Outcomes of Y90 Radioembolization for Hepatocellular Carcinoma in Patients Previously Treated with Transarterial Embolization.

The aim of this study was to evaluate outcomes of transarterial radioembolization (TARE) for hepatocellular carcinoma (HCC) in patients previously treated with transarterial embolization (TAE). In this retrospective study, all HCC patients who received TARE from 1/2012 to 12/2022 for treatment of residual or recurrent disease after TAE were identified. Overall survival (OS) was estimated using the Kaplan-Meier method. Univariate Cox regression was performed to determine significant predictors of OS after TARE. Twenty-one patients (median age 73.4 years, 18 male, 3 female) were included. Median dose to the perfused liver volume was 121 Gy (112-444, range), and 18/21 (85.7%) patients received 112-140 Gy. Median OS from time of HCC diagnosis was 32.9 months (19.4-61.4, 95% CI). Median OS after first TAE was 29.3 months (15.3-58.9, 95% CI). Median OS after first TARE was 10.6 months (6.8-27.0, 95% CI). ECOG performance status of 0 (p = 0.038), index tumor diameter < 4 cm (p = 0.022), and hepatic tumor burden < 25% (p = 0.018) were significant predictors of longer OS after TARE. TARE may provide a survival benefit for appropriately selected patients with HCC who have been previously treated with TAE.

Percutaneous Lung Biopsies With Robotic Systems: A Systematic Review of Available Clinical Solutions.

Objectives: This systematic review aims to assess existing research concerning the use of robotic systems to execute percutaneous lung biopsy. Methods: A systematic review was performed and identified 4 studies involving robotic systems used for lung biopsy. Outcomes assessed were operation time, radiation dose to patients and operators, technical success rate, diagnostic yield, and complication rate. Results: One hundred and thirteen robot-guided percutaneous lung biopsies were included. Technical success and diagnostic yield were close to 100%, comparable to manual procedures. Technical accuracy, illustrated by needle positioning, showed less frequent needle adjustments in robotic guidance than in manual guidance (P < .001): 2.7 ± 2.6 (range 1-4) versus 6 ± 4 (range 2-12). Procedure time ranged from comparable to reduced by 35% on average (20.1 ± 11.3 minutes vs 31.4 ± 10.2 minutes, P = .001) compared to manual procedures. Patient irradiation ranged from comparable to reduced by an average of 40% (324 ± 114.5 mGy vs 541.2 ± 446.8 mGy, P = .001). There was no significant difference in reported complications between manual biopsy and biopsies that utilized robotic guidance. Conclusion: Robotic systems demonstrate promising results for percutaneous lung biopsy. These devices provide adequate accuracy in probe placement and could both reduce procedural duration and mitigate radiation exposure to patients and practitioners. However, this review underscores the need for larger, controlled trials to validate and extend these findings.

Comparison of a Patient-Mounted Needle-Driving Robotic System versus Single-Rotation CT Fluoroscopy to Perform CT-Guided Percutaneous Lung Biopsies.

PURPOSE: To compare the effectiveness of percutaneous lung biopsy using a patient-mounted needle-driving robotic system with that using a manual insertion of needles under computed tomography (CT) fluoroscopy guidance. MATERIALS AND METHODS: In this institutional review board approved study, the cohort consisted of a series of patients who underwent lung biopsies following the intention-to-treat protocol from September 2022 to September 2023 using robot (n = 15) or manual insertion under single-rotation CT fluoroscopy (n = 66). Patient and procedure characteristics were recorded as well as outcomes. RESULTS: Although age, body mass index, and skin-to-target distance were not statistically different, target size varied (median, 8 mm [interquartile range, 6.5-9.5 mm] for robot vs 12 mm [8-18 mm] for single-rotation CT fluoroscopy; P = .001). No statistical differences were observed in technical success (86.7% [13/15] vs 89.4% [59/66], P = .673), Grade 3 adverse event (AE) (6.7% [1/15] vs 12.1% [8/66], P = .298), procedural time (28 minutes [22-32 minutes] vs 19 minutes [14.3-30.5 minutes], P = .086), and patient radiation dose (3.9 mSv [3.2-5.6 mSv] vs 4.6 mSv [3.3-7.5 mSv], P = .398). In robot-assisted cases, the median angle out of gantry plane was 10° (6.5°-16°), although it was null (0°-5°) for single-rotation CT fluoroscopy (P = .001). CONCLUSIONS: Robot-assisted and single-rotation CT fluoroscopy-guided percutaneous lung biopsies were similar in terms of technical success, diagnostic yield, procedural time, AEs, and radiation dose, although robot allowed for out-of-gantry plane navigation along the needle axis.

