Interpretable artificial intelligence-based app assists inexperienced radiologists in diagnosing biliary atresia from sonographic gallbladder images.

BACKGROUND: A previously trained deep learning-based smartphone app provides an artificial intelligence solution to help diagnose biliary atresia from sonographic gallbladder images, but it might be impractical to launch it in real clinical settings. This study aimed to redevelop a new model using original sonographic images and their derived smartphone photos and then test the new model's performance in assisting radiologists with different experiences to detect biliary atresia in real-world mimic settings. METHODS: A new model was first trained retrospectively using 3659 original sonographic gallbladder images and their derived 51,226 smartphone photos and tested on 11,410 external validation smartphone photos. Afterward, the new model was tested in 333 prospectively collected sonographic gallbladder videos from 207 infants by 14 inexperienced radiologists (9 juniors and 5 seniors) and 4 experienced pediatric radiologists in real-world mimic settings. Diagnostic performance was expressed as the area under the receiver operating characteristic curve (AUC). RESULTS: The new model outperformed the previously published model in diagnosing BA on the external validation set (AUC 0.924 vs 0.908, P = 0.004) with higher consistency (kappa value 0.708 vs 0.609). When tested in real-world mimic settings using 333 sonographic gallbladder videos, the new model performed comparable to experienced pediatric radiologists (average AUC 0.860 vs 0.876) and outperformed junior radiologists (average AUC 0.838 vs 0.773) and senior radiologists (average AUC 0.829 vs 0.749). Furthermore, the new model could aid both junior and senior radiologists to improve their diagnostic performances, with the average AUC increasing from 0.773 to 0.835 for junior radiologists and from 0.749 to 0.805 for senior radiologists. CONCLUSIONS: The interpretable app-based model showed robust and satisfactory performance in diagnosing biliary atresia, and it could aid radiologists with limited experiences to improve their diagnostic performances in real-world mimic settings.

Radiofrequency ablation induces tumor cell dissemination in a mouse model of hepatocellular carcinoma.

BACKGROUND: We tested the hypothesis that radiofrequency ablation (RFA) for hepatocellular carcinoma (HCC) promotes tumor cell release and explored a method for reducing these effects. METHODS: A green fluorescent protein-transfected orthotopic HCC model was established in 99 nude mice. In vivo flow cytometry was used to monitor circulating tumor cell (CTC) dynamics. Pulmonary fluorescence imaging and pathology were performed to investigate lung metastases. First, the kinetics of CTCs during the periablation period and the survival rate of CTCs released during RFA were investigated. Next, mice were allocated to controls, sham ablation, or RFA with/without hepatic vessel blocking (ligation of the portal triads) for evaluating the postablation CTC level, lung metastases, and survival over time. Moreover, the kinetics of CTCs, lung metastases, and mice survival were evaluated for RFA with/without ethanol injection. Pathological changes in tumors and surrounding parenchyma after ethanol injection were noted. Statistical analysis included t-test, ANOVA, and Kaplan-Meier survival curves. RESULTS: CTC counts were 12.3-fold increased during RFA, and 73.7% of RFA-induced CTCs were viable. Pre-RFA hepatic vessel blocking prevented the increase of peripheral CTCs, reduced the number of lung metastases, and prolonged survival (all p ≤ 0.05). Similarly, pre-RFA ethanol injection remarkably decreased CTC release during RFA and further decreased lung metastases with extended survival (all p ≤ 0.05). Histopathology revealed thrombus formation in blood vessels after ethanol injection, which may clog tumor cell dissemination during RFA. CONCLUSION: RFA induces viable tumor cell dissemination, and pre-RFA ethanol injection may provide a prophylactic strategy to reduce this underestimated effect. RELEVANCE STATEMENT: RFA for HCC promotes viable tumor cell release during ablation, while ethanol injection can prevent RFA induced tumor cell release. KEY POINTS: • RFA induced the release of viable tumor cells during the ablation procedure in an animal model. • Hepatic vessel blocking can suppress tumor cells dissemination during RFA. • Ethanol injection can prevent RFA-induced tumor cell release, presumably because of the formation of thrombosis.

A novel prognostic model based on AFP, tumor burden score and Albumin-Bilirubin grade for patients with hepatocellular carcinoma undergoing radiofrequency ablation.

