A Classification-Based Adaptive Segmentation Pipeline: Feasibility Study Using Polycystic Liver Disease and Metastases from Colorectal Cancer CT Images.

Automated segmentation tools often encounter accuracy and adaptability issues when applied to images of different pathology. The purpose of this study is to explore the feasibility of building a workflow to efficiently route images to specifically trained segmentation models. By implementing a deep learning classifier to automatically classify the images and route them to appropriate segmentation models, we hope that our workflow can segment the images with different pathology accurately. The data we used in this study are 350 CT images from patients affected by polycystic liver disease and 350 CT images from patients presenting with liver metastases from colorectal cancer. All images had the liver manually segmented by trained imaging analysts. Our proposed adaptive segmentation workflow achieved a statistically significant improvement for the task of total liver segmentation compared to the generic single-segmentation model (non-parametric Wilcoxon signed rank test, n = 100, p-value << 0.001). This approach is applicable in a wide range of scenarios and should prove useful in clinical implementations of segmentation pipelines.

Deep convolutional-neural-network-based metal artifact reduction for CT-guided interventional oncology procedures (MARIO).

BACKGROUND: Computed tomography (CT) is routinely used to guide cryoablation procedures. Notably, CT-guidance provides 3D localization of cryoprobes and can be used to delineate frozen tissue during ablation. However, metal-induced artifacts from ablation probes can make accurate probe placement challenging and degrade the ice ball conspicuity, which in combination could lead to undertreatment of potentially curable lesions. PURPOSE: In this work, we propose an image-based neural network (CNN) model for metal artifact reduction for CT-guided interventional procedures. METHODS: An image domain metal artifact simulation framework was developed and validated for deep-learning-based metal artifact reduction for interventional oncology (MARIO). CT scans were acquired for 19 different cryoablation probe configurations. The probe configurations varied in the number of probes and the relative orientations. A combination of intensity thresholding and masking based on maximum intensity projections (MIPs) was used to segment both the probes only and probes + artifact in each phantom image. Each of the probe and probe + artifact images were then inserted into 19 unique patient exams, in the image domain, to simulate metal artifact appearance for CT-guided interventional oncology procedures. The resulting 361 pairs of simulated image volumes were partitioned into disjoint training and test datasets of 304 and 57 volumes, respectively. From the training partition, 116 600 image patches with a shape of 128 × 128 × 5 pixels were randomly extracted to be used for training data. The input images consisted of a superposition of the patient and probe + artifact images. The target images consisted of a superposition of the patient and probe only images. This dataset was used to optimize a U-Net type model. The trained model was then applied to 50 independent, previously unseen CT images obtained during renal cryoablations. Three board-certified radiologists with experience in CT-guided ablations performed a blinded review of the MARIO images. A total of 100 images (50 original, 50 MARIO processed) were assessed across different aspects of image quality on a 4-point likert-type item. Statistical analyses were performed using Wilcoxon signed-rank test for paired samples. RESULTS: Reader scores were significantly higher for MARIO processed images compared to the original images across all metrics (all p < 0.001). The average scores of the overall image quality, iceball conspicuity, overall metal artifact, needle tip visualization, target region confidence, and worst metal artifact, needle tip visualization, iceball conspicuity, and target region confidence improved by 34.91%, 36.29%, 39.94%, 34.17%, 35.13%, and 45.70%, respectively. CONCLUSIONS: The proposed method of image-based metal artifact simulation can be used to train a MARIO algorithm to effectively reduce probe-related metal artifacts in CT-guided cryoablation procedures.

Sublobar Resection, Stereotactic Body Radiation Therapy, and Percutaneous Ablation Provide Comparable Outcomes for Lung Metastasis-Directed Therapy.