Early post-treatment MRI predicts long-term hepatocellular carcinoma response to radiation segmentectomy.

OBJECTIVES: Radiation segmentectomy using yttrium-90 plays an emerging role in the management of early-stage HCC. However, the value of early post-treatment MRI for response assessment is uncertain. We assessed the value of response criteria obtained early after radiation segmentectomy in predicting long-term response in patients with HCC. MATERIALS AND METHODS: Patients with HCC who underwent contrast-enhanced MRI before, early, and 12 months after radiation segmentectomy were included in this retrospective single-center study. Three independent radiologists reviewed images at baseline and 1st follow-up after radiation segmentectomy and assessed lesion-based response according to mRECIST, LI-RADS treatment response algorithm (TRA), and image subtraction. The endpoint was response at 12 months based on consensus readout of two separate radiologists. Diagnostic accuracy for predicting complete response (CR) at 12 months based on the 1st post-treatment MRI was calculated. RESULTS: Eighty patients (M/F 60/20, mean age 67.7 years) with 80 HCCs were assessed (median size baseline, 1.8 cm [IQR, 1.4-2.9 cm]). At 12 months, 74 patients were classified as CR (92.5%), 5 as partial response (6.3%), and 1 as progressive disease (1.2%). Diagnostic accuracy for predicting CR was fair to good for all readers with excellent positive predictive value (PPV): mRECIST (range between 3 readers, accuracy: 0.763-0.825, PPV: 0.966-1), LI-RADS TRA (accuracy: 0.700-0.825, PPV: 0.983-1), and subtraction (accuracy: 0.775-0.825, PPV: 0.967-1), with no difference in accuracy between criteria (p range 0.053 to > 0.9). CONCLUSION: mRECIST, LI-RADS TRA, and subtraction obtained on early post-treatment MRI show similar performance for predicting long-term response in patients with HCC treated with radiation segmentectomy. CLINICAL RELEVANCE STATEMENT: Response assessment extracted from early post-treatment MRI after radiation segmentectomy predicts complete response in patients with HCC with high PPV (≥ 0.96). KEY POINTS: • Early post-treatment response assessment on MRI predicts response in patients with HCC treated with radiation segmentectomy with fair to good accuracy and excellent positive predictive value. • There was no difference in diagnostic accuracy between mRECIST, LI-RADS, and subtraction for predicting HCC response to radiation segmentectomy.

Virtual and Augmented Reality in Interventional Radiology: Current Applications, Challenges, and Future Directions.

Virtual reality (VR) and augmented Reality (AR) are emerging technologies with the potential to revolutionize Interventional radiology (IR). These innovations offer advantages in patient care, interventional planning, and educational training by improving the visualization and navigation of medical images. Despite progress, several challenges hinder their widespread adoption, including limitations in navigation systems, cost, clinical acceptance, and technical constraints of AR/VR equipment. However, ongoing research holds promise with recent advancements such as shape-sensing needles and improved organ deformation modeling. The development of deep learning techniques, particularly for medical imaging segmentation, presents a promising avenue to address existing accuracy and precision issues. Future applications of AR/VR in IR include simulation-based training, preprocedural planning, intraprocedural guidance, and increased patient engagement. As these technologies advance, they are expected to facilitate telemedicine, enhance operational efficiency, and improve patient outcomes, marking a new frontier in interventional radiology.