PURPOSE: The aim of this study was to develop prognostic scores, including the tumor burden score (TBS) and albumin-bilirubin (ALBI) grade, for evaluating the outcomes of hepatocellular carcinoma (HCC) patients after radiofrequency ablation (RFA). MATERIALS AND METHODS: This retrospective study enrolled treatment-naïve HCC patients with BCLC 0-A who underwent RFA between January 2009 and December 2019. Regular follow-up was conducted after RFA to determine progression-free survival (PFS) and overall survival (OS). The patients were randomly allocated to the training or validation datasets in a 1:1 ratio. Preoperative prognostic scores were developed based on the results of multivariate analysis. The discriminatory ability of the scores was assessed using time-dependent AUC and compared with other models. RESULTS: Serum alpha-fetoprotein (AFP) level and TBS were identified as independent prognostic factors for PFS, while serum AFP, TBS, and ALBI were identified as independent prognostic factors for OS in HCC patients after RFA. The time-dependent AUCs of the AFP-TBS score for the 1-, 3-, and 5-year PFS were 0.651, 0.667, and 0.620, respectively, in the training set, and 0.657, 0.687, and 0.704, respectively, in the validation set. For the 1-, 3-, and 5-year OS, the time-dependent AUCs were 0.680, 0.712, and 0.666, respectively, in the training set, and 0.712, 0.706 and 0.726 in the validation set for the AFP-TBS-ALBI score (ATA). The C-indices and AIC demonstrated that the scores provided better clinical benefits compared to other models. CONCLUSION: The ATA/AT score, derived from clinical and objective laboratory variables, can assist in individually predicting the prognosis of HCC patients undergoing curative RFA.

Predicting Infectious Complications after Percutaneous Thermal Ablation of Liver Malignancies: A 12-year Single-Center Experience.

Background Infectious complications after percutaneous thermal ablation are seldom discussed, but better understanding of risk factors and early prediction is critical. Purpose To estimate the incidence of infectious complications after percutaneous thermal ablation of liver malignancies and to develop prediction models. Materials and Methods This single-center retrospective study reviewed the data of 3167 patients who underwent 7545 percutaneous US-guided thermal ablation procedures of liver malignancies between January 2010 and January 2022. All procedures with infectious complications were included as the case group. For each case, one treatment date-matched control subject without infection was randomly selected following a nested case-control design. Independent factors of overall and hepatobiliary infection were investigated with multivariable logistic regression. Results A total of 80 patients (median age, 59 years; IQR, 51-68 years; 64 men, 16 women) developed infectious complications after 80 ablation procedures; the incidence was 1.1% (80 of 7545 procedures). Of those with infection, 18% (14 of 80 patients) were severe, and 10% (eight of 80 patients) died as a result. Independent risk factors for overall infectious complication included prior biliary intervention (odds ratio [OR], 18.6; 95% CI: 4, 86; P < .001), prior transarterial chemoembolization (TACE) (OR, 2.4; 95% CI: 1.0, 5.8; P = .045), and the largest tumor size (OR, 1.9; 95% CI: 1.3, 2.8; P = .002); on this basis, subcapsular location was an additional risk factor of hepatobiliary infection. Prediction models for overall and hepatobiliary infection had an area under the receiver operating characteristics curve (AUC) of 0.77 and 0.82, respectively, both of which showed better AUC compared with the models, including prior biliary intervention alone (AUC = 0.63 and 0.65, respectively; P = .01 and P = .005, respectively). Conclusion Infectious complications after percutaneous thermal ablation of liver malignancies were uncommon but potentially fatal. Independent predictors were prior biliary intervention, prior transarterial chemoembolization, and the largest tumor size. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Ben-Arie and Sosna in this issue.

Ultrasound-Guided Human Islet Transplantation: Safety, Feasibility, and Efficacy Analysis.

RATIONALE AND OBJECTIVES: We aimed to analyze the safety, feasibility, and efficacy of human islet transplantation (IT) using ultrasound (US) throughout the entire procedure. MATERIALS AND METHODS: A total of 22 recipients (18 males; mean age 42.6 ± 17.5years) with 35 procedures were retrospective included. Under US guidance, percutaneous transhepatic portal catheterization was successfully performed through a right-sided transhepatic approach, and islets were infused into the main portal vein. Color Doppler and contrast-enhanced ultrasound were used to guide the procedure and monitor the complications. After infusion of the islet mass, the access track was embolized by embolic material. If hemorrhage persisted, US-guided radiofrequency ablation (RFA) was performed to stop bleeding. Factors that could affect the complication were analyzed. After transplantation, primary graft function was evaluated with a ß-score 1month after the last islet infusion. RESULTS: The technical success rates were 100% with a single puncture attempt. Six (17.1%) abdominal bleeding episodes were immediately stopped by US-guided RFA. No portal vein thrombosis were encountered. Dialysis (OR (Odd Ratio): 32.0; 95% CI: 1.561-656.054; and P = .025) was identified as a significant factor associated with bleeding. Primary graft function was optimal in eight patients (36.4%), suboptimal in 13 patients (59.1%), and poor in one patient (4.5%). CONCLUSION: In conclusion, whole-procedure US-guided IT is a safe, feasible, and effective method for diabetes. Complications are either self-limiting or manageable with noninvasive treatment.

