Prediction of microvascular invasion in hepatocellular carcinoma patients with MRI radiomics based on susceptibility weighted imaging and T2-weighted imaging.

BACKGROUND: The accurate identification of microvascular invasion (MVI) in patients with hepatocellular carcinoma (HCC) is of great clinical importance. PURPOSE: To develop a radiomics nomogram based on susceptibility-weighted imaging (SWI) and T2-weighted imaging (T2WI) for predicting MVI in early-stage (Barcelona Clinic Liver Cancer stages 0 and A) HCC patients. MATERIALS AND METHODS: A prospective cohort of 189 participants with HCC was included for model training and testing, and an additional 34 participants were enrolled for external validation. ITK-SNAP was used to manually segment the tumour, and PyRadiomics was used to extract radiomic features from the SWI and T2W images. Variance filtering, student's t test, least absolute shrinkage and selection operator regression and random forest (RF) were applied to select meaningful features. Four machine learning classifiers, including K-nearest neighbour, RF, logistic regression and support vector machine-based models, were established. Independent clinical and radiological risk factors were also determined to establish a clinical model. The best radiomics and clinical models were further evaluated in the validation set. In addition, a nomogram was constructed from the radiomic model and independent clinical factors. Diagnostic efficacy was evaluated by receiver operating characteristic curve analysis with fivefold cross-validation. RESULTS: AFP levels greater than 400 ng/mL [odds ratio (OR) 2.50; 95% confidence interval (CI) 1.239-5.047], tumour diameter greater than 5 cm (OR 2.39; 95% CI 1.178-4.839), and absence of pseudocapsule (OR 2.053; 95% CI 1.007-4.202) were found to be independent risk factors for MVI. The areas under the curve (AUCs) of the best radiomic model were 1.000 and 0.882 in the training and testing cohorts, respectively, while those of the clinical model were 0.688 and 0.6691. In the validation set, the radiomic model achieved better diagnostic performance (AUC = 0.888) than the clinical model (AUC = 0.602). The combination of clinical factors and the radiomic model yielded a nomogram with the best diagnostic performance (AUC = 0.948). CONCLUSION: SWI and T2WI-derived radiomic features are valuable for noninvasively and accurately identifying MVI in early-stage HCC. Furthermore, the integration of radiomics and clinical factors yielded a predictive nomogram with satisfactory diagnostic performance and potential clinical benefits.

Automated segmentation of brain metastases with deep learning: a multi-center, randomized crossover, multi-reader evaluation study.

BACKGROUND: Artificial intelligence has been proposed for brain metastasis (BM) segmentation but it has not been fully clinically validated. The aim of this study was to develop and evaluate a system for BM segmentation. METHODS: A deep-learning-based BM segmentation system (BMSS) was developed using contrast-enhanced MR images from 488 patients with 10,338 brain metastases. A randomized crossover, multi-reader study was then conducted to evaluate the performance of the BMSS for BM segmentation using data prospectively collected from 50 patients with 203 metastases at five centers. Five radiology residents and five attending radiologists were randomly assigned to contour the same prospective set in assisted and unassisted modes. Aided and unaided Dice similarity coefficients (DSCs) and contouring times per lesion were compared. RESULTS: The BMSS alone yielded a median DSC of 0.91 (95% confidence interval, 0.90-0.92) in the multi-center set and showed comparable performance between the internal and external sets (p = 0.67). With BMSS assistance, the readers increased the median DSC from 0.87 (0.87-0.88) to 0.92 (0.92-0.92) (p < 0.001) with a median time saving of 42% (40-45%) per lesion. Resident readers showed a greater improvement than attending readers in contouring accuracy (improved median DSC, 0.05 [0.05-0.05] vs. 0.03 [0.03-0.03]; p < 0.001), but a similar time reduction (reduced median time, 44% [40-47%] vs. 40% [37-44%]; p = 0.92) with BMSS assistance. CONCLUSIONS: The BMSS can be optimally applied to improve the efficiency of brain metastasis delineation in clinical practice.

Deep learning model based on contrast-enhanced MRI for predicting post-surgical survival in patients with hepatocellular carcinoma.