Safety and Efficacy of Hepatic Artery Embolization in Heavily Treated Patients with Intrahepatic Cholangiocarcinoma: Analysis of Clinicopathological and Radiographic Parameters Associated with Better Overall Survival.

The safety and efficacy of hepatic artery embolization (HAE) in treating intrahepatic cholangiocarcinoma (IHC) was evaluated. Initial treatment response, local tumor progression-free survival (L-PFS), and overall survival (OS) were evaluated in 34 IHC patients treated with HAE. A univariate survival analysis and a multivariate Cox proportional hazard analysis to identify independent factors were carried out. Objective response (OR) at 1-month was 79.4%. Median OS and L-PFS from the time of HAE was 13 (CI = 95%, 7.4-18.5) and 4 months (CI = 95%, 2.09-5.9), respectively. Tumor burden < 25% and increased tumor vascularity on preprocedure imaging and surgical resection prior to embolization were associated with longer OS (p < 0.05). Multivariate logistic regression analysis demonstrated that tumor burden < 25% and hypervascular tumors were independent risk factors. Mean post-HAE hospital stay was 4 days. Grade 3 complication rate was 8.5%. In heavily treated patients with IHC, after exhausting all chemotherapy and other locoregional options, HAE as a rescue treatment option appeared to be safe with a mean OS of 13 months. Tumor burden < 25%, increased target tumor vascularity on pre-procedure imaging, and OR on 1 month follow-up images were associated with better OS. Further studies with a control group are required to confirm the effectiveness of HAE in IHC.

Percutaneous liver interventions with robotic systems: a systematic review of available clinical solutions.

OBJECTIVE: Robotic-guided interventions are emerging techniques that are gradually becoming a common tool for performing biopsies and tumor ablations in liver. This systematic review aims to evaluate their advancements, challenges, and outcomes. METHODS: A systematic review was conducted using the PubMed database to identify relevant articles published between January 2000 and February 2023. Inclusion criteria focused on studies that assessed robotic systems for percutaneous liver biopsies and tumor ablations. Data extraction was performed to collect information on study characteristics; robotic systems; components and software; imaging modality; degree of freedom; and needle insertion methods. The outcome measures analyzed were procedure time, radiation dose, and accuracy. RESULTS: 10 studies met the inclusion criteria. The robotic devices used included MAXIO, EPIONE, ROBIO-EX, AcuBot, and ACE robotic systems. The data set consisted of 429 percutaneous thermal ablations and 57 biopsies, both robot-guided. On average, the mean deviation of probes was reduced by 30% (from 1.6 vs 3.3 mm to 2.4 vs 3.9 mm (p < 0.001)), and 40% (p < 0.05) fewer readjustments were required during the robotic-assisted interventions. Moreover, robotic systems contributed to a reduction in operating time, ranging from 15% (18.3 vs 21.7 min, p < 0.001) to 25% (63.5 vs 87.4 min, p < 0.001). Finally, the radiation dose delivered to both the patient and the operator was decreased by an average of 50% (p < 0.05) compared to manual procedures. ADVANCES IN KNOWLEDGE: Robotic systems could provide precise navigation and guidance during liver biopsies and percutaneous ablations.

Anti-Glycolytic Drugs in the Treatment of Hepatocellular Carcinoma: Systemic and Locoregional Options.

Hepatocellular cancer (HCC) is the most common primary liver cancer and the third leading cause of cancer-related death. Locoregional therapies, including transarterial embolization (TAE: bland embolization), chemoembolization (TACE), and radioembolization, have demonstrated survival benefits when treating patients with unresectable HCC. TAE and TACE occlude the tumor's arterial supply, causing hypoxia and nutritional deprivation and ultimately resulting in tumor necrosis. Embolization blocks the aerobic metabolic pathway. However, tumors, including HCC, use the "Warburg effect" and survive hypoxia from embolization. An adaptation to hypoxia through the Warburg effect, which was first described in 1956, is when the cancer cells switch to glycolysis even in the presence of oxygen. Hence, this is also known as aerobic glycolysis. In this article, the adaptation mechanisms of HCC, including glycolysis, are discussed, and anti-glycolytic treatments, including systemic and locoregional options that have been previously reported or have the potential to be utilized in the treatment of HCC, are reviewed.

