ML models. The DL model based on MRI may demonstrate better performance in the diagnosis of PAS compared to traditional clinical models or ML radiomics models, as further confirmed by the external verification set.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Hu et al. evaluated the DenseNet-121 model, which enhanced diagnostic accuracy and improved interobserver reliability among radiologists for PAS diagnosis.
“Comparatively, Hu et al. evaluated the DenseNet-121 model, a deep CNN across multiple centers. The model not only enhanced diagnostic accuracy but also improved interobserver reliability among radiologists.”
From Machine Learning in Maternity: Early Diagnosis of Placenta Accreta Spectrum (2026) · cites this paper as [37]
The Discussion (page unavailable) reports that the DenseNet-121 model had higher diagnostic accuracy for PAS than the clinical and machine-learning models, using data from two centers. The Segmentation and Feature Extraction section says one radiologist’s segmentation was reviewed by another, but the paper does not report improved interobserver reliability among radiologists. (LLM, conf=0.87)
Experimental — accuracy not yet measuredSaliency mapping techniques incorporated by Hu et al. visually highlighted the most predictive regions on MRI scans, which is crucial for clinical trust and adoption.
“Notably, Hu et al. incorporated saliency mapping techniques, which visually highlighted the most predictive regions on MRI scans. This interpretability aspect is crucial for clinical trust and adoption, as it allows physicians to understand and validate the decision-making process of the ML model [37].”
From Machine Learning in Maternity: Early Diagnosis of Placenta Accreta Spectrum (2026) · cites this paper as [37]
The Discussion (page unavailable) describes a visualization showing the placental implant location and weighting lesion areas, and says visual feature analysis demonstrated the model’s feasibility for clinical applications. It does not establish that the visualization allowed physicians to validate the model’s decisions or that it was crucial for clinical trust and adoption. (LLM, conf=0.83)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Explainable artificial intelligence (XAI) in deep learning-based medical image analysis
van der Velden, B. H. M., Kuijf, H. J., Gilhuijs, K. G. A., & Viergever, M. A. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102470 · Cited by 1,328 · cited 2× · see citing works
“Some studies have shown that DL models, with their trainable attention mechanisms, can better capture the relationship between biological information and avoid the generalization issues typically seen with machine learning (ML) models [13].”
Best Practice & Research Clinical Obstetrics & Gynaecology
Fisk, N., & Itskovitz‐Eldor, J. (2004) · Cited by 741 · see citing works
Epidemiology of placenta previa accreta: a systematic review and meta-analysis
Jauniaux, E. R. M., Grønbeck, L., Bunce, C., Langhoff‐Roos, J., & Collins, S. (2019) · BMJ Open · doi:10.1136/bmjopen-2019-031193 · Cited by 206 · see citing works
Jha, P., Pōder, L., Bourgioti, C., Bharwani, N., Lewis, S., Kamath, A., Nougaret, S., Soyer, P., Weston, M., Castillo, R. P., et al. (2020) · European Radiology · doi:10.1007/s00330-019-06617-7 · Cited by 173 · see citing works
Radiomics and Deep Learning: Hepatic Applications
Park, H. J., Park, B., & Lee, S. S. (2020) · Korean Journal of Radiology · doi:10.3348/kjr.2019.0752 · Cited by 160 · see citing works
Romeo, V., Ricciardi, C., Cuocolo, R., Stanzione, A., Verde, F., Sarno, L., Improta, G., Mainenti, P. P., D'Armiento, M., Brunetti, A., et al. (2019) · Magnetic Resonance Imaging · doi:10.1016/j.mri.2019.05.017 · Cited by 113 · see citing works