Shear wave elastography using high-frequency linear probe for kidney transplant monitoring: A methodological study.

OBJECTIVES: To investigate the influencing factors of the image quality of shear wave elastography (SWE) performed using a high-frequency probe and its reproducibility for renal allografts. METHODS: A total of 211 patients with transplanted kidneys who underwent SWE examination performed using high-frequency or low-frequency probes were recruited for the study. The reproducibility of inter- and intraobserver agreements were analysed by using the intraclass correlation coefficient (ICC). According to the colour filling of the area of interest and imaging noise when conducting SWE, the image quality was classified as three grades: "good", "common", and "poor". A logistic regression was used to analyse the independent factors for SWE quality. RESULTS: In the comparative analysis, high frequency, transection measurement and middle pole were selected as the appropriate measurement methods. Regarding reproducibility, the ICCs) of the intra- and interobserver agreements were 0.85 and 0.77, respectively. Multivariate analysis indicated that only the skin allograft distance and kidney width were independent variables for SWE quality. In the subgroup analysis of the skin-allograft distance, the "good" and "common" rates of images decreased as the distance increased, but the CV (coefficients of variation) showed the opposite trend. The SWE quality of kidney width <5.4 cm was significantly better than that of kidney width ≥5.4 cm. CONCLUSIONS: High-frequency SWE can be used in the evaluation of transplanted kidneys due to its good repeatability and high successful measurement rate, but we should pay attention to the influence of the skin-allograft distance and kidney width on SWE quality.

Overexpression of TREM1 is Associated with the Immune-Suppressive Microenvironment and Unfavorable Prognosis in Pan-Cancer.

Background: Triggering receptors expressed by myeloid cells-1 (TREM1) is a receptor belonging to the immunoglobulin superfamily and plays an important role in pro-inflammation in acute and chronic inflammatory disorders. However, the understanding of the immunomodulatory roles of TREM1 in the tumor microenvironment remains incomplete. Methods: The expression patterns of TREM1 mRNA in tumors and adjacent normal tissues were compared by analyzing data obtained from the Genotype-Tissue Expression and The Cancer Genome Atlas datasets. Survival analysis was performed to determine the prognostic value of TREM1. Functional enrichment analysis was applied to decipher the discrepancy in biological processes between high- and low-TREM1 groups across various cancers. The correlation between TREM1 and immune cell infiltration determined by using multiple algorithms was evaluated with the Pearson method. Four independent immunotherapy cohorts were adopted to validate the role of TREM1 as a biomarker. Results: TREM1 was elevated in most cancers as verified with clinical samples. Overexpression of TREM1 was linked with undesirable prognosis in patients. Further analysis revealed that TREM1 was positively correlated with immune response, pro-tumor pathways, and myeloid cell infiltration, while being negatively correlated with CD8+ T cell (including infiltration level and biological processes). Concordantly, tumors with high TREM1 levels were more resistant to immunotherapy. Through connective map analysis, therapeutically potential compounds like tozasertib and TPCA-1 were identified, which can be used synergistically with immunotherapy to improve the poor prognosis of patients with high TREM1 levels. Conclusion: Through a systematic and comprehensive pan-cancer analysis, we demonstrated that overexpression of TREM1 in tumors correlated closely with unfavorable outcome, infiltration of immune-suppressive cells, and immune regulation, which highlights its potential use as a tumor prognostic biomarker and a novel target for immunotherapy.

Artificial nano-red blood cells nanoplatform with lysosomal escape capability for ultrasound imaging-guided on-demand pain management.