Early Prediction of Response of Hepatocellular Carcinoma to Yttrium-90 Radiation Segmentectomy Using a Machine Learning MR Imaging Radiomic Approach.

PURPOSE: To assess the accuracy of a machine learning (ML) approach based on magnetic resonance (MR) imaging radiomic quantification obtained before treatment and early after treatment for prediction of early hepatocellular carcinoma (HCC) response to yttrium-90 transarterial radioembolization (TARE). MATERIALS AND METHODS: In this retrospective single-center study of 76 patients with HCC, baseline and early (1-2 months) post-TARE MR images were collected. Semiautomated tumor segmentation facilitated extraction of shape, first-order histogram, and custom signal intensity-based radiomic features, which were then trained (n = 46) using a ML XGBoost model and validated on a separate cohort (n = 30) not used in training to predict treatment response assessed at 4-6 months (based on modified Response and Evaluation Criteria in Solid Tumors criteria). Performance of this ML radiomic model was compared with those of models comprising clinical parameters and standard imaging characteristics using area under the receiver operating curve (AUROC) analysis for prediction of complete response (CR). RESULTS: Seventy-six tumors with a mean (±SD) diameter of 2.6 cm ± 1.6 were included. Sixty, 12, 1, and 3 patients were classified as having CR, partial response, stable disease, and progressive disease, respectively, at 4-6 months posttreatment on the basis of MR images. In the validation cohort, the radiomic model showed good performance (AUROC, 0.89) for prediction of CR, compared with models comprising clinical and standard imaging criteria (AUROC, 0.58 and 0.59, respectively). Baseline imaging features appeared to be more heavily weighted in the radiomic model. CONCLUSIONS: The use of ML modeling of radiomic data combining baseline and early follow-up MR imaging could predict HCC response to TARE. These models need to be investigated further in an independent cohort.

Factors affecting outcomes of Yttrium-90 radioembolization in heavily pretreated patients with intrahepatic cholangiocarcinoma.

PURPOSE: Transarterial radioembolization (TARE) is a liver-directed treatment for unresectable intrahepatic cholangiocarcinoma (ICC). The aim of this study is to evaluate factors affecting outcomes of TARE in heavily pretreated ICC patients. METHODS: We evaluated pretreated ICC patients who received TARE from January 2013 to December 2021. Prior treatments included systemic therapy, hepatic resection, and liver-directed therapies, including hepatic arterial infusion chemotherapy, external beam radiation, transarterial embolization, and thermal ablation. Patients were classified based on history of hepatic resection and genomic status based on next-generation sequencing (NGS). The primary endpoint was overall survival (OS) after TARE. RESULTS: Fourteen patients with median age 66.1 years (range, 52.4-87.5), 11 females and 3 males, were included. Prior therapies included systemic in 13/14 patients (93%), liver resection in 6/14 (43%), and liver-directed therapy in 6/14 (43%). Median OS was 11.9 months (range, 2.8-81.0). Resected patients had significantly longer median OS compared to unresected patients (16.6 versus 7.9 months; p = 0.038). Prior liver-directed therapy (p = 0.043), largest tumor diameter > 4 cm (p = 0.014), and > 2 hepatic segments involvement (p = 0.001) were associated with worse OS. Nine patients underwent NGS; 3/9 (33.3%) and had a high-risk gene signature (HRGS), defined as alterations in TP53, KRAS, or CDKN2A. Patients with a HRGS had worse median OS (10.0 versus 17.8 months; p = 0.024). CONCLUSIONS: TARE may be used as salvage therapy in heavily treated ICC patients. Presence of a HRGS may predict worse OS after TARE. Further investigation with more patients is recommended to validate these results.