Objective: To develop a deep learning model based on contrast-enhanced magnetic resonance imaging (MRI) data to predict post-surgical overall survival (OS) in patients with hepatocellular carcinoma (HCC). Methods: This bi-center retrospective study included 564 surgically resected patients with HCC and divided them into training (326), testing (143), and external validation (95) cohorts. This study used a three-dimensional convolutional neural network (3D-CNN) ResNet to learn features from the pretreatment MR images (T1WIpre, late arterial phase, and portal venous phase) and got the deep learning score (DL score). Three cox regression models were established separately using the DL score (3D-CNN model), clinical features (clinical model), and a combination of above (combined model). The concordance index (C-index) was used to evaluate model performance. Results: We trained a 3D-CNN model to get DL score from samples. The C-index of the 3D-CNN model in predicting 5-year OS for the training, testing, and external validation cohorts were 0.746, 0.714, and 0.698, respectively, and were higher than those of the clinical model, which were 0.675, 0.674, and 0.631, respectively (P = 0.009, P = 0.204, and P = 0.092, respectively). The C-index of the combined model for testing and external validation cohorts was 0.750 and 0.723, respectively, significantly higher than the clinical model (P = 0.017, P = 0.016) and the 3D-CNN model (P = 0.029, P = 0.036). Conclusions: The combined model integrating the DL score and clinical factors showed a higher predictive value than the clinical and 3D-CNN models and may be more useful in guiding clinical treatment decisions to improve the prognosis of patients with HCC.

Magnetic Resonance Deep Learning Radiomic Model Based on Distinct Metastatic Vascular Patterns for Evaluating Recurrence-Free Survival in Hepatocellular Carcinoma.

BACKGROUND: The metastatic vascular patterns of hepatocellular carcinoma (HCC) are mainly microvascular invasion (MVI) and vessels encapsulating tumor clusters (VETC). However, most existing VETC-related radiological studies still focus on the prediction of VETC status. PURPOSE: This study aimed to build and compare VETC-MVI related models (clinical, radiomics, and deep learning) associated with recurrence-free survival of HCC patients. STUDY TYPE: Retrospective. POPULATION: 398 HCC patients (349 male, 49 female; median age 51.7 years, and age range: 22-80 years) who underwent resection from five hospitals in China. The patients were randomly divided into training cohort (n = 358) and test cohort (n = 40). FIELD STRENGTH/SEQUENCE: 3-T, pre-contrast T1-weighted imaging spoiled gradient recalled echo (T1WI SPGR), T2-weighted imaging fast spin echo (T2WI FSE), and contrast enhanced arterial phase (AP), delay phase (DP). ASSESSMENT: Two radiologists performed the segmentation of HCC on T1WI, T2WI, AP, and DP images, from which radiomic features were extracted. The RFS related clinical characteristics (VETC, MVI, Barcelona stage, tumor maximum diameter, and alpha fetoprotein) and radiomic features were used to build the clinical model, clinical-radiomic (CR) nomogram, deep learning model. The follow-up process was done 1 month after resection, and every 3 months subsequently. The RFS was defined as the date of resection to the date of recurrence confirmed by radiology or the last follow-up. Patients were followed up until December 31, 2022. STATISTICAL TESTS: Univariate COX regression, least absolute shrinkage and selection operator (LASSO), Kaplan-Meier curves, log-rank test, C-index, and area under the curve (AUC). P < 0.05 was considered statistically significant. RESULTS: The C-index of deep learning model achieved 0.830 in test cohort compared with CR nomogram (0.731), radiomic signature (0.707), and clinical model (0.702). The average RFS of the overall patients was 26.77 months (range 1-80 months). DATA CONCLUSION: MR deep learning model based on VETC and MVI provides a potential tool for survival assessment. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 3.

Quantitative synthetic MRI for predicting locally advanced rectal cancer response to neoadjuvant chemoradiotherapy.