BACKGROUND: Prolonged survival of patients with metastatic disease has furthered interest in metastasis-directed therapy (MDT). RESEARCH QUESTION: There is a paucity of data comparing lung MDT modalities. Do outcomes among sublobar resection (SLR), stereotactic body radiation therapy (SBRT), and percutaneous ablation (PA) for lung metastases vary in terms of local control and survival? STUDY DESIGN AND METHODS: Medical records of patients undergoing lung MDT at a single cancer center between January 2015 and December 2020 were reviewed. Overall survival, local progression, and toxicity outcomes were collected. Patient and lesion characteristics were used to generate multivariable models with propensity weighted analysis. RESULTS: Lung MDT courses (644 total: 243 SLR, 274 SBRT, 127 PA) delivered to 511 patients were included with a median follow-up of 22 months. There were 47 local progression events in 45 patients, and 159 patients died. Two-year overall survival and local progression were 80.3% and 63.3%, 83.8% and 9.6%, and 4.1% and 11.7% for SLR, SBRT, and PA, respectively. Lesion size per 1 cm was associated with worse overall survival (hazard ratio, 1.24; P = .003) and LP (hazard ratio, 1.50; P < .001). There was no difference in overall survival by modality. Relative to SLR, there was no difference in risk of local progression with PA; however, SBRT was associated with a decreased risk (hazard ratio, 0.26; P = .023). Rates of severe toxicity were low (2.1%-2.6%) and not different among groups. INTERPRETATION: This study performs a propensity weighted analysis of SLR, SBRT, and PA and shows no impact of lung MDT modality on overall survival. Given excellent local control across MDT options, a multidisciplinary approach is beneficial for patient triage and longitudinal management.

Musculoskeletal Interventional Oncology: A Contemporary Review.

Musculoskeletal interventional oncology is an emerging field that addresses the limitations of conventional therapies for bone and soft-tissue tumors. The field's growth has been driven by evolving treatment paradigms, expanding society guidelines, mounting supportive literature, technologic advances, and cross-specialty collaboration with medical, surgical, and radiation oncology. Safe, effective, and durable pain palliation, local control, and stabilization of musculoskeletal tumors are increasingly achieved through an expanding array of contemporary minimally invasive percutaneous image-guided treatments, including ablation, osteoplasty, vertebral augmentation (with or without mechanical reinforcement via implants), osseous consolidation via percutaneous screw fixation (with or without osteoplasty), tumor embolization, and neurolysis. These interventions may be used for curative or palliative indications and can be readily combined with systemic therapies. Therapeutic approaches include the combination of different interventional oncology techniques as well as the sequential application of such techniques with other local treatments, including surgery or radiation. This article reviews the current practice of interventional oncology treatments for the management of patients with bone and soft-tissue tumors with a focus on emerging technologies and techniques.

Percutaneous Cryoablation of Plasmacytomas: Oncologic Effectiveness and Adverse Events.

PURPOSE: To evaluate the oncologic outcomes and adverse events associated with cryoablation of plasmacytomas. MATERIALS AND METHODS: Retrospective review of an institutional percutaneous ablation database showed that 43 patients underwent 46 percutaneous cryoablation procedures for treatment of 44 plasmacytomas between May 2004 and March 2021. The treatment of 25 (25 of 44, 56.8%) tumors was augmented with bone consolidation/cementoplasty. The median patient age was 64 years (interquartile range [IQR], 54-69), and 30 of 43 (69.8%) patients were men. The median maximum plasmacytoma diameter was 5.0 cm (IQR, 3.1-7.0). Thirty of 44 (68.2%) tumors were periacetabular, vertebral, or located in the iliac wing. Twenty-nine of 44 (65.9%) cryoablated plasmacytomas were recurrent tumors after prior external beam radiation therapy (EBRT). Survival analyses were performed using the Kaplan-Meier method. Adverse events were graded using Society of Interventional Radiology criteria. RESULTS: The 5-year estimated local tumor recurrence-free survival was 85.3% (95% CI, 74.1%-98.1%), the 5-year estimated new plasmacytoma-free survival was 49.9% (95% CI, 33.9%-73.4%), and the 5-year estimated overall survival was 70.4% (95% CI, 56.9%-87.1%). Nine of 46 (19.6%) major adverse events occurred in 8 patients, including 3 of 46 (6.5%) new or progressive pathologic fractures at the ablation site requiring surgical intervention, 3 of 46 (6.5%) nerve injuries, 1 of 46 (2.2%) avascular necrosis and femoral head collapse, 1 of 46 (2.2%) septic arthritis, and 1 of 46 (2.2%) acute renal failure caused by rhabdomyolysis. CONCLUSIONS: Percutaneous cryoablation is a viable treatment option for patients with plasmacytomas, including those with recurrent plasmacytomas after EBRT. Postcryoablation adverse events are relatively common.

Clinical Implementation of an Artificial Intelligence Algorithm for Magnetic Resonance-Derived Measurement of Total Kidney Volume.

