Imaging Reporting Research Articles

Comparison of AI software tools for automated detection, quantification and categorization of pulmonary nodules in the HANSE LCS trial

AbstractParticipant management in a lung cancer screening (LCS) depends on the assigned Lung Imaging Reporting and Data System (Lung-RADS) category, which is based on reliable detection and measurement of pulmonary nodules. The aim of this study was to compare the agreement of two AI-based software tools for detection, quantification and categorization of pulmonary nodules in an LCS program in Northern Germany (HANSE-trial). 946 low-dose baseline CT-examinations were analyzed by two AI software tools regarding lung nodule detection, quantification and categorization and compared to the final radiologist read. The relationship between detected nodule volumes by both software tools was assessed by Pearson correlation (r) and tested for significance using Wilcoxon signed-rank test. The consistency of Lung-RADS classifications between Software tool 1 (S1, Aview v2.5, Coreline Soft, Seoul, Korea) and Software tool 2 (S2, Prototype ‘’ChestCTExplore’’, software version ToDo, Siemens Healthineers, Forchheim, Germany) was evaluated by Cohen’s kappa (κ) and percentual agreement (PA).The derived volumes of true positive nodules were strongly correlated (r > 0.95), the volume derived by S2 was significantly higher than by S1 (P < 0.0001, mean difference: 6mm3). Moderate PA (62%) between S1 and S2 was found in the assignment of Lung-RADS classification (κ = 0.45). The PA of Lung-RADS classification to final read was 75% and 55% for S1 and S2, but the incorporation of S1 into the initial nodule detection and segmentation must be considered here. Significant nodule volume differences between AI software tools lead to different Lung-RADS scores in 38% of cases, which may result in altered participant management. Therefore, high performance and agreement of accredited AI software tools are necessary for a future national LCS program.

Effectiveness of ultrasound‐guided vacuum‐assisted excision for treating benign breast lesions

AbstractBackgroundUltrasound‑guided vacuum‐assisted excision (UGVAE) and breast biopsy are widely used for the diagnosis and treatment of both benign and suspicious breast lesions. In this retrospective study, we aimed to determine the safety of UGVAE for benign breast lesions and provide guidance for clinical practice.MethodsWe analyzed clinical and pathological data of female patients who had undergone UGVAE between January 2015 and December 2017 at our institution. All breast lesions were categorized according to the Breast Imaging Reporting and Data System (BI‐RADS) before performing UGVAE.ResultsIn our study cohort, UGVAE was used to resect 10,378 breast lesions from 5789 patients, and selected clinical and histopathological data were analyzed. The most common adverse events were postoperative bleeding (0.24%) and skin hypersensitivity (0.67%). The residual lesion rate was 2.27%. Fibroadenomas accounted for most of the benign lesions (7932 of 10,193; 77.82%). Breast cancer was diagnosed in 150 lesions from 128 patients. Multivariable binary logistic regression analyses showed that older age (odds ratio [OR] = 2.034, 95% confidence interval [CI]: 1.668–2.480, p < 0.001), higher BI‐RADS category (OR = 9.514, 95% CI: 6.790–13.332, p < 0.001), and larger legion size (OR = 1.048, 95% CI: 1.019–1.077, p = 0.001) were associated with an increased likelihood of breast cancer. Ninety‐six patients with breast cancer had undergone follow‐up treatment, achieving a 3‐year disease‐free survival rate of 97.2% and a 3‐year overall survival rate of 100%.ConclusionsUGVAE is a safe and effective means of removing benign breast lesions, causing minimal postoperative trauma and fewer complications compared with open surgery. Moreover, UGVAE had little impact on the follow‐up treatment and survival of patients diagnosed with breast cancer.

Abstract 4138722: Large Language Model based multi-agent Transcatheter Aortic Valve Implantation team to augment multidisciplinary meetings - proof of concept.