Postoperative pain management would benefit significantly from an anesthetic that could take effect in an on-demand manner. An ultrasound would be an appropriate tool for such nanoplatform because it is widely used in clinical settings for ultrasound-guided anesthesia. Herein, we report a nanoplatform for postoperative on-demand pain management that can effectively enhance their analgesic time while providing ultrasonic imaging. Levobupivacaine and perfluoropentane were put into dendritic mesoporous silica and covered with red blood cell membranes to make the pain relief last longer in living organisms. The generated nanoplatform with gas-producing capability is ultrasonic responsive and can finely escape from the lysosomal in cells under ultrasound irradiation, maximizing the anesthetic effect with minimal toxicity. Using an incision pain model in vivo, levobupivacaine's sustained and controlled release gives pain reduction for approximately 3 days straight. The duration of pain relief is over 20 times greater than with a single injection of free levobupivacaine. Effective pain management was reached in vivo, and the pain reduction was enhanced by repeated ultrasonic irradiation. There was no detectable systemic or tissue injury under either of the treatments. Thus, our results suggest that nanoplatform with lysosomal escape capability can provide a practical ultrasound imaging-guided on-demand pain management strategy. STATEMENT OF SIGNIFICANCE: On-demand pain management is essential to postoperative patients. However, the traditional on-demand pain management strategy is hampered by the limited tissue penetration depth of near-infrared stimuli and the lack of proper imaging guidance. The proposed research is significant because it provides a nanoplatform for deep penetrated ultrasound controlled pain management under clinical applicable ultrasound imaging guidance. Moreover, the nanoplatform with prolonged retention time and lysosomal escape capability can provide long-term pain alleviation. Therefore, our results suggest that nanoplatform with lysosomal escape capability can provide an effective strategy for ultrasound imaging-guided on-demand pain management.

Stereotactic body radiation therapy versus radiofrequency ablation in hepatocellular carcinoma: an up-date meta-analysis.

PURPOSE: Radiofrequency ablation (RFA) and stereotactic body radiotherapy (SBRT) are available locoregional curative treatments for nonsurgical Hepatocellular carcinoma (HCC) patients. We aimed to compare the clinical efficacy and safety of SBRT versus RFA for HCC. METHODS: A computerized bibliographic search was performed using PubMed, Embase, the Cochrane Library and Web of Science to identify comparative studies. The primary outcome was overall survival (OS), and the secondary outcomes were freedom from local progression (FFLP) and treatment-related complications. RESULTS: In total, there were 17 trials involving 22,180 patients. Patients receiving RFA showed significantly better 1-, 2- year OS (OR 0.69, 95% CI 0.50-0.96, P = 0.141,OR 0.69, 95% CI 0.53-0.89, P = 0.082), whereas SBRT resulted in significantly better 1-, 2-, 3- year FFLP (OR 2.19, 95% CI 1.44-3.34, P = 0.303; OR 1.57, 95% CI 1.12-2.19, P = 0.268; OR 2.22, 95% CI 1.70-2.90, P = 0.470). There were no significant differences for 3-, 5- year OS in both groups (OR 0.94, 95% CI 0.65-1.38, P = 0.001; OR 0.98, 95% CI 0.68-1.34, P = 0.016). The overall treatment-related complication rate did not differ significantly between the two treatment arms, while SBRT was significantly associated with Child-Pugh worsening. CONCLUSIONS: Though SBRT has excellent FFLP, RFA yields superior short-term survival for HCC. But the discrepancy between FFLP and OS requires further investigation.

Nomogram based on high-frequency shear wave elastography (SWE) to evaluate chronic changes after kidney transplantation.

OBJECTIVES: To construct a nomogram with high-frequency shear wave elastography (SWE) as a noninvasive method to accurately assess chronic changes in renal allografts. METHODS: A total of 191 renal transplantation patients (127 cases in the training group and 64 cases in the verification group) were included in this study. All patients received conventional ultrasound and high-frequency SWE examination, followed directly by biopsy the next day. The chronic changes were divided into mild, moderate, and severe. Multivariate logistic analyses were used to select significant variables, which were used to develop the nomogram. Nomogram models were assessed by receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA). RESULTS: The cutoff value of SWE in mild, moderate, and severe chronic changes was 18.9, 22.5, and 27.6 kPa, respectively. The areas under the curve (AUCs) of SWE in the differential diagnosis of mild and moderate to severe chronic changes and mild to moderate and severe chronic changes were 0.817 and 0.870, respectively. Multivariate analysis showed that time since transplantation, proteinuria, glomerular filtration rate, echogenicity, and SWE were independent diagnostic factors for moderate to severe chronic changes (all p < 0.05); thus, a nomogram was successfully developed. The AUCs of the nomogram in the training and validation groups were 0.905 and 0.938, respectively. The high agreement between the model predictions and the actual observations was confirmed by calibration plot and DCA. CONCLUSIONS: Based on SWE, the nomogram provided an insightful and applicable tool to evaluate chronic changes in renal allografts. KEY POINTS: • In kidney transplantation, compared with acute changes, chronic changes are significantly correlated with cortical stiffness. • SWE shows good performance in identifying mild to moderate and severe chronic changes, with an AUC of 0.870. • Time since transplantation, proteinuria, glomerular filtration rate, echogenicity, and SWE are independent diagnostic factors for moderate to severe chronic changes in renal allografts.