OBJECTIVE: To evaluate the performance of an internally developed and previously validated artificial intelligence (AI) algorithm for magnetic resonance (MR)-derived total kidney volume (TKV) in autosomal dominant polycystic kidney disease (ADPKD) when implemented in clinical practice. PATIENTS AND METHODS: The study included adult patients with ADPKD seen by a nephrologist at our institution between November 2019 and January 2021 and undergoing an MR imaging examination as part of standard clinical care. Thirty-three nephrologists ordered MR imaging, requesting AI-based TKV calculation for 170 cases in these 161 unique patients. We tracked implementation and performance of the algorithm over 1 year. A radiologist and a radiology technologist reviewed all cases (N=170) for quality and accuracy. Manual editing of algorithm output occurred at radiology or radiology technologist discretion. Performance was assessed by comparing AI-based and manually edited segmentations via measures of similarity and dissimilarity to ensure expected performance. We analyzed ADPKD severity class assignment of algorithm-derived vs manually edited TKV to assess impact. RESULTS: Clinical implementation was successful. Artificial intelligence algorithm-based segmentation showed high levels of agreement and was noninferior to interobserver variability and other methods for determining TKV. Of manually edited cases (n=84), the AI-algorithm TKV output showed a small mean volume difference of -3.3%. Agreement for disease class between AI-based and manually edited segmentation was high (five cases differed). CONCLUSION: Performance of an AI algorithm in real-life clinical practice can be preserved if there is careful development and validation and if the implementation environment closely matches the development conditions.

Radiofrequency Ablation of Benign and Malignant Thyroid Nodules.

Thyroid nodule treatment has significantly evolved over recent years with attempts to individualize treatment on the basis of the cause of the nodule and patient performance status. The risks and complications associated with surgery and radioactive iodine have promoted interest in additional therapies such as radiofrequency ablation (RFA). RFA creates an electrical current through a target tissue (thyroid nodule) with resultant tissue heating causing coagulative necrosis. National and international groups are beginning to recognize the role of RFA as a viable therapeutic option in the treatment of thyroid nodules. Based on numerous guidelines, RFA is indicated in the treatment of symptomatic benign nodules and autonomously functioning nodules when surgery is refused or when the patient would not tolerate surgery. The treatment of thyroid malignancy with RFA is controversial, with some groups advocating for its use in the treatment of small papillary thyroid cancers in specific scenarios. The most important aspect of RFA is the preprocedural workup and adequate patient selection. Procedural technique varies among centers. However, RFA is typically performed as a single-day-admission outpatient procedure. Methods such as hydrodissection and a moving shot technique are employed to ensure adequate coverage of the nodule without overtreating the peripheries and damaging sensitive structures. As a result, the procedure is well tolerated, and major complications such as recurrent laryngeal nerve injury and nodule rupture are very rare. In the proper patient cohort, thyroid RFA offers an efficacious and safe option in the management of thyroid nodules. An invited commentary by Filippiadis and Vrachliotis is available online. ©RSNA, 2022.

Improved Ultrasound Attenuation Estimation with Non-uniform Structure Detection and Removal.

Accurate detection of liver steatosis is important for liver disease management. Ultrasound attenuation coefficient estimation (ACE) has great potential in quantifying liver fat content. The ACE methods commonly assume uniform tissue characteristics. However, in vivo tissues typically contain non-uniform structures, which may bias the attenuation estimation and lead to large standard deviations. Here we propose a series of non-uniform structure detection and removal (NSDR) methods to reduce the impact from non-uniform structures during ACE analysis. The effectiveness of NSDR was validated through phantom and in vivo studies. In a pilot clinical study, ACE with NSDR provided more robust in vivo performance as compared with ACE without NSDR, indicating its potential for in vivo applications.

A prospective trial of CT-guided percutaneous microwave ablation for lung tumors.