Introduction: Multidisciplinary team (MDT) discussions are integral to Transcatheter Aortic Valve Implantation (TAVI) decision making. Large language model (LLM) ubiquity and low-code no-code platforms have enabled clinician lead solution development. Specialised chatbots or ‘agents’ have evolved into multi-agent systems that can personify human collaboration. We assess the performance of an artificial intelligence (AI) multi-agent TAVI MDT. Methods: Four de-identified TAVI cases from two metropolitan Australian hospitals were assessed by a mock human TAVI MDT (h-MDT) and an AI multi-agent TAVI MDT (ai-MDT). The ai-MDT was created with Agentflow within Flowise AITM and had a hierarchical multi-agent architecture suited to complex reasoning required for TAVI MDT simulation (figure). LLM limitations necessitated the ai-MDT rely on imaging reports rather than clinical images. The h-MDT and ai-MDT consisted of similar team members. Outputs from the h-MDT and ai-MDT was adjudicated by a panel of four blinded TAVI doctors that determined if output was human vs AI and assigned a SMIC score (4-12, 4=good, 12=poor) that assessed structure, missing information, incorrect information and clinical utility. Time durations for h-MDT and ai-MDT were recorded. Results: Adjudicators differentiated human vs. AI output 100% of the time and ai-MDT output had better SMIC scores than h-MDT (U-stat 213, p=0.0011). ai-MDT outperformed h-MDT in the domains of structure, missing information and clinical utility but was not statistically different in the incorrect information domain (U-stat 132, p=0.88). The average time for each case in h-MDT was 15 minutes and 45 seconds compared to 97 seconds for ai-MDT. Conclusion: This demonstrates the potential of using LLM based multi-agent systems as a clinical adjunct in highly specialized multidisciplinary clinical meetings. AI responses were superior for structure, clinical utility and missing information and non-inferior for incorrect information compared to humans, which highlights that hallucinations remain an issue with generative AI. Time was saved but image interpretation still requires human input, for now. Cognitive AI continues to require human supervision for implementation.

Shareable artificial intelligence to extract cancer outcomes from electronic health records for precision oncology research

Abstract Databases that link molecular data to clinical outcomes can inform precision cancer research into novel prognostic and predictive biomarkers. However, outside of clinical trials, cancer outcomes are typically recorded only in text form within electronic health records (EHRs). Artificial intelligence (AI) models have been trained to extract outcomes from individual EHRs. However, patient privacy restrictions have historically precluded dissemination of these models beyond the centers at which they were trained. In this study, the vulnerability of text classification models trained directly on protected health information to membership inference attacks is confirmed. A teacher-student distillation approach is applied to develop shareable models for annotating outcomes from imaging reports and medical oncologist notes. ‘Teacher’ models trained on EHR data from Dana-Farber Cancer Institute (DFCI) are used to label imaging reports and discharge summaries from the Medical Information Mart for Intensive Care (MIMIC)-IV dataset. ‘Student’ models are trained to use these MIMIC documents to predict the labels assigned by teacher models and sent to Memorial Sloan Kettering (MSK) for evaluation. The student models exhibit high discrimination across outcomes in both the DFCI and MSK test sets. Leveraging private labeling of public datasets to distill publishable clinical AI models from academic centers could facilitate deployment of machine learning to accelerate precision oncology research.

Documentation of incidentally noted hepatic steatosis to emergency department patients: A retrospective study

AbstractBackgroundHepatic steatosis is a common incidental finding on emergency department (ED) imaging studies, occurring in up to 10% of studies, and carries significant long‐term morbidity. Frequently considered an unimportant finding, it is unknown how often ED patients are informed of hepatic steatosis. Our objective was to examine hepatic steatosis inclusion in ED discharge materials.MethodsData from discharged patients at an urban academic ED (>90,000 visits) with abdominal imaging (computed tomography or ultrasound) from 2019 to 2022 were screened. Patients with radiology reports documenting hepatic steatosis were included. Two trained data abstractors analyzed discharge materials and coded disclosure of steatosis (present/absent). Data abstraction also noted how the finding was included (e.g., follow‐up instructions, provision of radiology report). Factors associated with patient disclosure were examined through regression models including age, race, ethnicity, insurance, number of imaging studies during encounter, type of imaging study, and inclusion of hepatic steatosis in the impression section of radiology report.ResultsOf 10,677 radiology reports, 1209 (11.3%) had documented hepatic steatosis. The mean (±SD) age was 47.1 (±14.1) years; 56.4% were female, 53.5% were White, and 30.7% Hispanic. Only 173 of 1209 patients (14.3%) received any discharge documentation of hepatic steatosis. In 65% of cases where disclosure occurred, the imaging report was pasted verbatim into discharge material. There were no significant differences in discharge documentation by demographic groups. Mention within the radiology report impression (rather than the report body alone) occurred in 73.8% of records and was the strongest predictor of disclosure to patients (adjusted odds ratio 2.18, 95% confidence interval 1.39–3.54).ConclusionsDespite the high prevalence of hepatic steatosis in radiology reports, rates of documentation in patient‐facing discharge materials are notably low, exposing a possible communication failure with consequences for diagnosis.