Intensity-adjustable pain management with prolonged duration based on phase-transitional nanoparticles-assisted ultrasound imaging-guided nerve blockade.

BACKGROUND: The lack of a satisfactory strategy for postoperative pain management significantly impairs the quality of life for many patients. However, existing nanoplatforms cannot provide a longer duration of nerve blockage with intensity-adjustable characteristics under imaging guidance for clinical applications. RESULTS: To overcome this challenge, we proposed a biocompatible nanoplatform that enables high-definition ultrasound imaging-guided, intensity-adjustable, and long-lasting analgesia in a postoperative pain management model in awake mice. The nanoplatform was constructed by incorporating perfluoropentane and levobupivacaine with red blood cell membranes decorated liposomes. The fabricated nanoplatform can achieve gas-producing and can finely escape from immune surveillance in vivo to maximize the anesthetic effect. The analgesia effect was assessed from both motor reactions and pain-related histological markers. The findings demonstrated that the duration of intensity-adjustable analgesia in our platform is more than 20 times longer than free levobupivacaine injection with pain relief for around 3 days straight. Moreover, the pain relief was strengthened by repeatable ultrasound irradiation to effectively manage postoperative pain in an intensity-adjustable manner. No apparent systemic and local tissue injury was detected under different treatments. CONCLUSION: Our results suggest that nanoplatform can provide an effective strategy for ultrasound imaging-guided intensity-adjustable pain management with prolonged analgesia duration and show considerable transformation prospects.

Circulating tumour cell counts and ultrasomics signature-based nomogram for preoperative prediction of early recurrence of hepatocellular carcinoma after radical treatment.

METHODS: Between December 2017 and December 2018, 153 HCC patients (134 males and 19 females; mean age, 56.0 ± 10.2 years; range, 28-78 years) treated with radical therapy were enrolled in our retrospective study and were divided into a training cohort (n = 107) and a validation cohort (n = 46). All patients underwent preoperative CTC tests and CEUS examinations before treatment. The ultrasomics signature was extracted and built from CEUS images. Univariate and multivariate logistic regression analyses were used to identify the significant variables related to ER, which were then combined to build a predictive nomogram. The performance of the nomogram was evaluated by its discrimination, calibration and clinical utility. The predictive model was further evaluated in the internal validation cohort. RESULTS: HBV DNA, serum AFP level, CTC status, tumour size and ultrasomics score were identified as independent predictors associated with ER (all p < 0.05). Multivariable logistic regression analysis showed that the CTC status (OR = 7.02 [95% CI, 2.07 to 28.38], p = 0.003) and ultrasomics score (OR = 148.65 [95% CI, 25.49 to 1741.72], p < 0.001) were independent risk factors for ER. The nomogram based on ultrasomics score, CTC status, serum AFP level and tumour size exhibited C-indexes of 0.933 (95% CI, 0.878 to 0.988) and 0.910 (95% CI, 0.765 to 1.055) in the training and validation cohorts, respectively, fitting well in calibration curves. Decision curve analysis further confirmed the clinical usefulness of the nomogram. CONCLUSION: The nomogram incorporating CTC, ultrasomics features and independent clinical risk factors achieved satisfactory preoperative prediction of ER in HCC patients after radical treatment. ADVANCES IN KNOWLEDGE: 1. CTC status and ultrasomics score were identified as independent predictors associated with ER of HCC after radical treatment. 2. The nomogram constructed by ultrasomics score generated by 17 ultrasomics features, combined with CTCs and independent clinical risk factors such as AFP and tumour size. 3. The nomogram exhibited satisfactory discriminative power, and could be clinically useful in the preoperative prediction of ER after radical treatment in HCC patients.

Preoperative prognostic value of alfa-fetoprotein density in patients with hepatocellular carcinoma undergoing radiofrequency ablation.