Image-guided percutaneous locoregional therapies for hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is the most common primary liver cancer and the 3rd leading cause of cancer death worldwide. Treatment options include surgical resection, liver transplantation, image-guided percutaneous locoregional options, external beam radiation therapy (EBRT) and systemic therapies. Treatment choice depends on the stage of the disease and patient's characteristics including performance status and liver function. Barcelona Clinic Liver Cancer (BCLC) staging system, with its recent 2022 update, is one of the most widely endorsed staging system. Locoregional therapies (LRT) are recommended for very early stage (BCLC-0), early stage (BCLC-A), and the two first subgroups of intermediate stage (BCLC-B). Image-guided percutaneous locoregional therapies include ablation, mainly thermal ablation with radiofrequency (RFA), microwave ablations (MWA) and cryoablation, transarterial embolization (TAE, also known as bland embolization), transarterial chemoembolization (TACE), drug-eluding beads-transarterial chemoembolization (DEB-TACE), combination of ablation with embolization, transarterial radioembolization (TARE) also known as selective internal radioembolization therapy, and hepatic artery infusion (HAI). While ablation is recognized as a curative therapy, all intra-arterial therapies are considered non-curative options. There is growing evidence that TARE, through radiation segmentectomy, can be considered a curative intent treatment in appropriate selective patients. In this article, we will review indications, complications, and outcomes of locoregional therapies for HCC.

Fertility after Transcatheter Arterial Embolization for Obstetric Hemorrhage: An Urban Health Care System Observational Study.

OBJECTIVE: This study aimed to investigate pregnancy rate, pregnancy outcomes, and resumption of menses after transcatheter arterial embolization (TAE) for obstetric hemorrhage (OH). STUDY DESIGN: Sixty-seven patients who underwent TAE for OH from 2006 to 2020 within an urban, multihospital health care system were identified retrospectively. Selected patients were interviewed by phone to complete a survey with a primary outcome of self-reported pregnancy in those seeking pregnancy. Secondary outcomes included pregnancy outcomes and resumption of menses. Univariate testing of association of pregnancy and miscarriage rate with embolic agent was performed using Fisher's exact test. RESULTS: Thirty-three of 50 patients (66%) meeting the inclusion criteria completed the survey on fertility, a median of 47 (range, 13-123) months after TAE for OH. Of the 13 patients who attempted pregnancy, there was a pregnancy rate of 77% and miscarriage rate of 38%. Those who delivered live newborns conceived spontaneously, carried to term, and delivered a healthy newborn via cesarean section at a weight appropriate for gestational age. Thirty (91%) patients resumed menstruation, and the majority with unchanged frequency. Most patients underwent bilateral uterine artery embolization with radial artery access (54%). The most common embolic agents used were gelfoam only (30%) and glue only (24%). There was no statistically significant association between embolic agent and pregnancy or miscarriage rate. CONCLUSION: Spontaneous pregnancy with live birth and resumption of menses can occur in a majority of patients after TAE for OH. KEY POINTS: · Most patients who attempted pregnancy after TAE for OH achieved pregnancy.. · Most patients who became pregnant conceived spontaneously and delivered healthy newborns at term.. · Most patients resumed menstruation after TAE for OH.. · There was no significant association between type of embolic and pregnancy or miscarriage rate..

Leveraging the virtual learning environment to enhance medical student engagement with interventional radiology.