Abstract 4139198: A Systematic Approach to Prompting Large Language Models for Automated Feature Extraction from Cardiovascular Imaging Reports

Introduction: Cardiovascular radiology reports contain valuable diagnostic information linked to images, but the unstructured text format makes feature extraction difficult on a large scale. Large language models (LLMs) allow for feature extraction where string parsing alone is insufficient, but require careful prompting for accurate results. Hypothesis: We hypothesize that a systematic prompting approach using LLMs can expedite the extraction of features from unstructured text in transesophageal echocardiography (TEE) reports. Methods: The data consisted of 7106 intraoperative TEE reports, 600 of which were manually reviewed to obtain pre- and post-intervention ground truth values for left ventricular ejection fraction (LVEF), right ventricular systolic function (RVSF), and tricuspid regurgitation (TR). Reports are paired with an imaging study consisting of 50-200 clips. For each feature considered, 100 of the 600 labeled reports were used to engineer a prompt in Llama-2 that maximized feature extraction accuracy. Results: We found that using multiple, shorter prompts yielded higher accuracy than did fewer, longer prompts. Additionally, when imposing semantic information onto a numerical scale, prompt engineering in combination with string parsing (Figure 1) gave the best results. When evaluated on the 500 labeled reports withheld for testing, the finalized prompts had accuracies of 94.1%, 94.8%, and 91.3% for LVEF, RVSF, and TR, respectively. Using this strategy, 5000 intraoperative TEE reports were analyzed and used to train and evaluate a regression model for LVEF estimation from TEE clips (Figure 2). Conclusion: We have shown that performing prompt engineering on Llama-2 can be used to extract features from unstructured TEE reports in an accurate manner. As an extension of these methods, automated feature prediction from echocardiograms can be used to create rapid, low-cost, and accessible cardiac assessments.

Diagnostic Accuracy of Spectral Doppler Indices and Sonoelastography in Predicting Malignancy in Breast Imaging Reporting and Database System 3 Breast Lesions With Histopathology as the Reference Standard

Diagnostic Accuracy of Spectral Doppler Indices and Sonoelastography in Predicting Malignancy in Breast Imaging Reporting and Database System 3 Breast Lesions With Histopathology as the Reference Standard

A Machine Learning Approach for Breast Cancer Risk Prediction in Digital Mammography

Breast cancer is among the most prevalent cancers in the female population globally. Therefore, screening campaigns as well as approaches to identify patients at risk are particularly important for the early detection of suspect lesions. This study aims to propose a workflow for the automatic classification of patients based on one of the most relevant risk factors in breast cancer, which is represented by breast density. The proposed classification methodology takes advantage of the features automatically extracted from mammographic images, as digital mammography represents the major screening tool in women. Textural features were extracted from the breast parenchyma through a radiomics approach, and they were used to train different machine learning algorithms and neural network models to classify the breast density according to the standard Breast Imaging Reporting and Data System (BI-RADS) guidelines. Both binary and multiclass tasks have been carried out and compared in terms of performance metrics. Preliminary results show interesting classification accuracy (93.55% for the binary task and 82.14% for the multiclass task), which are promising compared to the current literature. As the proposed workflow relies on straightforward and computationally efficient algorithms, it could serve as a basis for a fast-track protocol for the screening of mammograms to reduce the radiologists’ workload.

Magnetic Resonance Elastography Combined With PI-RADS v2.1 for the Identification of Clinically Significant Prostate Cancer.

Multiparametric MRI may cause overdiagnosis of clinically significant prostate cancer (csPCa) with the Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1). To investigate the diagnostic performance of stiffness as a standalone and complementary marker to PI-RADS v2.1 for diagnosing csPCa. Prospective. One hundred forty-seven participants with pathologically confirmed prostate lesions (≥1 cm), including 71 with csPCa. T1-weighted fast spin-echo, T2-weighted fast spin-echo, single-shot echo-planar diffusion-weighted imaging, fast 3D gradient-echo T1-weighted dynamic contrast-enhanced imaging, and 3D single-shot spin-echo based echo-planar MR elastography at 3.0 T. The PI-RADS v2.1 score was assessed by three radiologists independently. Lesion shear stiffness (SS) values at 60 Hz and 90 Hz were measured. A modified PI-RADS integrating stiffness with PI-RADS v2.1 was developed. Diagnostic performance for csPCa was compared between stiffness, PI-RADS v2.1 and the modified PI-RADS. Spearman's correlation, Fleiss κ and intraclass correlation, Pearson correlation, one-way analysis of variance, area under the receiver operating characteristic curve (AUC), and the Delong test. Significance level was P < 0.05. In the peripheral zone, csPCa (N = 35) had significantly higher SS than non-csPCa at 60 Hz (3.22 ± 0.66 kPa vs. 2.56 ± 0.56 kPa) and at 90 Hz (5.64 ± 1.30 kPa vs. 4.48 ± 0.84 kPa). PI-RADS v2.1 showed 100% sensitivity, 58% specificity, and 0.79 AUC for detecting csPCa. SS achieved 97% sensitivity, 52% specificity, and 0.80 AUC at 60 Hz, while SS had 63% sensitivity, 87% specificity, and 0.78 AUC at 90 Hz. The modified PI-RADS, combing SS at 60 Hz with PI-RADS v2.1, resulted in a significantly increased AUC (0.86) compared to that of PI-RADS v2.1, with a sensitivity of 97% and specificity of 75%. Stiffness can help identifying csPCa in the peripheral zone. Combining stiffness with the PI-RADS v2.1 improved the diagnostic accuracy and specificity for csPCa. 1 TECHNICAL EFFICACY: Stage 2.