Background: Percutaneous ablation is an alternative treatment for lung cancer in non-operable patients. This is a prospective clinical trial for percutaneous microwave ablation (pMWA) of biopsy-proven lung cancer to demonstrate safety and efficacy. Methods: A prospective trial from 6-1-2016 to 1-1-2019 enrolled patients with biopsy-proven primary or metastatic lung cancer <3 cm in size and 1 cm away from the pleura for pMWA with the Emprint Ablation System with Thermosphere Technology for Phase I analysis, (Clinicaltrials.gov; #NCT0267302). Patients were followed for 1 year with PET/CT and PET/MR to determine patterns of recurrence and efficacy of ablation. Results: After 12 patients consented for biopsy, 6 patients underwent treatment of 7 lesions, 3/6 women, median age of 67 (IQR, 65-70) years, body mass index (BMI): 27.8 (IQR, 21.4-32.1) kg/m2, lesion distance to pleura 24.4 (IQR, 13-38) mm, lesion size of 10.7 (IQR, 6-14) mm, and ablation duration time 5.9 (IQR, 3-10) minutes. pMWA were completed at 75 W. Twelve adverse events were reported (1 Grade 3, 3 Grade 2, and 8 Grade 1 events) with Grade 4 or 5 events. Mean % change after ablation in forced expiratory volume in one second (FEV1) was -2% and diffusion capacity for carbon monoxide (DLCO) was -1%. After 2-3 months, the lesions would decrease in size, rim thickness, fluorodeoxyglucose (FDG) activity, and T2 signal. FDG activity after 6 months was below blood pool in all cases. The ablation zones stabilized by 6-12 months. One patient expired during the study from pneumonia unrelated to ablation without local recurrence. Of the seven ablations during the 1 year follow-up, there was local tumor recurrence at 271 days following ablation at the apex of the ablation zone, subsequently successfully treated with percutaneous cryoablation (Cryo). Conclusions: pMWA appears to be a safe and effective mechanism for treatment of primary and secondary tumors of the lung, with possible preservation of pulmonary function.

Innovative Fetal Therapy for a Giant Congenital Pulmonary Airway Malformation with Hydrops.

INTRODUCTION: Congenital pulmonary airway malformations (CPAMs) complicated by hydrops portend significant morbidity and mortality, with fetal survival estimates less than 10%. CASE PRESENTATION: We report successful use of ultrasound-guided radiofrequency ablation at 21-week gestation in a hydropic fetus with CPAM, with subsequent resolution of hydrops. Thirty-two-week MRI noted persistent mediastinal shift, and US at 36 weeks and 5 days noted polyhydramnios. Maternal gestational hypertension prompted delivery at 37 weeks, with a cesarean section performed after a failed trial of labor. The infant required CPAP at 100% and weaned to 21%. Tachypnea persisted, and chest CT on day of life 2 demonstrated multiple large cysts in the right lower lobe with anterior pneumothorax. On day of life 3, she successfully underwent a thoracoscopic right lower lobectomy. Adhesions to the chest wall and rib abnormalities were noted. She was extubated to CPAP at the conclusion of the procedure. She was able to wean to 21% on POD2 and transitioned to oral feeds. Her chest tube was removed with resultant ex vacuo pneumothorax noted. She remained asymptomatic and was discharged home on room air POD11. Pathology confirmed a type 1 CPAM. CONCLUSION: In utero radiofrequency ablation may be an adjunct to the management of large CPAM.

Bone ablations in peripheral skeleton: Rationale, techniques and evidence.

This article aims to disclose a consensus on the rationale, approaches, and the outcomes of bone ablations in the peripheral skeleton. Despite less numerous prospective studies about peripheral metastasis, interventional radiology has a role in this setting. Scrupulous attention for selection criteria, ablation technique, procedural steps, and clinical and imaging follow-up are required to provide optimal multidisciplinary care for oncologic patients.

Reverberation clutter signal suppression in ultrasound attenuation estimation using wavelet-based robust principal component analysis.

Objective. Ultrasound attenuation coefficient estimation (ACE) has diagnostic potential for clinical applications such as quantifying fat content in the liver. Previously, we have proposed a system-independent ACE technique based on spectral normalization of different frequencies, called the reference frequency method (RFM). This technique does not require a well-calibrated reference phantom for normalization. However, this method may be vulnerable to severe reverberation clutter introduced by the body wall. The clutter superimposed on liver echoes may bias the estimation.Approach. We proposed to use robust principal component analysis, combined with wavelet-based sparsity promotion, to suppress the severe reverberation clutters. The capability to mitigate the reverberation clutters was validated through phantom andin vivostudies.Main Results. In the phantom studies with added reverberation clutters, higher normalized cross-correlation and smaller mean absolute errors were attained as compared to RFM results without the proposed method, demonstrating the capability to reconstruct tissue signals from reverberations. In a pilot patient study, the correlation between ACE and proton density fat fraction (PDFF), a measurement of liver fat by MRI as a reference standard, was investigated. The proposed method showed an improvement of the correlation (coefficient of determination,R = 0.82) as compared with the counterpart without the proposed method (R = 0.69).Significance: The proposed method showed the feasibility of suppressing the reverberation clutters, providing an important basis for the development of a robust ACE with large reverberation clutters.