OBJECTIVES: To investigate the value of pre-treatment quantitative synthetic MRI (SyMRI) for predicting a good response to neoadjuvant chemoradiotherapy (nCRT) in patients with locally advanced rectal cancer. METHODS: This prospective study enrolled 63 patients with locally advanced rectal cancer scheduled to undergo preoperative chemoradiotherapy from January 2019 to June 2021. T1 relaxation time (T1), T2 relaxation time (T2), proton density (PD) from synthetic MRI, and apparent diffusion coefficient (ADC) from diffusion-weighted imaging (DWI) were measured. Independent-sample t-test, the Mann-Whitney U test, the Delong test, and receiver operating characteristic curve (ROC) analyses were used to predict the pathologic complete response (pCR) and T-downstaging. RESULTS: Among the 63 patients, 19 (30%) achieved pCR and 44 (70%) did not, and 24 (38%) achieved T-downstaging, while 44 (62%) did not. The mean T1 and T2 values were significantly lower in the pCR group compared with those in the non-pCR group and in the T-downstage group compared with those in the non-T-downstage group (all p < 0.05). There were no significant differences in the PD and ADC values between the two groups. There were no significant differences between the mean values of T1 and T2 for predicting pCR after CRT (AUC, 0.767 vs. 0.831, p = 0.37). There were no significant differences between the AUC values of T1 and T2 values for the assessment of post-CRT T-downstaging (AUC, 0.746 vs. 0.820, p = 0.506). CONCLUSIONS: In patients with locally advanced rectal cancer, the synthetic MRI-derived T1 relaxation time and T2 relaxation time values are promising imaging markers for predicting a good response to neoadjuvant chemoradiotherapy. KEY POINTS: • Mean T1 and T2 values were significantly lower in the pathologic complete response group and the T-downstage group. • There were no significant differences in the proton density and apparent diffusion coefficient values between the two groups.

False-negative and false-positive outcomes of computer-aided detection on brain metastasis: Secondary analysis of a multicenter, multireader study.

BACKGROUND: Errors have seldom been evaluated in computer-aided detection on brain metastases. This study aimed to analyze false negatives (FNs) and false positives (FPs) generated by a brain metastasis detection system (BMDS) and by readers. METHODS: A deep learning-based BMDS was developed and prospectively validated in a multicenter, multireader study. Ad hoc secondary analysis was restricted to the prospective participants (148 with 1,066 brain metastases and 152 normal controls). Three trainees and 3 experienced radiologists read the MRI images without and with the BMDS. The number of FNs and FPs per patient, jackknife alternative free-response receiver operating characteristic figure of merit (FOM), and lesion features associated with FNs were analyzed for the BMDS and readers using binary logistic regression. RESULTS: The FNs, FPs, and the FOM of the stand-alone BMDS were 0.49, 0.38, and 0.97, respectively. Compared with independent reading, BMDS-assisted reading generated 79% fewer FNs (1.98 vs 0.42, P < .001); 41% more FPs (0.17 vs 0.24, P < .001) but 125% more FPs for trainees (P < .001); and higher FOM (0.87 vs 0.98, P < .001). Lesions with small size, greater number, irregular shape, lower signal intensity, and located on nonbrain surface were associated with FNs for readers. Small, irregular, and necrotic lesions were more frequently found in FNs for BMDS. The FPs mainly resulted from small blood vessels for the BMDS and the readers. CONCLUSIONS: Despite the improvement in detection performance, attention should be paid to FPs and small lesions with lower enhancement for radiologists, especially for less-experienced radiologists.

Application of synthetic magnetic resonance imaging and DWI for evaluation of prognostic factors in cervical carcinoma: a prospective preliminary study.

OBJECTIVES: To determine the values of quantitative metrics derived from synthetic MRI (SyMRI) and apparent diffusion coefficient (ADC) in evaluating the prognostic factors of cervical carcinoma (CC). METHODS: In this prospective study, 74 patients with pathologically confirmed CC were enrolled. Pretreatment quantitative metrics including T1, T2 and ADC values were obtained from SyMRI and diffusion-weighted imaging (DWI) sequences. The values of all metrics were compared for different prognostic features using Student's t-test or Mann-Whitney U-test. The receiver operating characteristic (ROC) curve and multivariate logistic regression analysis were utilized to evaluate the diagnostic performance of quantitative variables. RESULTS: T1 and T2 values of parametrial involvement (PMI)-negative were significantly higher than those of PMI-positive (p = 0.002 and < 0.001), while ADC values did not show a significant difference. The area under curve (AUC) of T1 and T2 values for identifying PMI were 0.743 and 0.831. Only the T2 values showed a significant difference between the lymphovascular space involvement (LVSI)-negative and LVSI-positive (p < 0.001), and the AUC of T2 values for discriminating LVSI was 0.814. The differences of T1, T2, and ADC values between the well/moderately and the poorly differentiated CC were significant (all p < 0.001). The AUCs of T1, T2 and ADC values for predicting differentiation grades were 0.762, 0.830, and 0.808. The combined model of all metrics proved to achieve good diagnostic performance with the AUC of 0.866. CONCLUSION: SyMRI may be a potential noninvasive tool for assessing the prognostic factors such as PMI, LVSI, and differentiation grades in CC. Moreover, the overall diagnostic performances of synthetic quantitative metrics were superior to the ADC values, especially in identifying PMI and LVSI. ADVANCES IN KNOWLEDGE: This is the first study to assess the utility of SyMRI-derived parameters and ADC value in evaluating the prognostic factors in CC.