OBJECTIVES: To examine the prognostic value of preoperative alfa-fetoprotein (AFP) density and other clinical factors in patients undergoing percutaneous radiofrequency ablation (RFA) of hepatocellular carcinoma (HCC). METHODS: From January 2010 to December 2018, a total of 543 patients undergoing RFA for HCC meeting the Milan criteria were included at our institution. AFP density was calculated as absolute AFP pre-ablation divided by the total volume of all HCC lesions. The survival rates according to AFP density were estimated using the Kaplan-Meier method and compared using the log-rank test. Univariate and multivariate Cox proportional-hazards regression analyses were used to assess predictors of overall survival (OS) and progression-free survival (PFS). RESULTS: The Kaplan-Meier 1-, 3-, and 5-year OS rates were 98.8%, 88.5%, and 70.4%, respectively, for the low AFP density group, and 98.3%, 74.9%, and 49.4%, respectively, for the high AFP density group. The corresponding PFS rates were 78.9%, 56.7%, and 40.9% (low AFP density group), and 63.6%, 40.8%, and 27.5% (high AFP density group). High AFP density was associated with significantly reduced PFS and OS (both p < 0.001). Multivariate analysis suggested that AFP density was a predictor of OS and PFS. CONCLUSIONS: Serum AFP density may serve as a promising predictor of survival in patients with HCC undergoing RFA. High AFP density could identify patients who might be prone to recurrence or progression and need close surveillance.

High-Frequency US for BK Polyomavirus-associated Nephropathy after Kidney Transplant.

Background BK polyomavirus-associated nephropathy (BKPyVAN) is an important cause of chronic renal allograft dysfunction. However, US features indicative of BKPyVAN have not been fully evaluated. Purpose To assess the value of high-frequency US for the diagnosis of BKPyVAN in kidney transplant recipients. Materials and Methods In this prospective cohort study, participants who tested positive for BK viruria after kidney transplant from September 2019 to January 2021 were evaluated with high-frequency US 1 day before biopsy. Clinical characteristics and US features were compared between participants with and without BKPyVAN. Significant predictors associated with BKPyVAN were determined using logistic regression analyses. The area under the receiver operating characteristic curve (AUC) was used to evaluate diagnostic performance. Results A total of 105 participants who underwent kidney transplant (mean age, 38 years ± 11 [SD]; 63 men) were evaluated; 45 participants were diagnosed with BKPyVAN. Multivariable analysis demonstrated that eccentric hydronephrosis and subcapsular hypoechoic areas were independent factors for BKPyVAN. The AUC for predicting BKPyVAN according to subcapsular hypoechoic areas was 0.66 (95% CI: 0.55, 0.77), with a specificity of 92% (55 of 60 participants). The AUC of combined US (eccentric hydronephrosis plus subcapsular hypoechoic area) and clinical (urine BKPyV DNA load [BKPyV-DNA] plus BK viremia) features was 0.90, with a specificity of 92% (55 of 60 participants). Parenchymal hyperechoic and subcapsular hypoechoic areas were independent factors for differentiating BKPyVAN from transplant rejection. The pooled specificity of subcapsular hypoechoic areas was 96% (21 of 22 participants), with an AUC of 0.67 (95% CI: 0.54, 0.80). For the combination of US (parenchymal echogenicity plus subcapsular hypoechoic area) and clinical (urine BKPyV-DNA plus time since transplant) features, the AUC reached 0.92 and specificity was 82% (18 of 22 participants). Conclusion High-frequency US characteristics are valuable for diagnosing BK polyomavirus-associated nephropathy (BKPyVAN) and distinguishing BKPyVAN from rejection in kidney transplant recipients. © RSNA, 2022 Online supplemental material is available for this article.

Versatile multi-constrained planning for thermal ablation of large liver tumors.

The surgical planning of large hepatic tumor ablation remains a challenging task that relies on fulfilling multiple medical constraints, especially for the ablation based on configurations of multiple electrodes. The placement of the electrodes to completely ablate the tumor as well as their insertion trajectory to their final position have to be planned to cause as little damage to healthy anatomical structures as possible to allow a fast rehabilitation. In this paper, we present a novel, versatile approach for the computer-assisted planning of multi-electrode thermal ablation of large liver tumors based on pre-operative CT data with semantic annotations. This involves both the specification of the number of required electrodes and their distribution to adequately ablate the tumor region without damaging too much healthy tissue. To determine the insertion trajectory of the electrodes to their final position, we additionally incorporate a series of medical constraints into our optimization, which allows a global analysis where obstacles such as bones are taken into account and damage to healthy tissue is mitigated. Compared with the state-of-the-art method, our method achieves compact ablation regions without relying on assumptions on a potential needle path for optimal global search and, hence, is suitable for guiding clinicians through the planning of the tumor ablation. We also demonstrate the feasibility of our approach in various experiments of clinical data and demonstrate that our approach not only allows completely ablating the tumor region but also reducing the damage of healthy tissue in comparison to the previous state-of-the-art method.