PURPOSE: To evaluate medical student engagement with Interventional Radiology (IR) before and after a virtual elective course. METHODS: The elective was nine, one-hour lectures over ten weeks. An anonymous pre and post-course survey was administered to students. The hypothesis was that this course would increase student engagement with IR. Respondents answered nine questions to score their interest in, exposure to, familiarity with, and understanding of IR using a five-point Likert scale. Demographics were reported for the pre-course group only. A Wilcoxon signed-ranked test was performed to assess for significant mean change in pre and post-course responses. Among the 276 registered students, there were 144 individual, complete responses for the pre-course survey, and 60 paired responses for both surveys. RESULTS: Thirty-seven percent of respondents were first or second year medical students. Thirty percent of participants were enrolled at an institution outside of the United States, 26% are the first in their family to attend college, and 41% identified as female. Thirty-six percent reported this virtual course was one of their earliest experiences with IR. There was a significant increase in student exposure to IR generally, familiarity with IR compared to other specialties, familiarity with the IR training pathway(s), understanding of what an Interventional Radiologist does, understanding of the difference between IR and Diagnostic Radiology, and understanding of when to consult IR for patient care after completion of the course. CONCLUSION: A virtual IR elective is an effective means to increase exposure to, familiarity with, and understanding of IR.

Safety and Feasibility of a Novel Externally Cooled Bronchoscopic Radiofrequency Ablation Catheter for Ablation of Peripheral Lung Tumours: A First-In-Human Dose Escalation Study.

BACKGROUND: Radiofrequency ablation (RFA) is an established modality for percutaneous ablation of non-small cell lung cancer (NSCLC) in medically inoperable patients but is underutilized clinically due to side effects. We have developed a novel, completely endobronchial RFA catheter with an externally cooled electrode. OBJECTIVES: The objective of this study was to establish the safety and feasibility of bronchoscopic RFA using a novel, externally cooled catheter for ablation of peripheral NSCLC. METHODS: Patients with stage I biopsy-confirmed NSCLC underwent bronchoscopic RFA of tumour 7 days prior to lobectomy. The RFA catheter was delivered bronchoscopically to peripheral NSCLC lesions, guided by radial endobronchial ultrasound, with positioning confirmed using intra-procedural cone beam CT. Pre-operative CT chest and histologic examination of resected specimens were used to establish distribution/uniformity of ablation and efficacy of tumour ablation. RESULTS: RFA in the first patient was complicated by dispersal of heated saline due to cough, resulting in ICU admission. The patient recovered fully and underwent uncomplicated lobectomy. Subsequently, the protocol was altered to mandate neuromuscular blockade with a pre-determined dose escalation, with algorithm-restricted energy (kJ) and irrigated saline volume (mL) constraints. A further 10 patients consented and seven underwent successful bronchoscopic RFA of peripheral NSCLC. No significant adverse events were noted. Ablation zone included tumour in all cases (proportion of tumour ablated ranged 8-72%), with uniform necrosis of tissue within ablation zones observed at higher energy levels. Ablation zone diameter correlated with RFA energy delivered (R2 = 0.553), with maximum long axis diameter of ablation zone 3.1 cm (22.9 kJ). CONCLUSION: Bronchoscopic RFA using an externally cooled catheter is feasible, appears safe, and achieves uniform ablation within the treatment zone. Uncontrolled escape of heated saline poses a major safety risk but can be prevented procedurally through neuromuscular blockade and by limiting irrigation.

RadImageNet: An Open Radiologic Deep Learning Research Dataset for Effective Transfer Learning.

Purpose: To demonstrate the value of pretraining with millions of radiologic images compared with ImageNet photographic images on downstream medical applications when using transfer learning. Materials and Methods: This retrospective study included patients who underwent a radiologic study between 2005 and 2020 at an outpatient imaging facility. Key images and associated labels from the studies were retrospectively extracted from the original study interpretation. These images were used for RadImageNet model training with random weight initiation. The RadImageNet models were compared with ImageNet models using the area under the receiver operating characteristic curve (AUC) for eight classification tasks and using Dice scores for two segmentation problems. Results: The RadImageNet database consists of 1.35 million annotated medical images in 131 872 patients who underwent CT, MRI, and US for musculoskeletal, neurologic, oncologic, gastrointestinal, endocrine, abdominal, and pulmonary pathologic conditions. For transfer learning tasks on small datasets-thyroid nodules (US), breast masses (US), anterior cruciate ligament injuries (MRI), and meniscal tears (MRI)-the RadImageNet models demonstrated a significant advantage (P < .001) to ImageNet models (9.4%, 4.0%, 4.8%, and 4.5% AUC improvements, respectively). For larger datasets-pneumonia (chest radiography), COVID-19 (CT), SARS-CoV-2 (CT), and intracranial hemorrhage (CT)-the RadImageNet models also illustrated improved AUC (P < .001) by 1.9%, 6.1%, 1.7%, and 0.9%, respectively. Additionally, lesion localizations of the RadImageNet models were improved by 64.6% and 16.4% on thyroid and breast US datasets, respectively. Conclusion: RadImageNet pretrained models demonstrated better interpretability compared with ImageNet models, especially for smaller radiologic datasets.Keywords: CT, MR Imaging, US, Head/Neck, Thorax, Brain/Brain Stem, Evidence-based Medicine, Computer Applications-General (Informatics) Supplemental material is available for this article. Published under a CC BY 4.0 license.See also the commentary by Cadrin-Chênevert in this issue.