Diagnostic performance of synthetic magnetic resonance imaging in the prognostic evaluation of rectal cancer.

Background: Numerous factors are related to the prognosis of rectal cancer, including T stage, N stage, metastasis, extramural venous invasion (EMVI), circumferential resection margin (CRM), and tumor differentiation. However, it is still a challenge to precisely evaluate them before therapy; therefore, we investigate whether synthetic magnetic resonance imaging and apparent diffusion coefficient (ADC) values could help predict the prognostic factors of rectal cancer. Methods: Eighty-seven patients (55 men and 32 women; mean age, 59±11 years) with pathologically confirmed rectal cancer were enrolled. Preoperative quantitative metrics, including T1, T2, proton density (PD), and ADC values, were measured with diffusion-weighted imaging (DWI) acquired by a single-shot echo-planar sequence and synthetic magnetic resonance imaging acquired by a multi-dynamic multi-echo sequence at 3.0 T, in patients with rectal cancer by two radiologists. We evaluated the diagnostic performance of synthetic magnetic resonance imaging using the independent sample t-test or Mann-Whitney U test and receiver operating characteristic (ROC) curve and multivariate logistic regression analyses and compared the area under the ROC curve of quantitative values using the DeLong test. Results: The T2 and PD values showed a significant reduction among patients with poor differentiation and lymph node metastasis in rectal cancer. The area under the ROC curve values of T2 and PD values for predicting magnetic resonance imaging N stage and differentiation were 0.734, 0.682, and 0.673, 0.686, respectively. Moreover, combining T2 and PD values for magnetic resonance imaging N stage slightly improved the area under the ROC curve value of 0.774 (95% CI, 0.673-0.876). In the present study, the ADC and T1 values were not significant in the differentiation or clinical stage of rectal cancer (RC). Conclusions: Quantitative T2 and PD values obtained by synthetic magnetic resonance imaging might be used for evaluating prognostic factors of rectal cancer noninvasively. Furthermore, combining T2 and PD values further improved the diagnostic performance of magnetic resonance imaging N staging in rectal cancer. The ADC and T1 values were not significant in the differentiation or clinical stage of RC.

Discriminating rectal cancer grades using restriction spectrum imaging.

PURPOSE: Restriction spectrum imaging (RSI) is a novel diffusion MRI model that separates water diffusion into several microscopic compartments. The restricted compartment correlating to the tumor cellularity is expected to be a potential indicator of rectal cancer aggressiveness. Our aim was to assess the ability of RSI model for rectal tumor grading. METHODS: Fifty-eight patients with different rectal cancer grading confirmed by biopsy were involved in this study. DWI acquisitions were performed using single-shot echo-planar imaging (SS-EPI) with multi-b-values at 3 T. We applied a three-compartment RSI model, along with ADC model and diffusion kurtosis imaging (DKI) model, to DWI images of 58 patients. ROC and AUC were used to compare the performance of the three models in differentiating the low grade (G1 + G2) and high grade (G3). Mean ± standard deviation, ANOVA, ROC analysis, and correlation analysis were used in this study. RESULTS: The volume fraction of restricted compartment C1 from RSI was significantly correlated with grades (r = 0.403, P = 0.002). It showed significant difference between G1 and G3 (P = 0.008) and between G2 and G3 (P = 0.01). As for the low-grade and high-grade discrimination, significant difference was found in C1 (P < 0.001). The AUC of C1 for differentiation between low-grade and high-grade groups was 0.753 with a sensitivity of 72.0% and a specificity of 70.0%. CONCLUSION: The three-compartment RSI model was able to discriminate the rectal cancer of low and high grades. The results outperform the traditional ADC model and DKI model in rectal cancer grading.

