Development and validation of an interpretable model integrating multimodal information for improving ovarian cancer diagnosis.

Ovarian cancer, a group of heterogeneous diseases, presents with extensive characteristics with the highest mortality among gynecological malignancies. Accurate and early diagnosis of ovarian cancer is of great significance. Here, we present OvcaFinder, an interpretable model constructed from ultrasound images-based deep learning (DL) predictions, Ovarian-Adnexal Reporting and Data System scores from radiologists, and routine clinical variables. OvcaFinder outperforms the clinical model and the DL model with area under the curves (AUCs) of 0.978, and 0.947 in the internal and external test datasets, respectively. OvcaFinder assistance led to improved AUCs of radiologists and inter-reader agreement. The average AUCs were improved from 0.927 to 0.977 and from 0.904 to 0.941, and the false positive rates were decreased by 13.4% and 8.3% in the internal and external test datasets, respectively. This highlights the potential of OvcaFinder to improve the diagnostic accuracy, and consistency of radiologists in identifying ovarian cancer.

Development and validation of ultrasound-based radiomics model to predict germline BRCA mutations in patients with breast cancer.

BACKGROUND: Identifying breast cancer (BC) patients with germline breast cancer susceptibility gene (gBRCA) mutation is important. The current criteria for germline testing for BC remain controversial. This study aimed to develop a nomogram incorporating ultrasound radiomic features and clinicopathological factors to predict gBRCA mutations in patients with BC. MATERIALS AND METHODS: In this retrospective study, 497 women with BC who underwent gBRCA genetic testing from March 2013 to May 2022 were included, including 348 for training (84 with and 264 without a gBRCA mutation) and 149 for validation(36 patients with and 113 without a gBRCA mutation). Factors associated with gBRCA mutations were identified to establish a clinicopathological model. Radiomics features were extracted from the intratumoral and peritumoral regions (3 mm and 5 mm) of each image. The least absolute shrinkage and selection operator regression algorithm was used to select the features and logistic regression analysis was used to construct three imaging models. Finally, a nomogram that combined clinicopathological and radiomics features was developed. The models were evaluated based on the area under the receiver operating characteristic curve (AUC), calibration, and clinical usefulness. RESULTS: Age at diagnosis, family history of BC, personal history of other BRCA-related cancers, and human epidermal growth factor receptor 2 status were independent predictors of the clinicopathological model. The AUC of the imaging radiomics model combining intratumoral and peritumoral 3 mm areas in the validation set was 0.783 (95% confidence interval [CI]: 0.702-0.862), which showed the best performance among three imaging models. The nomogram yielded better performance than the clinicopathological model in validation sets (AUC: 0.824 [0.755-0.894] versus 0.659 [0.563-0.755], p = 0.007). CONCLUSION: The nomogram based on ultrasound images and clinicopathological factors performs well in predicting gBRCA mutations in BC patients and may help to improve clinical decisions about genetic testing.

Deep Learning-assisted Diagnosis of Breast Lesions on US Images: A Multivendor, Multicenter Study.

Purpose: To evaluate the diagnostic performance of a deep learning (DL) model for breast US across four hospitals and assess its value to readers with different levels of experience. Materials and Methods: In this retrospective study, a dual attention-based convolutional neural network was built and validated to discriminate malignant tumors from benign tumors by using B-mode and color Doppler US images (n = 45 909, March 2011-August 2018), acquired with 42 types of US machines, of 9895 pathologic analysis-confirmed breast lesions in 8797 patients (27 men and 8770 women; mean age, 47 years ± 12 [SD]). With and without assistance from the DL model, three novice readers with less than 5 years of US experience and two experienced readers with 8 and 18 years of US experience, respectively, interpreted 1024 randomly selected lesions. Differences in the areas under the receiver operating characteristic curves (AUCs) were tested using the DeLong test. Results: The DL model using both B-mode and color Doppler US images demonstrated expert-level performance at the lesion level, with an AUC of 0.94 (95% CI: 0.92, 0.95) for the internal set. In external datasets, the AUCs were 0.92 (95% CI: 0.90, 0.94) for hospital 1, 0.91 (95% CI: 0.89, 0.94) for hospital 2, and 0.96 (95% CI: 0.94, 0.98) for hospital 3. DL assistance led to improved AUCs (P < .001) for one experienced and three novice radiologists and improved interobserver agreement. The average false-positive rate was reduced by 7.6% (P = .08). Conclusion: The DL model may help radiologists, especially novice readers, improve accuracy and interobserver agreement of breast tumor diagnosis using US.Keywords: Ultrasound, Breast, Diagnosis, Breast Cancer, Deep Learning, Ultrasonography Supplemental material is available for this article. © RSNA, 2023.