Zheng, Y., Zhou, D., Liu, H., & Wen, M. (2022) · European Radiology · doi:10.1007/s00330-022-08830-3 · Cited by 104 · see citing works
Radiomics in precision medicine for gastric cancer: opportunities and challenges
Chen, Q., Zhang, L., Liu, S., You, J., Chen, L., Jin, Z., Zhang, S., & Zhang, B. (2022) · European Radiology · doi:10.1007/s00330-022-08704-8 · Cited by 101 · see citing works
Kapoor, H., Hanaoka, M. M., Dawkins, A. A., & Khurana, A. (2020) · Placenta · doi:10.1016/j.placenta.2020.11.004 · Cited by 87 · see citing works
Do, Q. N., Lewis, M. A., Xi, Y., Madhuranthakam, A. J., Happe, S. K., Dashe, J. S., Lenkinski, R. E., Khan, A., & Twickler, D. M. (2020) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.26883 · Cited by 61 · see citing works
Placenta accreta spectrum-a catastrophic situation in obstetrics
Pegu, B., Chitra, T., Nayak, D., & Subbaiah, M. (2021) · Obstetrics & Gynecology Science · doi:10.5468/ogs.20345 · Cited by 59 · see citing works
Chen, Y., Wang, L., Dong, X., et al. (2023) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-023-00818-9 · Cited by 55 · see citing works
Leveraging deep learning to improve vaccine design
Hederman, A. P., & Ackerman, M. E. (2023) · Trends in Immunology · doi:10.1016/j.it.2023.03.002 · Cited by 40 · see citing works
Luetkens, J. A., Nowak, S., Mesropyan, N., Block, W., Praktiknjo, M., Chang, J., Bauckhage, C., Sifa, R., Sprinkart, A. M., Faron, A., et al. (2022) · Scientific Reports · doi:10.1038/s41598-022-12410-2 · Cited by 34 · see citing works
Peng, L., Zhang, X., Liu, J., Liu, Y., Huang, J., Chen, J., Su, Y., Yang, Z., & Song, T. (2022) · European Radiology · doi:10.1007/s00330-022-08821-4 · Cited by 34 · see citing works
Radiomics and Deep Learning to Predict Pulmonary Nodule Metastasis at CT
Sohn, J. H., & Fields, B. K. K. (2024) · Radiology · doi:10.1148/radiol.233356 · Cited by 34 · see citing works
Ren, H., Mori, N., Mugikura, S., Shimizu, H., Kageyama, S., Saito, M., & Takase, K. (2021) · Abdominal Radiology · doi:10.1007/s00261-021-03226-1 · Cited by 30 · see citing works
Romeo, V., Sarno, L., Volpe, A., Ginocchio, M. I., Esposito, R., Mainenti, P. P., Petretta, M., Liuzzi, R., D'Armiento, M. R., Martinelli, P., et al. (2019) · Abdominal Radiology · doi:10.1007/s00261-019-02185-y · Cited by 27 · see citing works
Deep learning-based segmentation of the placenta and uterus on MR images
Shahedi, M., Spong, C. Y., Dormer, J. D., Do, Q. N., Xi, Y., Lewis, M. A., Herrera, C. L., Madhuranthakam, A. J., Twickler, D. M., & Fei, B. (2021) · Journal of medical imaging · doi:10.1117/1.jmi.8.5.054001 · Cited by 26 · see citing works
Ye, Z., Xuan, R., Ouyang, M., Wang, Y., Xu, J., & Jin, W. (2022) · Abdominal Radiology · doi:10.1007/s00261-022-03673-4 · Cited by 24 · see citing works
Prenatal prediction and typing of placental invasion using MRI deep and radiomic features
Xuan, R., Li, T., Wang, Y., Xu, J., & Jin, W. (2021) · BioMedical Engineering OnLine · doi:10.1186/s12938-021-00893-5 · Cited by 21 · see citing works
A multimodal deep learning model for predicting severe hemorrhage in placenta previa
Akazawa, M., & Hashimoto, K. (2023) · Scientific Reports · doi:10.1038/s41598-023-44634-1 · Cited by 17 · see citing works
MI-DenseCFNet: deep learning–based multimodal diagnosis models for Aureus and Aspergillus pneumonia
Liu, T., Zhang, Z., Zhou, Q.-H., et al. (2024) · European Radiology · doi:10.1007/s00330-023-10578-3 · Cited by 17 · see citing works
Self-citationBrain MRI Deep Learning and Bayesian Inference System Augments Radiology Resident Performance
Rudie, J. D., Duda, J., Duong, M. T., Chen, P., Xie, L., Kurtz, R. M., Ware, J. B., Choi, J., Mattay, R. R., Botzolakis, E. J., et al. (2021) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-021-00470-1 · Cited by 16 · see citing works
