Search

PURPOSE: To explore the predictive potential of serial computed tomography (CT) radiology reports for pancreatic cancer survival using natural language processing (NLP). METHODS: Deep-transfer-learning–based NLP models were retrospectively trained and tested with serial, free-text CT reports, and survival information of consecutive patients diagnosed with pancreatic cancer in a Korean tertiary hospital was extracted. Randomly selected patients with pancreatic cancer and their serial CT reports from an independent tertiary hospital in the United States were included in the external testing data set. The concordance index (c-index) of predicted survival and actual survival, and area under the receiver operating characteristic curve (AUROC) for predicting 1-year survival were calculated. RESULTS: Between January 2004 and June 2021, 2,677 patients with 12,255 CT reports and 670 patients with 3,058 CT reports were allocated to training and internal testing data sets, respectively. ClinicalBERT (Bidirectional Encoder Representations from Transformers) model trained on the single, first CT reports showed a c-index of 0.653 and AUROC of 0.722 in predicting the overall survival of patients with pancreatic cancer. ClinicalBERT trained on up to 15 consecutive reports from the initial report showed an improved c-index of 0.811 and AUROC of 0.911. On the external testing set with 273 patients with 1,947 CT reports, the AUROC was 0.888, indicating the generalizability of our model. Further analyses showed our model's contextual interpretation beyond specific phrases. CONCLUSION: Deep-transfer-learning–based NLP model of serial CT reports can predict the survival of patients with pancreatic cancer. Clinical decisions can be supported by the developed model, with survival information extracted solely from serial radiology reports. Survival of patients with pancreatic cancer can be predicted solely by leveraging an NLP model on serial CT reports. [ABSTRACT FROM AUTHOR]

Background: According to the Liver Imaging Reporting and Data System (LI-RADS), the LI-RADS category M (LR-M), which are probably or definitely malignant but are not specific for hepatocellular carcinomas (HCCs), does not exclude HCCs. A gap in knowledge remains, including their characteristics and recurrence of HCCs categorized as LR-M.Purpose: To compare the characteristics of HCCs categorized as LR-M with HCCs categorized as LR-4 or LR-5 (LR-4/5) using the LI-RADS version 2018 and evaluate the relationship of these categories with the risk of early recurrence after curative resections of single HCCs.Study Type: Retrospective.Subjects: Two hundred and eighty-one patients (mean age, 57 years; 191 men and 90 women) who underwent curative resections for single HCCs and preoperative contrast-enhanced MRI between 2015 and 2017.Field Strength/sequence: 3T Dual gradient-echo T1 WI with in- and opposed-phase, turbo spin-echo T2 WI, diffusion-weighted echo-planar images, and three-dimensional gradient-echo T1 WI before and after administration of contrast agent.Assessment: MRI features according to the LI-RADS version 2018 were evaluated and LI-RADS category were assigned for each observation. Clinical, imaging, and histopathological features were compared based on LI-RADS categorization. Early recurrence rates (<2 years) and associated factors were also evaluated.Statistical Tests: Fisher's exact test, two-sample t test after satisfying assumption of normality through Shapiro-Wilk test, Fleiss κ coefficient, Cox proportional hazards regression analysis, Kaplan-Meier method, and log-rank test.Results: Forty-one HCCs (14.6%) were categorized as LR-M and 240 HCCs (85.4%) were categorized as LR-4/5. LR-M HCCs showed poorer differentiation than LR-4/5 HCCs. In the multivariate analysis, the LR-M category was an independent predictor for early recurrence (hazard ratio, 1.904; 95% confidence interval, 1.024-3.542; P < 0.05). Early recurrence rates were significantly higher in patients with LR-M HCCs than in patients with LR-4/5 HCCs (32.0% vs. 18.4%, respectively, P < 0 05).Data Conclusion: Compared to LR-4/5 HCCs, LR-M HCCs were associated with poorer tumor differentiation and higher early recurrence rates after curative resections of single HCCs.Level Of Evidence: 3 Technical Efficacy Stage: 2. [ABSTRACT FROM AUTHOR]

Purpose: The present study aimed to investigate the frequency and extent of compensatory common bile duct (CBD) dilatation after cholecystectomy, assess the time between cholecystectomy and CBD dilatation, and identify potentially useful CT findings suggestive of obstructive CBD dilatation., Materials and Methods: This retrospective study included 121 patients without biliary obstruction who underwent multiple CT scans before and after cholecystectomy at a single center between 2009 and 2011. The maximum short-axis diameters of the CBD and intrahepatic duct (IHD) were measured on each CT scan. In addition, the clinical and CT findings of 11 patients who were initially excluded from the study because of CBD stones or periampullary tumors were examined to identify distinguishing features between obstructive and non-obstructive CBD dilatation after cholecystectomy., Results: The mean (standard deviation) short-axis maximum CBD diameter of 121 patients was 5.6 (± 1.9) mm in the axial plane before cholecystectomy but increased to 7.9 (± 2.6) mm after cholecystectomy ( p < 0.001). Of the 106 patients with a pre-cholecystectomy axial CBD diameter of < 8 mm, 39 (36.8%) showed CBD dilatation of ≥ 8 mm after cholecystectomy. Six of the 17 patients with longterm (> 2 years) serial follow-up CT scans (35.3%) eventually showed a significant (> 1.5-fold) increase in the axial CBD diameter, all within two years after cholecystectomy. Of the 121 patients without obstruction or related symptoms, only one patient (0.1%) showed IHD dilatation > 3 mm after cholecystectomy. In contrast, all 11 patients with CBD obstruction had abdominal pain and abnormal laboratory indices, and 81.8% (9/11) had significant dilatation of the IHD and CBD., Conclusion: Compensatory non-obstructive CBD dilatation commonly occurs after cholecystectomy to a similar extent as obstructive dilatation. However, the presence of relevant symptoms, significant IHD dilatation, or further CBD dilatation 2-3 years after cholecystectomy should raise suspicion of CBD obstruction., Competing Interests: Conflicts of Interest: The authors have no potential conflicts of interest to disclose., (Copyrights © 2024 The Korean Society of Radiology.)

