Dataset Information

Deep learning analysis of contrast-enhanced spectral mammography to determine histoprognostic factors of malignant breast tumours.

ABSTRACT:

Objective

To evaluate if a deep learning model can be used to characterise breast cancers on contrast-enhanced spectral mammography (CESM).

Methods

This retrospective mono-centric study included biopsy-proven invasive cancers with an enhancement on CESM. CESM images include low-energy images (LE) comparable to digital mammography and dual-energy subtracted images (DES) showing tumour angiogenesis. For each lesion, histologic type, tumour grade, estrogen receptor (ER) status, progesterone receptor (PR) status, HER-2 status, Ki-67 proliferation index, and the size of the invasive tumour were retrieved. The deep learning model used was a CheXNet-based model fine-tuned on CESM dataset. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was calculated for the different models: images by images and then by majority voting combining all the incidences for one tumour.

Results

In total, 447 invasive breast cancers detected on CESM with pathological evidence, in 389 patients, which represented 2460 images analysed, were included. Concerning the ER, the deep learning model on the DES images had an AUC of 0.83 with the image-by-image analysis and of 0.85 for the majority voting. For the triple-negative analysis, a high AUC was observable for all models, in particularity for the model on LE images with an AUC of 0.90 for the image-by-image analysis and 0.91 for the majority voting. The AUC for the other histoprognostic factors was lower.

Conclusion

Deep learning analysis on CESM has the potential to determine histoprognostic tumours makers, notably estrogen receptor status, and triple-negative receptor status.

Key points

• A deep learning model developed for chest radiography was adapted by fine-tuning to be used on contrast-enhanced spectral mammography. • The adapted models allowed to determine for invasive breast cancers the status of estrogen receptors and triple-negative receptors. • Such models applied to contrast-enhanced spectral mammography could provide rapid prognostic and predictive information.

SUBMITTER: Dominique C

PROVIDER: S-EPMC8800426 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Contrast-enhanced MRI (CE-MRI) is the most sensitive technique for breast cancer detection. Contrast-enhanced mammography (CEM) is emerging as a possible alternative to CE-MRI.PurposeTo evaluate the diagnostic performance of a low radiation dose contrast-enhanced mammography (L-CEM) in women with suspicious findings on conventional imaging compared to CE-MRI of the breast.Study typeProspective, single center.PopulationWomen with suspicious findings on mammography, tomosynthesis, or ultrasound, and no contraindications for L-CEM or CE-MRI. Eighty women were included.Field strength/sequence1.5 and 3T CE-MRI, standard protocol for breast, with dedicated coils, according to international guidelines. L-CEM was performed using a dedicated prototype.AssessmentThree, off-site, blinded readers evaluated the images according to the BI-RADS lexicon in a randomized order, each in two separate reading sessions. Histology served as a gold standard.Statistical testLesion detection rate, sensitivity, specificity, and negative and positive predictive values (NPV, PPV) were calculated and compared with multivariate statistics.ResultsIncluded were 80 women (mean age, 54.3 years ±11.2 standard deviation) with 93 lesions (32 benign, 61 malignant). The detection rate was significantly higher with CE-MRI (92.5-94.6%; L-CEM 79.6-91.4%, P = 0.014). Sensitivity (L-CEM 65.6-90.2%; CE-MRI 83.6-93.4%, P = 0.086) and NPV (L-CEM 59.6-71.4%; CE-MRI 63.0-76.5%, P = 0.780) did not differ between the modalities. Specificity (L-CEM 46.9-96.9%; CE-MRI 37.5-53.1%, P = 0.001) and PPV (L-CEM 76.4-97.6%; CE-MRI 73.3-77.3%, P = 0.007) were significantly higher with L-CEM. Variations between readers were significant for sensitivity and NPV. The accuracy of L-CEM was as good as CE-MRI (75.3-76.3% vs. 72.0-75.3%, P = 0.514).Data conclusionL-CEM showed a high sensitivity and accuracy in women with suspicious findings on conventional imaging. Compared to CE-MRI, L-CEM has the potential to increase specificity and PPV. L-CEM might help to reduce false-positive biopsies while obtaining sensitivity comparable to that of CE-MRI LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:589-595.

Project description:BackgroundLiver cancer remains the leading cause of cancer death globally, and the treatment strategies are distinct for each type of malignant hepatic tumors. However, the differential diagnosis before surgery is challenging and subjective. This study aims to build an automatic diagnostic model for differentiating malignant hepatic tumors based on patients' multimodal medical data including multi-phase contrast-enhanced computed tomography and clinical features.MethodsOur study consisted of 723 patients from two centers, who were pathologically diagnosed with HCC, ICC or metastatic liver cancer. The training set and the test set consisted of 499 and 113 patients from center 1, respectively. The external test set consisted of 111 patients from center 2. We proposed a deep learning model with the modular design of SpatialExtractor-TemporalEncoder-Integration-Classifier (STIC), which take the advantage of deep CNN and gated RNN to effectively extract and integrate the diagnosis-related radiological and clinical features of patients. The code is publicly available at https://github.com/ruitian-olivia/STIC-model .ResultsThe STIC model achieved an accuracy of 86.2% and AUC of 0.893 for classifying HCC and ICC on the test set. When extended to differential diagnosis of malignant hepatic tumors, the STIC model achieved an accuracy of 72.6% on the test set, comparable with the diagnostic level of doctors' consensus (70.8%). With the assistance of the STIC model, doctors achieved better performance than doctors' consensus diagnosis, with an increase of 8.3% in accuracy and 26.9% in sensitivity for ICC diagnosis on average. On the external test set from center 2, the STIC model achieved an accuracy of 82.9%, which verify the model's generalization ability.ConclusionsWe incorporated deep CNN and gated RNN in the STIC model design for differentiating malignant hepatic tumors based on multi-phase CECT and clinical features. Our model can assist doctors to achieve better diagnostic performance, which is expected to serve as an AI assistance system and promote the precise treatment of liver cancer.

Project description:BackgroundIn the Dutch breast cancer screening program, women recalled with a BI-RADS 0 score are referred for additional imaging, while those with BI-RADS 4/5 scores are also directed to an outpatient breast clinic. Approximately six out of ten women are recalled without being diagnosed with a malignancy. However, these recalls require additional imaging and doctor visits, which result in patient anxiety and increased health care costs. Conventional types of imaging used for additional imaging are full-field digital mammography and tomosynthesis. Contrast-enhanced spectral mammography has proved to have higher sensitivity and specificity than conventional imaging in women recalled from screening. Therefore, the aim is to study if CESM instead of conventional imaging is a more accurate, patient-friendly, and cost-effective strategy in the work-up of women recalled from breast cancer screening.MethodsThis prospective, multicenter, randomized controlled trial will be conducted at four centers and will include 528 patients recalled for suspicious breast lesions from the Dutch breast cancer screening program. Participants are randomized in two groups: (1) standard care using conventional breast imaging techniques as initial imaging after recall versus (2) work-up primarily based on CESM. Written informed consent will be collected prior to study inclusion. The primary outcome is the diagnostic accuracy for detection of breast cancer. Secondary outcomes are numbers of additional diagnostic exams, days until final diagnosis, health care costs, and experienced patient anxiety.DiscussionBased on previously published retrospective studies, we expect to demonstrate in this prospective multicenter randomized controlled trial, that using CESM as a primary work-up tool in women recalled from breast cancer screening is a more accurate, cost-effective, and patient-friendly strategy.Trial registrationNetherlands Trial Register, NL6413/NTR6589. Registered on 6 July, 2017.