Nomograms for Predicting Hepatocellular Carcinoma Recurrence and Overall Postoperative Patient Survival.

Background: Few studies have focused on the prognosis of patients with hepatocellular carcinoma (HCC) of Barcelona Clinic Liver Cancer (BCLC) stage 0‒C in terms of early recurrence and 5-years overall survival (OS). We sought to develop nomograms for predicting 5-year OS and early recurrence after curative resection of HCC, based on a clinicopathological‒radiological model. We also investigated whether different treatment methods influenced the OS of patients with early recurrence. Methods: Retrospective data, including clinical pathology, radiology, and follow-up data, were collected for 494 patients with HCC who underwent hepatectomy. Nomograms estimating OS and early recurrence were constructed using multivariate Cox regression analysis, based on the random survival forest (RSF) model. We evaluated the discrimination and calibration abilities of the nomograms using concordance indices (C-index), calibration curves, and Kaplan‒Meier curves. OS curves of different treatments for patients who had recurrence within 2 years after curative surgery were depicted and compared using the Kaplan-Meier method and the log-rank test. Results: Multivariate Cox regression revealed that BCLC stage, non-smooth margin, maximum tumor diameter, age, aspartate aminotransferase levels, microvascular invasion, and differentiation were prognostic factors for OS and were incorporated into the nomogram with good predictive performance in the training (C-index: 0.787) and testing cohorts (C-index: 0.711). A nomogram for recurrence-free survival was also developed based on four prognostic factors (BCLC stage, non-smooth margin, maximum tumor diameter, and microvascular invasion) with good predictive performance in the training (C-index: 0.717) and testing cohorts (C-index: 0.701). In comparison to the BCLC staging system, the C-index (training cohort: 0.787 vs. 0.678, 0.717 vs. 0.675; external cohort 2: 0.748 vs. 0.624, 0.729 vs. 0.587 respectively, for OS and RFS; external cohort1:0.716 vs. 0.627 for RFS, all p value<0.05), and model calibration curves all showed improved performance. Patients who underwent surgery after tumor recurrence had a higher reOS than those who underwent comprehensive treatments and supportive care. Conclusions: The nomogram, based on clinical, pathological, and radiological factors, demonstrated good accuracy in estimating OS and recurrence, which can guide follow-up and treatment of individual patients. Reoperation may be the best option for patients with recurrence in good condition.

Development and validation of a deep-learning model for detecting brain metastases on 3D post-contrast MRI: a multi-center multi-reader evaluation study.

BACKGROUND: Accurate detection is essential for brain metastasis (BM) management, but manual identification is laborious. This study developed, validated, and evaluated a BM detection (BMD) system. METHODS: Five hundred seventy-three consecutive patients (10 448 lesions) with newly diagnosed BMs and 377 patients without BMs were retrospectively enrolled to develop a multi-scale cascaded convolutional network using 3D-enhanced T1-weighted MR images. BMD was validated using a prospective validation set comprising an internal set (46 patients with 349 lesions; 44 patients without BMs) and three external sets (102 patients with 717 lesions; 108 patients without BMs). The lesion-based detection sensitivity and the number of false positives (FPs) per patient were analyzed. The detection sensitivity and reading time of three trainees and three experienced radiologists from three hospitals were evaluated using the validation set. RESULTS: The detection sensitivity and FPs were 95.8% and 0.39 in the test set, 96.0% and 0.27 in the internal validation set, and ranged from 88.9% to 95.5% and 0.29 to 0.66 in the external sets. The BMD system achieved higher detection sensitivity (93.2% [95% CI, 91.6-94.7%]) than all radiologists without BMD (ranging from 68.5% [95% CI, 65.7-71.3%] to 80.4% [95% CI, 78.0-82.8%], all P < .001). Radiologist detection sensitivity improved with BMD, reaching 92.7% to 95.0%. The mean reading time was reduced by 47% for trainees and 32% for experienced radiologists assisted by BMD relative to that without BMD. CONCLUSIONS: BMD enables accurate BM detection. Reading with BMD improves radiologists' detection sensitivity and reduces their reading times.