Enhancing the diagnostic capacity of [18F]PSMA-1007 PET/MRI in primary prostate cancer staging with artificial intelligence and semi-quantitative DCE: an exploratory study

BackgroundTo investigate the ability of artificial intelligence (AI)-based and semi-quantitative dynamic contrast enhanced (DCE) multiparametric MRI (mpMRI), performed within [18F]-PSMA-1007 PET/MRI, in differentiating benign from malignant prostate tissues in patients with primary prostate cancer (PC).ResultsA total of seven patients underwent whole-body [18F]-PSMA-1007 PET/MRI examinations including a pelvic mpMRI protocol with T2w, diffusion weighted imaging (DWI) and DCE image series. Conventional analysis included visual reading of PET/MRI images and Prostate Imaging Reporting & Data System (PI-RADS) scoring of the prostate. On the prostate level, we performed manual segmentations for time-intensity curve parameter formation and semi-quantitative analysis based on DCE segmentation data of PC-suspicious lesions. Moreover, we applied a recently introduced deep learning (DL) pipeline previously trained on 1010 independent MRI examinations with systematic biopsy-enhanced histopathological targeted biopsy lesion ground truth in order to perform AI-based lesion detection, prostate segmentation and derivation of a deep learning PI-RADS score. DICE coefficients between manual and automatic DL-acquired segmentations were compared. On patient-based analysis, PET/MRI revealed PC-suspicious lesions in the prostate gland in 6/7 patients (Gleason Score-GS ≥ 7b) that were histologically confirmed. Four of these patients also showed lymph node metastases, while two of them had bone metastases. One patient with GS 6 showed no PC-suspicious lesions. Based on DCE segmentations, a distinction between PC-suspicious and normal appearing tissue was feasible with the parameters fitted maximum contrast ratio (FMCR) and wash-in-slope. DICE coefficients (manual vs. deep learning) were comparable with literature values at a mean of 0.44. Further, the DL pipeline could identify the intraprostatic PC-suspicious lesions in all six patients with clinically significant PC.ConclusionFirstly, semi-quantitative DCE analysis based on manual segmentations of time-intensity curves was able to distinguish benign from malignant tissues. Moreover, DL analysis of the MRI data could detect clinically significant PC in all cases, demonstrating the feasibility of AI-supported approaches in increasing diagnostic certainty of PSMA-radioligand PET/MRI.

Assessment of Reason for Exam Imaging Reporting and Data System (RI-RADS) in inpatient diagnostic imaging referrals

ObjectivesTo test the Reason for Exam Imaging Reporting and Data System (RI-RADS) in assessing the quality of radiology requests in an Italian cohort of inpatients; to evaluate the interobserver reliability of RI-RADS.MethodsA single-center quality care study was designed to retrospectively identify consecutive radiology request forms for computed tomography, magnetic resonance imaging, and conventional radiography examinations. One radiologist scored the requests using the RI-RADS. The association between RI-RADS and clinical request variables (urgent request, on-call requests, indication for imaging, requesting specialty, imaging modality, and body region) was evaluated. We calculated interobserver agreement between four readers in a subset of 450 requests.ResultsWe included 762 imaging requests. RI-RADS grades A (adequate request), B (barely adequate request), C (considerably limited request), D (deficient request), and X were assigned to 8 (1%), 49 (7%), 237 (31%), 404 (53%), and 64 (8%) of cases, respectively. In the multivariate analysis, the indication for imaging, body region, and requesting specialty significantly influenced the RI-RADS. Indications for imaging with a high risk of poor RI-RADS grade were routine preoperative imaging and device check requests. The upper extremity was the body region with the highest risk of poor RI-RADS grade. Requesting specialties with a high risk of poor RI-RADS grade were cardiovascular surgery, intensive care medicine, and orthopedics. The analysis of the interobserver agreement revealed substantial agreement for the RI-RADS grade.ConclusionThe majority of radiology exam requests were inadequate according to RI-RADS, especially those for routine imaging. RI-RADS demonstrated substantial reliability, suggesting that it can be satisfactorily employed in clinical settings.Critical relevant statementThe implementation of RI-RADS can provide a framework for standardizing radiology requests, thereby enabling quality assurance and promoting a culture of quality improvement.Key PointsRI-RADS aims to grade the completeness of radiology requests.Over half of the imaging requests were RI-RADS D grade; RI-RADS demonstrated substantial reliability.Most radiology requests were inadequate and RI-RADS could classify them in clinical practice.Graphical