Contrast-enhanced ultrasound liver imaging reporting and data system v2017: patient outcomes after treatment for early-stage hepatocellular carcinoma nodules with category 3-5 and category M.

OBJECTIVE: To evaluate correlation between contrast-enhanced ultrasonography Liver Imaging Reporting and Data System (CEUS LI-RADS; v. 2017) categories (LR 3-5 vs LR-M) and outcomes in patients with early-stage hepatocellular carcinoma (HCC) after initial therapy. METHODS: In this retrospective study, 272 patients with high risks for HCC and solitary clinically or pathologically confirmed HCC were identified between January 2010 and December 2015. Patients were initially treated by resection and radiofrequency ablation (RFA) according to the Barcelona Clinic Liver Cancer staging system and were followed up until December 31, 2018. Recurrence-free survival (RFS) and overall survival (OS) were compared between nodules assigned as LR 3-5 or LR M according to CEUS LI-RADS v. 2017 by using the Kaplan-Meier curve, log-rank test, and Cox proportional hazard model. RESULTS: Early washout is the key determinating whether a nodule is classed as LR-M. Treatment procedures and LI-RADS category showed an independent correlation with OS and RFS (p < 0.05). LR 3-5 category were more correlated with better OS (88.6 months and 74.2 months, respectively; p = 0.017) compared with LR-M. Surgical resection demonstrated longer OS and RFS than RFA in LR-M patients and longer OS in LR 3-5 patients (p < 0.05). Besides, there was no significantly difference in OS and RFS between two categories in resection (p > 0.05), while for patients treated with RFA, LR 3-5 patients showed significant longer OS and RFS than LR-M patients (p < 0.05). CONCLUSION: Patients with HCC assigned as LR-M showed worse RFS and OS and surgical resection tended to be a more effective treatment for these patients. ADVANCES IN KNOWLEDGE: Putting forward a theory that CEUS LI-RADS categories could independently predict the outcome for patients with solitary HCC at early-stage after initial treatment.

Deep Learning of Liver Contrast-Enhanced Ultrasound to Predict Microvascular Invasion and Prognosis in Hepatocellular Carcinoma.

Background and Aims: Microvascular invasion (MVI) is a well-known risk factor for poor prognosis in hepatocellular carcinoma (HCC). This study aimed to develop a deep convolutional neural network (DCNN) model based on contrast-enhanced ultrasound (CEUS) to predict MVI, and thus to predict prognosis in patients with HCC. Methods: A total of 436 patients with surgically resected HCC who underwent preoperative CEUS were retrospectively enrolled. Patients were divided into training (n = 301), validation (n = 102), and test (n = 33) sets. A clinical model (Clinical model), a CEUS video-based DCNN model (CEUS-DCNN model), and a fusion model based on CEUS video and clinical variables (CECL-DCNN model) were built to predict MVI. Survival analysis was used to evaluate the clinical performance of the predicted MVI. Results: Compared with the Clinical model, the CEUS-DCNN model exhibited similar sensitivity, but higher specificity (71.4% vs. 38.1%, p = 0.03) in the test group. The CECL-DCNN model showed significantly higher specificity (81.0% vs. 38.1%, p = 0.005) and accuracy (78.8% vs. 51.5%, p = 0.009) than the Clinical model, with an AUC of 0.865. The Clinical predicted MVI could not significantly distinguish OS or RFS (both p > 0.05), while the CEUS-DCNN predicted MVI could only predict the earlier recurrence (hazard ratio [HR] with 95% confidence interval [CI 2.92 [1.1-7.75], p = 0.024). However, the CECL-DCNN predicted MVI was a significant prognostic factor for both OS (HR with 95% CI: 6.03 [1.7-21.39], p = 0.009) and RFS (HR with 95% CI: 3.3 [1.23-8.91], p = 0.011) in the test group. Conclusions: The proposed CECL-DCNN model based on preoperative CEUS video can serve as a noninvasive tool to predict MVI status in HCC, thereby predicting poor prognosis.