Contemporary Management of Acute Pulmonary Embolism: Evolution of Catheter-based Therapy.

Acute pulmonary embolism (PE) affects more than 100 000 people in the United States annually and is the third leading cardiovascular cause of death. The standard management for PE is systemic anticoagulation therapy. However, a subset of patients experience hemodynamic decompensation, despite conservative measures. Traditionally, these patients have been treated with systemic administration of thrombolytic agents or open cardiac surgery, although attempts at endovascular treatment have a long history that dates back to the 1960s. The technology for catheter-based therapy for acute PE is rapidly evolving, with multiple devices approved over the past decade. Currently available devices fall into two broad categories of treatment methods: catheter-directed thrombolysis and percutaneous suction thrombectomy. Catheter-directed thrombolysis is the infusion of thrombolytic agents directly into the occluded pulmonary arteries to increase local delivery and decrease the total dose. Suction thrombectomy involves the use of small- or large-bore catheters to mechanically aspirate a clot from the pulmonary arteries without the need for a thrombolytic agent. A thorough understanding of the various risk stratification schemes and the available evidence for each device is critical for optimal treatment of this complex entity. Multiple ongoing studies will improve our understanding of the role of catheter-based therapy for acute PE in the next 5-10 years. A multidisciplinary approach through PE response teams has become the management standard at most institutions. An invited commentary by Bulman and Weinstein is available online. Online supplemental material is available for this article. ©RSNA, 2022.

Quantitative chest computed tomography combined with plasma cytokines predict outcomes in COVID-19 patients.

Despite extraordinary international efforts to dampen the spread and understand the mechanisms behind SARS-CoV-2 infections, accessible predictive biomarkers directly applicable in the clinic are yet to be discovered. Recent studies have revealed that diverse types of assays bear limited predictive power for COVID-19 outcomes. Here, we harness the predictive power of chest computed tomography (CT) in combination with plasma cytokines using a machine learning and k-fold cross-validation approach for predicting death during hospitalization and maximum severity degree in COVID-19 patients. Patients (n = 152) from the Mount Sinai Health System in New York with plasma cytokine assessment and a chest CT within five days from admission were included. Demographics, clinical, and laboratory variables, including plasma cytokines (IL-6, IL-8, and TNF-α), were collected from the electronic medical record. We found that CT quantitative alone was better at predicting severity (AUC 0.81) than death (AUC 0.70), while cytokine measurements alone better-predicted death (AUC 0.70) compared to severity (AUC 0.66). When combined, chest CT and plasma cytokines were good predictors of death (AUC 0.78) and maximum severity (AUC 0.82). Finally, we provide a simple scoring system (nomogram) using plasma IL-6, IL-8, TNF-α, ground-glass opacities (GGO) to aerated lung ratio and age as new metrics that may be used to monitor patients upon hospitalization and help physicians make critical decisions and considerations for patients at high risk of death for COVID-19.

Predicting Patient Demographics From Chest Radiographs With Deep Learning.