Image Features and Diagnostic Value of Contrast-Enhanced Ultrasound for Ductal Carcinoma In Situ of the Breast: Preliminary Findings.

To explore the image features and the diagnostic value of contrast-enhanced ultrasound (CEUS) for ductal carcinoma in situ (DCIS) of the breast. A total of 96 female patients with a solitary and histologically proven DCIS were analyzed retrospectively, and 100 female cases of invasive ductal carcinoma (IDC) lesions were used as the control group. The Breast Imaging Reporting and Data System (BI-RADS) category of breast lesions was assessed according to conventional ultrasound features. The DCIS lesions were classified into mass type and non-mass type. The CEUS characteristics of these breast lesions were retrospectively analyzed qualitatively and quantitatively. The final gold standard was biopsy or surgery with histo-pathological examination. Comparing the ultrasound images of DCIS with that of IDC, there were significant differences in echo pattern, calcification morphology, and calcification distribution (p < .05 for all). There was a significant difference between DCIS and IDC in enhancement intensity, perfusion defects, peripheral high enhancement, intratumoral vessels, and arrival time (AT) (p < .05 for all). In the logistic multivariate regression analysis, two indicators linked with DCIS were recognized: perfusion defects (p = .002) and peripheral high enhancement (p < .001). In forecasting DCIS, the logistic regression equation resulted in an AUC of 0.689, a specificity of 0.720, and a sensitivity of 0.563. CEUS showed differences in enhancement characteristics between DCIS and IDC, with perfusion defects and peripheral high enhancement being associated with DCIS.

Is There a Correlation Between the Location of the Clot in the Pulmonary Arterial Tree with the Location of the Mucus Plug in the Pulmonary Bronchial Tree in Patients with chronic obstructive pulmonary disease Experiencing Pulmonary Embolism? Novel Findings.

The study aimed to investigate the impact of mucus plugs on the localization of clots in pulmonary embolisms among chronic obstructive pulmonary disease (COPD) patients. The retrospective study examined 200 participants diagnosed with both COPD and pulmonary embolism. Of these, 100 patients exhibited mucus plugs in the segmental and subsegmental branches of the pulmonary bronchial tree, while the remaining 100 did not, using computed tomography images for diagnosis. Data collection encompassed a comprehensive review of patient records, including medical history and imaging reports, to determine the presence of mucus plugs and the localization of clots in pulmonary embolism cases. Patients with mucus plugs exhibited a notably longer duration of COPD (P = .021) and a higher mean pulmonary arterial occlusion index (23 vs. 12, P = .001). Moreover, the prevalence of clots in major pulmonary arteries was significantly elevated in the mucus plug group compared to the non-mucus plug group (P < .05). Conversely, patients without mucus plugs displayed a higher incidence of clots in segmental and subsegmental arteries (P < .001). Strong positive correlations existed between mucus plugs in segmental branches and clots in major pulmonary arteries, with moderate to strong correlation coefficients (0.51 to 0.62, P < .05). Additionally, strong negative correlations were observed between mucus plugs in segmental branches and clots in segmental and subsegmental arteries, with correlation coefficients (CC) ranging from -0.74 to -0.76 (P < .05). Similarly, a significant negative correlation was noted between mucus plugs in subsegmental branches and clots in subsegmental arteries (CC: -0.68 and -0.71, P < .05). The results suggest that mucus plugs may be associated with increased severity of COPD, higher pulmonary arterial occlusion index, and altered clot distribution within the pulmonary artery tree. These findings emphasize the importance of recognizing mucus plugs as a key consideration in COPD assessment and management, potentially influencing disease severity, vascular remodeling, and thrombotic risk management.