Machine Learning-Based Ultrasomics Improves the Diagnostic Performance in Differentiating Focal Nodular Hyperplasia and Atypical Hepatocellular Carcinoma.

BACKGROUND: The typical enhancement patterns of hepatocellular carcinoma (HCC) on contrast-enhanced ultrasound (CEUS) are hyper-enhanced in the arterial phase and washed out during the portal venous and late phases. However, atypical variations make a differential diagnosis both challenging and crucial. We aimed to investigate whether machine learning-based ultrasonic signatures derived from CEUS images could improve the diagnostic performance in differentiating focal nodular hyperplasia (FNH) and atypical hepatocellular carcinoma (aHCC). PATIENTS AND METHODS: A total of 226 focal liver lesions, including 107 aHCC and 119 FNH lesions, examined by CEUS were reviewed retrospectively. For machine learning-based ultrasomics, 3,132 features were extracted from the images of the baseline, arterial, and portal phases. An ultrasomics signature was generated by a machine learning model. The predictive model was constructed using the support vector machine method trained with the following groups: ultrasomics features, radiologist's score, and combination of ultrasomics features and radiologist's score. The diagnostic performance was explored using the area under the receiver operating characteristic curve (AUC). RESULTS: A total of 14 ultrasomics features were chosen to build an ultrasomics model, and they presented good performance in differentiating FNH and aHCC with an AUC of 0.86 (95% confidence interval [CI]: 0.80, 0.89), a sensitivity of 76.6% (95% CI: 67.5%, 84.3%), and a specificity of 80.5% (95% CI: 70.6%, 85.9%). The model trained with a combination of ultrasomics features and the radiologist's score achieved a significantly higher AUC (0.93, 95% CI: 0.89, 0.96) than that trained with the radiologist's score (AUC: 0.84, 95% CI: 0.79, 0.89, P < 0.001). For the sub-group of HCC with normal AFP value, the model trained with a combination of ultrasomics features, and the radiologist's score remain achieved the highest AUC of 0.92 (95% CI: 0.87, 0.96) compared to that with the ultrasomics features (AUC: 0.86, 95% CI: 0.74, 0.89, P < 0.001) and radiologist's score (AUC: 0.86, 95% CI: 0.79, 0.91, P < 0.001). CONCLUSIONS: Machine learning-based ultrasomics performs as well as the staff radiologist in predicting the differential diagnosis of FNH and aHCC. Incorporating an ultrasomics signature into the radiologist's score improves the diagnostic performance in differentiating FNH and aHCC.

Inter-reader agreement of CEUS LI-RADS among radiologists with different levels of experience.

OBJECTIVES: To investigate the inter-reader agreement of contrast-enhanced ultrasound (CEUS) of Liver Imaging Reporting and Data System version 2017 (LI-RADS v2017) categories among radiologists with different levels of experience. MATERIALS AND METHODS: From January 2014 to December 2014, a total of 326 patients at high risk of hepatocellular carcinoma (HCC) who underwent CEUS were included in this retrospective study. All lesions were classified according to LI-RADS v2017 by six radiologists with different levels of experiences: two residents, two fellows, and two specialists. Kappa coefficient was used to assess consistency of LI-RADS categories and major features among radiologists with different levels of experience. The diagnostic performance of HCC was described by accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC). RESULTS: Inter-reader agreement among radiologists of different experience levels was substantial agreement for arterial phase hyperenhancement, washout appearance, and early or late washout. Inter-reader agreement for LI-RADS categories was moderate to substantial. When LR-5 was used as criteria to determinate HCC, the AUC of LI-RADS for HCC was 0.67 for residents, 0.72 for fellows, and 0.78 for specialist radiologists. When compared between residents and specialists, accuracy, sensitivity, and AUC were significantly different (all p < 0.05). However, there were no significant differences in specificity, PPV, and NPV between the two groups. CONCLUSION: CEUS LI-RADS showed good diagnostic consistency among radiologists with different levels of experience, and consistency increased with experience levels. KEY POINTS: • The inter-reader agreement for LI-RADS categories was moderate to substantial agreement (κ, 0.60-0.80). • When compared between residents and specialists, accuracy, sensitivity, and AUC showed significantly different (all p < 0.05). However, there were no significant differences for specificity, PPV, and NPV between these two groups. • Among the radiologists with more than 1 year of experience, there was no significant difference in the diagnostic performance of HCC, suggesting that CEUS LI-RADS is a good standardized categorization system for high-risk patients.