BACKGROUND: Deep learning models are increasingly informing medical decision making, for instance, in the detection of acute intracranial hemorrhage and pulmonary embolism. However, many models are trained on medical image databases that poorly represent the diversity of the patients they serve. In turn, many artificial intelligence models may not perform as well on assisting providers with important medical decisions for underrepresented populations. PURPOSE: Assessment of the ability of deep learning models to classify the self-reported gender, age, self-reported ethnicity, and insurance status of an individual patient from a given chest radiograph. METHODS: Models were trained and tested with 55,174 radiographs in the MIMIC Chest X-ray (MIMIC-CXR) database. External validation data came from two separate databases, one from CheXpert and another from a multihospital urban health care system after institutional review board approval. Macro-averaged area under the curve (AUC) values were used to evaluate performance of models. Code used for this study is open-source and available at https://github.com/ai-bias/cxr-bias, and pixelstopatients.com/models/demographics. RESULTS: Accuracy of models to predict gender was nearly perfect, with 0.999 (95% confidence interval: 0.99-0.99) AUC on held-out test data and 0.994 (0.99-0.99) and 0.997 (0.99-0.99) on external validation data. There was high accuracy to predict age and ethnicity, ranging from 0.854 (0.80-0.91) to 0.911 (0.88-0.94) AUC, and moderate accuracy to predict insurance status, with AUC ranging from 0.705 (0.60-0.81) on held-out test data to 0.675 (0.54-0.79) on external validation data. CONCLUSIONS: Deep learning models can predict the age, self-reported gender, self-reported ethnicity, and insurance status of a patient from a chest radiograph. Visualization techniques are useful to ensure deep learning models function as intended and to demonstrate anatomical regions of interest. These models can be used to ensure that training data are diverse, thereby ensuring artificial intelligence models that work on diverse populations.

Integrated use of PD-1 inhibition and transarterial chemoembolization for hepatocellular carcinoma: evaluation of safety and efficacy in a retrospective, propensity score-matched study.

BACKGROUND: Immune checkpoint inhibitors (ICIs) have revolutionized treatment of advanced hepatocellular carcinoma. Integrated use of transarterial chemoembolization (TACE), a locoregional inducer of immunogenic cell death, with ICI has not been formally assessed for safety and efficacy outcomes. METHODS: From a retrospective multicenter dataset of 323 patients treated with ICI, we identified 31 patients who underwent >1 TACE 60 days before or concurrently, with nivolumab at a single center. We derived a propensity score-matched cohort of 104 patients based on Child-Pugh Score, portal vein thrombosis, extrahepatic metastasis and alpha fetoprotein (AFP) who received nivolumab monotherapy. We described overall survival (OS), progression-free survival (PFS), objective responses according to modified RECIST criteria and safety in the multimodal arm in comparison to monotherapy. RESULTS: Over a median follow-up of 9.3 (IQR 4.0-16.4) months, patients undergoing multimodal immunotherapy with TACE achieved a significantly longer median (95% CI) PFS of 8.8 (6.2-23.2) vs 3.7 (2.7-5.4) months (log-rank 0.15, p<0.01) in the monotherapy group. Multimodal immunotherapy with TACE demonstrated a numerically longer OS compared with ICI monotherapy with a median 35.1 (16.1-Not Evaluable) vs 16.6 (15.7-32.6) months (log-rank 0.41, p=0.12). In the multimodal treatment group, there were three (10%) grade 3 or higher adverse events (AEs) attributed to immunotherapy compared with seven (6.7%) in the matched ICI monotherapy arm. There were no AEs grade 3 or higher attributed to TACE in the multimodal treatment arm. At 3 months following each TACE in the multimodal arm, there was an overall objective response rate of 84%. There were no significant changes in liver functional reserve 1 month following each TACE. Four patients undergoing multimodal treatment were successfully bridged to transplant. CONCLUSIONS: TACE can be safely integrated with programmed cell death 1 blockade and may lead to a significant delay in tumor progression and disease downstaging in selected patients.