Retrospective analysis in in situ ductal carcinoma; 11 years of experience

Objective: To describe our experience in the diagnosis, treatment and follow up of patients with ductal carcinoma in situ (DCIS). Materials and methods: 953 breast carcinomas treated in the Mastology service of the Department of Gynecology and Obstetrics of the Hospital Aleman of Buenos Aires, between January 2004 and December 2014, are retrospectively analyzed. 206 biopsies (BRQ 25-mammotome 151-core biopsy 30) of ductal carcinomas in situ were identified, resulting after definitive pathological evaluation in 172 (18%) pure ductal carcinomas in situ of the total of 953 patients analyzed. Results: 206 biopsies (BRQ 25-mammotome 151-core biopsy 30) of ductal carcinomas in situ were identified, resulting after definitive pathological evaluation in 172 (18%) pure ductal carcinomas in situ of the total of 953 patients analyzed. The imaging report in the total of the 206 patients showed microcalcifications in 80.1%. The diagnosis of carcinoma in situ was made in 181 (87%) patients by preoperative microinvasive procedures and in the remaining 25 (13%) patients by radiosurgical biopsy (BRQ). There was evidence of 18.8% underdiagnosis after microinvasive procedures. In our case series, 84.3% were GH3/GH2 while 15.7% were GH 1. When comparing the size of the surgical specimens and correlating it with the 34 cases of invasive and microinvasive carcinoma, it showed that 54.8% of invasion in those tumors greater than 30 mm, 50.4% in those that exceeded 21 mm and in no case in those less than 10 mm. 20% of multicentric lesions were associated with invasive tumor. Sentinel lymph node technique was performed in 23.8% patients in the first surgery, resulting negative in all cases. When evaluating radiation and hormonal treatment, radiotherapy was performed in 131 patients (85.6%) and hormonal treatment was performed in 75% of the patients. In the follow-up until December 2014, 11 relapses (5.23%) were recorded. Conclusion: Our results are consistent with the international indexed literature in reference to diagnosis, treatment and recurrence rate pure of DCIS.

CT/MRI LI-RADS 2024 Update: Treatment Response Assessment

The CT/MRI Liver Imaging Reporting and Data System version 2024 Treatment Response Assessment (TRA) algorithm now has two independent cores with tailored TRA criteria after nonradiation or radiation-based liver-directed therapy and incorporates MRI ancillary features (diffusion restriction and mild to moderate T2 hyperintensity).

Prognostic Implications of MRI-assessed Intratumoral Fat in Hepatocellular Carcinoma: An Asian and European Cohort Study.

Background The clinicopathologic-radiologic and prognostic characteristics of intratumoral fat in hepatocellular carcinoma (HCC) are critical for personalized treatment but remain understudied. Purpose To investigate the clinicopathologic-radiologic associations and prognostic implications of MRI-assessed intratumoral fat in HCCs. Materials and Methods This retrospective cohort study included consecutive adult patients who underwent resection for solitary HCCs and preoperative contrast-enhanced MRI from two tertiary-care hospitals in East Asia (March 2011 to December 2021) and Western Europe (September 2012 to December 2019). MRI scans were independently evaluated by three radiologists at each hospital. Based on Liver Imaging Reporting and Data System (LI-RADS) version 2018, intratumoral fat was defined as "fat in mass more than adjacent liver," and the homogeneous subtype was defined as intratumoral fat "in absence of mosaic and nodule-in-nodule architecture." Recurrence-free survival (RFS) and overall survival (OS) were estimated using the Kaplan-Meier method and compared using the log-rank test. Cox regression analyses were conducted to identify factors associated with RFS and OS. Results A total of 933 patients were included in the Asian (n = 736; median age, 53 years [IQR, 45-62 years]; 626 male) and European (n = 207; median age, 64 years [IQR, 55-70 years]; 161 male) cohorts. MRI-assessed intratumoral fat was detected in 30% (215 of 726) and 31% (64 of 207) of patients in the Asian and European cohorts, respectively (P = .72). In both cohorts, the steatohepatitic subtype, nonperipheral washout, enhancing capsule, and mosaic architecture were more frequent in tumors with intratumoral fat (P value range, <.001 to .04). Intratumoral fat in general was not associated with RFS or OS in either cohort (P value range, .48-.97). However, in the Asian cohort, homogeneous intratumoral fat was associated with longer RFS (hazard ratio [HR], 0.60; P = .009) and OS (HR, 0.33; P = .008) in multivariable Cox regression analyses. Conclusion MRI-assessed intratumoral fat was more frequent in steatohepatitic HCCs and associated with nonperipheral washout, enhancing capsule, and mosaic architecture. Although intratumoral fat was generally nonprognostic, homogeneous intratumoral fat was associated with longer RFS and OS in the Asian cohort. Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Harmath in this issue.