Comparison of diagnostic accuracy and utility of artificial intelligence-optimized ACR TI-RADS and original ACR TI-RADS: a multi-center validation study based on 2061 thyroid nodules.

OBJECTIVE: To determine if artificial intelligence-based modification of the Thyroid Imaging Reporting Data System (TI-RADS) would be better than the current American College of Radiology (ACR) TI-RADS for risk stratification of thyroid nodules. METHODS: A total of 2061 thyroid nodules (in 1859 patients) sampled with fine-needle aspiration or operation were retrospectively analyzed between January 2017 and July 2020. Two radiologists blinded to the pathologic diagnosis evaluated nodule features in five ultrasound categories and assigned TI-RADS scores by both ACR TI-RADS and AI TI-RADS. Inter-rater agreement was assessed by asking another two radiologists to score a set of 100 nodules independently. The reference standard was postoperative pathological or cytopathological diagnosis according to the Bethesda system. Inter-rater agreement was determined using intraclass correlation coefficient (ICC). RESULTS: AI TI-RADS assigned lower TI-RADS risk levels than ACR TI-RADS (p < 0.001) and had larger area under receiver operating characteristic curve (0.762 vs. 0.679, p < 0.001). The sensitivities of ACR TI-RADS and AI TI-RADS were similar (86.7% vs. 82.2%, p = 0.052), but specificity was higher with AI TI-RADS (70.2% vs. 49.2%, p < 0.001). AI TI-RADS downgraded 743 (48.63%) benign nodules, indicating that 328 (42.3% of 776 biopsied nodules) unnecessary fine-needle aspirations (FNA) could have been avoided. Inter-rater agreement was better with AI TI-RADS than with ACR TI-RADS (ICC, 0.808 vs. 0.861, p < 0.001). CONCLUSION: AI TI-RADS can achieve meaningful reduction in the number of benign thyroid nodules recommended for biopsy and significantly improve specificity despite a slight decrease in sensitivity. KEY POINTS: • AI TI-RADS assigned lower TI-RADS risk levels than ACR TI-RADS, showing similar sensitivity but higher specificity. • Half of the benign nodules can be downgraded of which 42.3% of biopsy nodules avoided unnecessary fine-needle aspiration (FNA). • AI TI-RADS had a better overall inter-rater agreement.

Value of dual-phase, contrast-enhanced CT combined with ultrasound for the diagnosis of metastasis to central lymph nodes in patients with papillary thyroid cancer.

AIMS: To evaluate the diagnostic performance of computed tomography combined with ultrasound (CT/US) in metastatic central lymph nodes (CLNs) compared with US in patients with papillary thyroid cancer (PTC). MATERIAL AND METHODS: Six-hundred patients with surgically proven PTC who underwent both US and CT examination before CLN dissection were evaluated retrospectively. All cases were divided into four subgroups according to the tumor size and number. Diagnostic performances of US, CT and CT/US were evaluated. RESULTS: Among 600 patients with CLN dissection, CT/US showed higher sensitivity (89.10%) and accuracy (83.00%) than US alone (76.06%, 76.50%). In the subgroup of solitary non-microcarcinomas, the AUC of CT/US was significantly higher than that for US alone (0.827 vs. 0.722, P < 0.05). For the subgroup of solitary or multiple microcarcinomas, the performance efficiency of CT had no obvious advantage over that of US (0.698 vs. 0.740, 0.798 vs. 0.802, P > 0.05). For the subgroup of multiple non-microcarcinomas, CT, US and US/CT had high diagnostic efficacy (AUC > 0.8, Accuracy >80%. P > 0.05), and there was no significant difference between them. CONCLUSIONS: In the subgroup of solitary non-microcarcinoma, CT combined with US provided better diagnostic efficacy, and for those cases, complementary CT examination was recommended. In other subgroups, the diagnostic efficacy of US/CT was similar to that of US alone, and there was no significant benefit from additional CT examination.

Ultrasonic RF time series for early assessment of the tumor response to chemotherapy.