BACKGROUND. In LI-RADS version 2018, observations showing at least one of five targetoid appearances in different sequences or postcontrast phases are categorized LR-M, indicating likely non-hepatocellular carcinoma (HCC) malignancy. OBJECTIVE. The purpose of this study was to evaluate interobserver agreement for LI-RADS targetoid appearances among a large number of radiologists of varying experience and the diagnostic performance of targetoid appearances for differentiating HCC from non-HCC malignancy. METHODS. This retrospective study included 100 patients (76 men, 24 women; mean age, 58 ± 9 [SD] years) at high risk of HCC who underwent gadoxetic acid-enhanced MRI within 30 days before hepatic tumor resection (25 randomly included patients with non-HCC malignancy [13, intrahepatic cholangiocarcinoma; 12, combined HCC-cholangiocarcinoma]; 75 matched patients with HCC). Eight radiologists (four more experienced [8-15 years]; four less experienced [1-5 years]) from seven institutions independently assessed observations for the five targetoid appearances and LI-RADS categorization. Interobserver agreement and diagnostic performance for non-HCC malignancy were evaluated. RESULTS. Interobserver agreement was poor for peripheral washout (κ = 0.20); moderate for targetoid transitional phase or hepatobiliary phase appearance (κ = 0.33), delayed central enhancement (κ = 0.37), and targetoid restriction (κ = 0.43); and substantial for rim arterial phase hyperenhancement (κ = 0.61). Agreement was fair for at least one targetoid appearance (κ = 0.36) and moderate for at least two, three, or four targetoid appearances (κ = 0.43-0.51). Agreement for individual targetoid appearances was not significantly different between more experienced and less experienced readers other than for targetoid restriction (κ = 0.63 vs 0.43; p = .001). Agreement for at least one targetoid appearance was fair among more experienced (κ = 0.29) and less experienced (κ = 0.37) reviewers. Agreement for at least two, three, or four targetoid appearances was moderate to substantial among more experienced reviewers (κ = 0.45-0.63) and moderate among less experienced reviewers (κ = 0.42-0.56). Existing LR-M criteria of at least one targetoid appearance had median accuracy for non-HCC malignancy of 62%, sensitivity of 84%, and specificity of 54%. For all reviewers, accuracy was highest when at least three (median accuracy, 79%; sensitivity, 68%; specificity, 82%) or four (median accuracy, 80%; sensitivity, 54%; specificity, 88%) targetoid appearances were required. CONCLUSION. Targetoid appearances and LR-M categorization exhibited considerable interobserver variation among both more and less experienced reviewers. CLINICAL IMPACT. Requiring multiple targetoid appearances for LR-M categorization improved interobserver agreement and diagnostic accuracy for non-HCC malignancy.

The field of medical image analysis has been attracted to deep learning. Various deep learning-based techniques have been introduced to aid diagnosis in the CT image of the patient. The auxiliary model for diagnosis that we proposed is to detect colorectal tumors in the CT image. The model is combined with two contrary networks of 'Detection Transformer" and 'Hourglass". Furthermore., to improve the performance of the model., we propose an efficient connection method for two contrary models by using intermediate prediction information. A total of 3.,509 patients (193.,567 CT images) were applied to the experiment and our model outperforms the conventional models in colorectal tumor detection. Clinical Relevance - The proposed model in this paper automatically detects colorectal tumors and provides the bounding box in the CT images. Colorectal tumor is one of the common diseases. In addition, the mortality rate is so high that in-time treatment is required. The model we present here has a sensitivity (or recall) of 84.73 % for tumor detection and a precision of 88.25 % in the patient CT data. The in-slice performance of the tumor detection shows an IoU of 0.56, a sensitivity of 0.67, and a precision of 0.68.

Purpose: To construct a standard dataset of contrast-enhanced CT images of liver tumors to test the performance and safety of artificial intelligence (AI)-based algorithms for clinical decision support systems (CDSSs)., Materials and Methods: A consensus group of medical experts in gastrointestinal radiology from four national tertiary institutions discussed the conditions to be included in a standard dataset. Seventy-five cases of hepatocellular carcinoma, 75 cases of metastasis, and 30-50 cases of benign lesions were retrieved from each institution, and the final dataset consisted of 300 cases of hepatocellular carcinoma, 300 cases of metastasis, and 183 cases of benign lesions. Only pathologically confirmed cases of hepatocellular carcinomas and metastases were enrolled. The medical experts retrieved the medical records of the patients and manually labeled the CT images. The CT images were saved as Digital Imaging and Communications in Medicine (DICOM) files., Results: The medical experts in gastrointestinal radiology constructed the standard dataset of contrast-enhanced CT images for 783 cases of liver tumors. The performance and safety of the AI algorithm can be evaluated by calculating the sensitivity and specificity for detecting and characterizing the lesions., Conclusion: The constructed standard dataset can be utilized for evaluating the machine-learningbased AI algorithm for CDSS., Competing Interests: Conflicts of Interest: The authors have no potential conflicts of interest to disclose., (Copyrights © 2021 The Korean Society of Radiology.)