Detecting Clinically Significant Prostate Cancer in PI-RADS 3 Lesions Using T2w-Derived Radiomics Feature Maps in 3T Prostate MRI

Prostate Imaging Reporting and Data System version 2.1 (PI-RADS) category 3 lesions are a challenge in the clinical workflow. A better detection of the infrequently occurring clinically significant prostate cancer (csPCa) in PI-RADS 3 lesions is an important objective. The purpose of this study was to evaluate if feature maps calculated from T2-weighted (T2w) 3 Tesla (3T) MRI can help detect csPCa in PI-RADS category 3 lesions. In-house biparametric 3T prostate MRI examinations acquired between January 2019 and June 2023 because of elevated prostate-specific antigen (PSA) levels were retrospectively screened. Inclusion criteria were a PI-RADS 3 lesion and available results of an ultrasound-guided targeted and systematic biopsy. Exclusion criteria were a simultaneous PI-RADS category 4 or 5 lesion and hip replacement. Target lesions with the International Society of Urological Pathology (ISUP) grade group 1 were rated clinically insignificant PCa (ciPCa) and ≥2 csPCa. This resulted in 52 patients being included in the final analysis, of whom 11 (21.1%), 8 (15.4%), and 33 (63.5%) patients had csPCa, ciPCa, and no PCa, respectively, with the latter two groups being combined as non-csPCa. Eight of the csPCas were located in the peripheral zone (PZ) and three in the transition zone (TZ). In the non-csPCa group, 29 were located in the PZ and 12 in the TZ. Target lesions were marked with volumes of interest (VOIs) on axial T2w images. Axial T2w images were then converted to 93 feature maps. VOIs were copied into the maps, and feature quantity was retrieved directly. Features were tested for significant differences with the Mann–Whitney U-test. Univariate models for single feature performance and bivariate models implementing PSA density (PSAD) were calculated. Ten map-derived features differed significantly between the csPCa and non-csPCa groups (AUCs: 0.70–0.84). The diagnostic performance for TZ lesions (AUC: 0.83–1.00) was superior to PZ lesions (AUC: 0.74–0.85). In the bivariate models, performance in the PZ improved with AUCs &gt;0.90 throughout. Parametric feature maps alone and as bivariate models with PSAD can (?) noninvasively identify csPCa in PI-RADS 3 lesions and could serve as a quantitative tool reducing ambiguity in PI-RADS 3 lesions.

Optimal timing of anticoagulation after acute ischaemic stroke with atrial fibrillation (OPTIMAS): a multicentre, blinded-endpoint, phase 4, randomised controlled trial

The optimal timing of anticoagulation for patients with acute ischaemic stoke with atrial fibrillation is uncertain. We investigated the efficacy and safety of early compared with delayed initiation of direct oral anticoagulants (DOACs) in patients with acute ischaemic stroke associated with atrial fibrillation. We performed a multicentre, open-label, blinded-endpoint, parallel-group, phase 4, randomised controlled trial at 100 UK hospitals. Adults with atrial fibrillation and a clinical diagnosis of acute ischaemic stroke and whose physician was uncertain of the optimal timing for DOAC initiation were eligible for inclusion in the study. We randomly assigned participants (1:1) to early (ie, ≤4 days from stroke symptom onset) or delayed (ie, 7-14 days) anticoagulation initiation with any DOAC, using an independent online randomisation service with random permuted blocks and varying block length, stratified by stroke severity at randomisation. Participants and treating clinicians were not masked to treatment assignment, but all outcomes were adjudicated by a masked independent external adjudication committee using all available clinical records, brain imaging reports, and source images. The primary outcome was a composite of recurrent ischaemic stroke, symptomatic intracranial haemorrhage, unclassifiable stroke, or systemic embolism incidence at 90 days in a modified intention-to-treat population. We used a gatekeeper approach by sequentially testing for a non-inferiority margin of 2 percentage points, followed by testing for superiority. OPTIMAS is registered with ISRCTN (ISRCTN17896007) and ClinicalTrials.gov (NCT03759938), and the trial is ongoing. Between July 5, 2019, and Jan 31, 2024, 3648 patients were randomly assigned to early or delayed DOAC initiation. 27 participants did not fulfil the eligibility criteria or withdrew consent to include their data, leaving 3621 patients (1814 in the early group and 1807 in the delayed group; 1981 men and 1640 women) in the modified intention-to-treat analysis. The primary outcome occurred in 59 (3·3%) of 1814 participants in the early DOAC initiation group compared with 59 (3·3%) of 1807 participants in the delayed DOAC initiation group (adjusted risk difference [RD] 0·000, 95% CI -0·011 to 0·012). The upper limit of the 95% CI for the adjusted RD was less than the non-inferiority margin of 2 percentage points (pnon-inferiority=0·0003). Superiority was not identified (psuperiority=0·96). Symptomatic intracranial haemorrhage occurred in 11 (0·6%) participants allocated to the early DOAC initiation group compared with 12 (0·7%) participants allocated to the delayed DOAC initiation group (adjusted RD 0·001, -0·004 to 0·006; p=0·78). Early DOAC initiation within 4 days after ischaemic stroke associated with atrial fibrillation was non-inferior to delayed initiation for the composite outcome of ischaemic stroke, intracranial haemorrhage, unclassifiable stroke, or systemic embolism at 90 days. Our findings do not support the current common and guideline-supported practice of delaying DOAC initiation after ischaemic stroke with atrial fibrillation. British Heart Foundation.