Romeo, V., & Maurea, S. (2019) · EBioMedicine · doi:10.1016/j.ebiom.2019.11.049 · Cited by 15 · see citing works
Routine screening for placenta accreta spectrum
Bhide, A. (2023) · Best Practice & Research Clinical Obstetrics & Gynaecology · doi:10.1016/j.bpobgyn.2023.102392 · Cited by 11 · see citing works
Zong, M., Pei, X., Yan, K., Luo, D., Zhao, Y., Wang, P., & Chen, L. (2023) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.29023 · Cited by 11 · see citing works
Correlation of placental thickness and PAS disorders: findings from MRI
Lu, T., Wang, Y., Guo, A., Deng, Y., Wu, C., Li, X., & Liu, S. (2022) · Abdominal Radiology · doi:10.1007/s00261-022-03420-9 · Cited by 10 · see citing works
Zhang, Y., Liu, P., Tang, L., Lin, P.-M., Li, R., Luo, H., & Luo, P. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107130 · Cited by 9 · see citing works
Maurea, S., Verde, F., Mainenti, P. P., Barbuto, L., Iacobellis, F., Romeo, V., Liuzzi, R., Raia, G., Dominicis, G. D., Santangelo, C., et al. (2021) · European Journal of Radiology · doi:10.1016/j.ejrad.2021.110078 · Cited by 8 · see citing works
External Validation of a Multivariable Prediction Model for Placenta Accreta Spectrum
Singh, S., Carusi, D. A., Wang, P., Reitman-Ivashkov, E., Landau, R., Fields, K. G., Weiniger, C. F., & Farber, M. K. (2022) · Anesthesia & Analgesia · doi:10.1213/ane.0000000000006222 · Cited by 6 · see citing works
Parameter tuning in machine learning based on radiomics biomarkers of lung cancer
Luo, Y., Li, Y., Zhang, Y., et al. (2022) · Journal of X-Ray Science and Technology · doi:10.3233/xst-211096 · Cited by 1 · see citing works
Flagged first, then by citations
Research works printed but not linked: 2
- [17]17. Model for Placenta Accreta Spectrum. Anesthesia and analgesia 2023;137:537-47.
- [22]22. Huang S, Cai N, Pacheco PP, Narrandes S, Wang Y, Xu W. Applications of Support Vector Machine (SVM) Learning in Cancer Genomics. Cancer genomics & proteomics 2018;15:41-51.
Cited by
Sorted by citations
Machine Learning in Maternity: Early Diagnosis of Placenta Accreta Spectrum
Waszczuk, D., Manikandan, V., Wong, B. L., Martin, F., Bhargava, N., Mondal, A., Loy, M., Strnad, D., Panchatcharam, M., Sadanala, G., et al. (2026) · Cureus · doi:10.7759/cureus.106592
- Authors
- Hu, Y., Liu, T., Pang, S., Ling, X., Wang, Z., & Li, W.
- Affiliations
- Hebei North University (CN)
- Lanzhou University Second Hospital (CN)
- First Hospital of Qinhuangdao (CN)
- Corresponding author
- Li, Wenfei
- Venue
- Journal of Imaging Informatics in Medicine
- Published
- 2025-03-24
- Volume
- 38
- Issue
- 6
- Pages
- 3899-3908
- ISSN
- 0897-1889 (ISSN-L)
- 2948-2933
- PMC
- PMC12701209
- OpenAlex
- W4408845293
- Cited by
- 4
- References
- 32
- Citations by year
- 2026: 3 · 2025: 1
- FWCI
- 6.09
- Type
- journal article
- Language
- en
- Open access
- green
- pmc.ncbi.nlm.nih.gov
- Licence
- Locations
- Journal of Imaging Informatics in Medicine (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (publishedVersion)
- Topics
- Health Sciences › Medicine › Pediatrics, Perinatology and Child Health › Maternal and fetal healthcare
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- placenta accreta spectrum, deep learning diagnosis, DenseNet121, magnetic resonance imaging, radiomics, high-risk pregnancy, cesarean section history, placental thickness, placenta previa, risk factors, external validation
- MeSH terms
- Deep Learning, Convolutional Neural Networks, Adult, Female, Humans, Magnetic Resonance Imaging, Placenta Accreta, Pregnancy, Retrospective Studies, Support Vector Machine
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