Percutaneous thermal ablation of hepatic tumors: local control efficacy and risk factors for artificial ascites failure.

OBJECTIVE: To investigate the risk factors affecting the technical failure of artificial ascites (AA) formation and to evaluate the local control efficacy of percutaneous thermal ablation assisted by the AA for hepatic tumors. METHODS: A total of 341 patients with 362 hepatic tumors who underwent thermal ablation assisted by AA were reviewed retrospectively. The technical success of AA, the volume of liquid, and local efficacy after ablation were assessed. Predictive factors for the technical failure of AA formation and local tumor progression (LTP) were analyzed using univariate and multivariate analysis. RESULTS: The technical success rate of AA formation was 81.8% (296/362). The amount of fluid was higher when the tumor was located in the left lobe of the liver than when it was located in the right lobe (median 950 ml versus 700 ml, p < 0.001). Previous hepatic resection (OR: 12.63, 95% CI: 2.93-54.45, p < 0.001), ablation (OR: 6.48, 95% CI: 1.36-30.92, p = 0.019) and upper-abdomen surgery (OR: 11.34, 95% CI: 1.96-65.67, p = 0.007) were the independent risk factors of AA failure. In the AA success group, the complete ablation rate was higher and the LTP rate was lower than that in the AA failure group (98.7 versus 92.4%, p = 0.012; 8.8 versus 21.2%, p = 0.004). Multivariate analysis identified AA failure (p = 0.004), tumor size (>3.0 cm) (p = 0.002) and metastatic liver tumor (p = 0.008) as independent risk factors for LTP. CONCLUSION: History of hepatic resection, ablation and upper abdomen surgery were significant predictive factors affecting the technical failure of AA formation. Successful introduction of AA before thermal ablation can achieve better local tumor control efficacy.

Treatment effect of radiofrequency ablation versus liver transplantation and surgical resection for hepatocellular carcinoma within Milan criteria: a population-based study.

OBJECTIVES: Restricted mean survival time (RMST) has been increasingly used to assess the treatment effect. We aimed to evaluate a treatment effect of radiofrequency ablation (RFA) versus liver transplantation (LT) and surgical resection (SR) for hepatocellular carcinoma (HCC) within Milan criteria by using an adjusted RMST. METHODS: A total of 7,218 HCC patients (RFA, 3,327; LT, 2,332; SR 1,523) within Milan criteria were eligible for this retrospectively study. The RMST using inverse probability of treatment weighting (IPTW) adjustment were applied to estimate the treatment effect between RFA and LT, RFA, and SR groups. RESULTS: The 3-, 5-, and 10-year IPTW-adjusted difference in RMST of OS for LT over RFA were + 4.5, + 12.4, and + 36.3 months, respectively. For SR versus RFA group, the survival benefit was + 2.3, + 6.1, and + 15.8 months at 3, 5, and 10 years, respectively. But the incremental survival benefit of SR over RFA was only half than that of LT over RFA. In the subgroup of solitary tumor ≤ 2 cm, the adjusted RMST of RFA versus SR was comparable with no statistical differences. Beyond that, in comparison with RFA, a notably greater efficacy of LT and SR was consistently across all subgroups with solitary HCC > 2.0 cm, AFP positive or negative, and fibrosis score 0-4 or 5-6. CONCLUSIONS: RMST provides a measure of absolute survival benefit at a specific time point. Using IPTW-adjusted RMST, we showed that the incremental survival benefit of SR over RFA was about half than that of LT over RFA. KEY POINTS: • The restricted mean survival time offers an intuitive, clinically meaningful interpretation to quantify the treatment effect than the hazard ratio. • Liver transplantation and surgical resection provided better overall survival compared to radiofrequency ablation for HCC patients within Milan criteria, but RFA and SR provide equivalent long-term overall survival for solitary HCC ≤ 2 cm. • The incremental survival benefit of surgical resection over radiofrequency ablation was only half than that of liver transplantation over radiofrequency ablation.