A Risk Stratification Study of Ultrasound Images of Thyroid Nodules Based on Improved DETR

ABSTRACTThe Chinese Thyroid Imaging Reporting and Data System (C‐TIRADS) standard is based on the Chinese current medical context. However, at present, there is a lack of C‐TIRADS‐based automatic computer‐aided diagnosis system for thyroid nodule ultrasound images, and the existing algorithms for detecting and recognizing thyroid nodules are basically for the dichotomous classification of benign and malignant. We used the DETR (detection transformer) model as a baseline model and carried out model enhancements to address the shortcomings of unsatisfactory classification accuracy and difficulty in detecting small thyroid nodules in the DETR model. First, to investigate the method of acquiring multi‐scale features of thyroid nodule ultrasound images, we choose TResNet‐L as the feature extraction network and introduce multi‐scale feature information and group convolution, thereby enhancing the model's multi‐label classification accuracy. Second, a parallel decoder architecture for multi‐label thyroid nodule ultrasound image classification is designed to enhance the learning of correlation between pathological feature class labels, aiming to improve the multi‐label classification accuracy of the detection model. Third, the loss function of the detection model is improved. We propose a linear combination of Smooth L1‐Loss and CIoU Loss as the model's bounding box loss function and asymmetric loss as the model's multi‐label classification loss function, aiming to further improve the detection model's detection accuracy for small thyroid nodules. The experiment results show that the improved DETR model achieves an AP of 92.4% and 81.6% with IoU thresholds of 0.5 and 0.75, respectively.

Comparison of British Thyroid Association and TIRADS classifications and their impact on the radiological and surgical management of indeterminate thyroid nodules

BackgroundThe British Thyroid Association (BTA) Guidelines for the Management of Thyroid Cancer advocate for fine-needle aspiration cytology for all thyroid nodules graded indeterminate (U3) at ultrasound assessment. This approach raises concerns regarding potential over-diagnosis of low-risk lesions. Conversely, equivalent Thyroid Imaging Reporting and Data Systems (TIRADS) guidelines permit surveillance or discharge of indeterminate thyroid nodules of certain sizes. This service analysis analyses how guideline choice impacts the fine-needle aspiration cytology rate and subsequent surgical management of indeterminate thyroid nodules. MethodsAll patients with an indeterminate (U3) thyroid nodule identified on ultrasound over a 12-month period were included. Indeterminate thyroid nodules were retrospectively rescored using three equivalent TIRADS classifications by three independent reviewers, blinded to the histopathology. Hypothetical differences in fine-needle aspiration cytology rates and surgical activity were then compared. ResultsNinety-six nodules were identified. Retrospective application of TIRADS guidelines resulted in a hypothetical 44.8-55.2% reduction in fine-needle aspiration cytology performed for indeterminate thyroid nodules compared to BTA. A statistically significant increase in rates of surgical activity for indeterminate thyroid nodules was observed between BTA guidance and all retrospectively applied TIRADS guidelines (p<0.001). Of four confirmed thyroid cancers, three would have been unanimously removed. ConclusionUnder BTA guidance, increased fine-needle aspiration cytology rates for indeterminate thyroid nodules resulted in significantly increased surgical activity in our cohort compared to retrospectively applied TIRADS guidelines.