Ultrasound radio-frequency (RF) time series have been shown to carry tissue typing information. To evaluate the potential of RF time series for early prediction of tumor response to chemotherapy, 50MCF-7 breast cancer-bearing nude mice were randomized to receive cisplatin and paclitaxel (treatment group; n = 26) or sterile saline (control group; n = 24). Sequential ultrasound imaging was performed on days 0, 3, 6, and 8 of treatment to simultaneously collect B-mode images and RF data. Six RF time series features, slope, intercept, S1, S2, S3, and S4, were extracted during RF data analysis and contrasted with microstructural tumor changes on histopathology. Chemotherapy administration reduced tumor growth relative to control on days 6 and 8. Compared with day 0, intercept, S1, and S2 were increased while slope was decreased on days 3, 6, and 8 in the treatment group. Compared with the control group, intercept, S1, S2, S3, and S4 were increased, and slope was decreased, on days 3, 6, and 8 in the treatment group. Tumor cell density decreased significantly in the latter on day 3. We conclude that ultrasonic RF time series analysis provides a simple way to noninvasively assess the early tumor response to chemotherapy.

The Degree of Contrast Washout on Contrast-Enhanced Ultrasound in Distinguishing Intrahepatic Cholangiocarcinoma from Hepatocellular Carcinoma.

We aim to assess the role and degree of contrast washout in the differential diagnosis of intrahepatic cholangiocarcinoma (ICC) from hepatocellular carcinoma (HCC) on contrast-enhanced ultrasound (CEUS). Fifty-six histopathology-confirmed ICC nodules and 184 HCC nodules were included in this study. The nodules' washout degree on CEUS at 1, 2 and 3 min was semi-quantitatively and qualitatively assessed using gray-scale video signal intensity. Semi-quantitative assessment showed that the washout degree of ICCs at 1, 2 and 3 min were significantly lower than those of HCCs (p < 0.001) and similar results were found in the same size range subgroups. There were no significant differences in the washout degree of ICCs between patients with chronic hepatitis and those without. The areas under receiver operating characteristic curves, using the nodules' washout degree at 1, 2 and 3 min to differentiate ICC from HCC, were 0.957, 0.979 and 0.982, respectively. The qualitative assessment showed the washout of ICCs was more rapid and obvious than that of HCCs. At 3 min, moderate and marked washout were observed in all ICCs, but in only 12.5% HCCs (p < 0.001). In conclusion, ICCs displayed much higher degree of contrast washout than HCCs on CEUS, which allowed for differentiation from HCCs.

Evaluation of renal urothelial carcinoma by contrast-enhanced ultrasonography.

PURPOSE: To observe ultrasonographic features of urothelial carcinoma in renal pelvis and evaluate contrast-enhanced ultrasound (CEUS) in diagnosis. MATERIALS AND METHODS: Fifty-two patients with urothelial carcinoma underwent preoperative conventional US, colour Doppler flow imaging (CDFI) and CEUS. RESULTS: Of 52 total lesions, 41 (78.8%) could be clearly identified by US, and 49 (94.2%) were enhanced by CEUS. Among US-imaged lesions, 39 (95.1%) were solid tumours, and two (4.9%) were mixed solid-cystic; 25 (61.0%) were isoechoic, 11 (26.8%) hypoechoic, and five (12.2%) hyperechoic. Analysis of tumour blood flow by CDFI characterised 17 avascular lesions (41.5% of total), 16 hypovascular (39.0%), and 8 hypervascular (19.5%). The resistance index ranged from 0.65 to 0.88 (mean of 0.71). Enhancement was seen in 49 lesions after injection of SonoVue. A slow enhancement pattern was observed in 36 lesions (73.5%) relative to renal cortex, and 13/49 (26.5%) showed simultaneous enhancement. At peak enhancement, 38 lesions (77.6%) were hypo-enhanced, six (12.2%) iso-enhanced, and five (10.2%) hyper-enhanced. There were 12 lesions with intertumoural necrosis or haemorrhage (24.5%) that were heterogeneously enhanced, and 37 (75.5%) were homogeneously enhanced. A fast washout pattern was observed in 46 lesions (93.9%), synchronous washout in two (4.08%), and slow washout in one (2.04%). CONCLUSIONS: Slow-in, fast-out, and hypo-enhancement properties are associated with renal urothelial carcinoma and may thus have diagnostic value. We found that CEUS is able to identify tumours that are ambiguous by conventional US, and it thus significantly improves the confidence of diagnosis.