Consensus Guidelines for the Definition of Time-to-Event End Points in Image-guided Tumor Ablation: Results of the SIO and DATECAN Initiative.

There is currently no consensus regarding preferred clinical outcome measures following image-guided tumor ablation or clear definitions of oncologic end points. This consensus document proposes standardized definitions for a broad range of oncologic outcome measures with recommendations on how to uniformly document, analyze, and report outcomes. The initiative was coordinated by the Society of Interventional Oncology in collaboration with the Definition for the Assessment of Time-to-Event End Points in Cancer Trials, or DATECAN, group. According to predefined criteria, based on experience with clinical trials, an international panel of 62 experts convened. Recommendations were developed using the validated three-step modified Delphi consensus method. Consensus was reached on when to assess outcomes per patient, per session, or per tumor; on starting and ending time and survival time definitions; and on time-to-event end points. Although no consensus was reached on the preferred classification system to report complications, quality of life, and health economics issues, the panel did agree on using the most recent version of a validated patient-reported outcome questionnaire. This article provides a framework of key opinion leader recommendations with the intent to facilitate a clear interpretation of results and standardize worldwide communication. Widespread adoption will improve reproducibility, allow for accurate comparisons, and avoid misinterpretations in the field of interventional oncology research. Published under a CC BY 4.0 license. Online supplemental material is available for this article. See also the editorial by Liddell in this issue.

The ECLIPSE Study: Efficacy of Cryoablation on Metastatic Lung Tumors With a 5-Year Follow-Up.

INTRODUCTION: The ECLIPSE study aimed to evaluate the feasibility and efficacy of cryoablation (CA) for local tumor control in patients with pulmonary metastatic disease in 5 years of follow-up. METHODS: ECLIPSE was a prospective, multicenter, single-arm study that included patients treated with CA if they had one to five metastatic lung tumors, each with a diameter of less than or equal to 3.5 cm. Patients were followed up in the course of 5 years. The primary end point was local tumor control, both per tumor and per patient; secondary end points included cancer-specific survival, overall survival, and quality of life (QoL). QoL was evaluated using the Karnofsky Performance Score, the Eastern Cooperative Oncology Group performance score, and the Short Form-12 health survey. RESULTS: The study included 40 patients across four sites (three in United States and one in Europe). A total of 60 metastatic pulmonary tumors were treated with 48 CA procedures. Overall local tumor control rates were 87.9% (29 of 33) and 79.2% (19 of 24) per tumor, 83.3% (20 of 24) and 75.0% (15 of 20) per patient, at 3 and 5 years, respectively. A total of five treated patients had local progression throughout the duration of the study. Disease-specific survival rate was 74.8% at 3 years and 55.3% at 5 years, whereas overall survival at 3 and 5 years was 63.2% and 46.7%, respectively. Patient QoL scores did not reach statistical significance. CONCLUSIONS: CA is an effective means of long-term local tumor control in patients with metastatic pulmonary tumors.

Musculoskeletal Oncologic Interventions: Proceedings from the Society of Interventional Radiology and Society of Interventional Oncology Research Consensus Panel.

Musculoskeletal interventions are increasingly used with palliative and curative intent in the multidisciplinary treatment of oncology patients with bone and soft-tissue tumors. There is an unmet need for high-quality evidence to guide broader application and adoption of minimally invasive interventional technologies to treat these patients. Therefore, the Society of Interventional Radiology Foundation and the Society of Interventional Oncology collaborated to convene a research consensus panel to prioritize a research agenda addressing the gaps in the current evidence. This article summarizes the panel's proceedings and recommendations for future basic science and clinical investigation to chart the course for interventional oncology within the musculoskeletal system. Key questions that emerged addressed the effectiveness of ablation within specific patient populations, the effect of combination of ablation with radiotherapy and/or immunotherapy, and the potential of standardization of techniques, including modeling and monitoring, to improve the consistency and predictability of treatment outcomes.

Combined Effects of Masking and Distance on Aerosol Exposure Potential.

OBJECTIVE: To quantify the efficacy of masking and "social distancing" on the transmission of airborne particles from a phantom aerosol source (simulating an infected individual) to a nearby target (simulating a healthy bystander) in a well-controlled setting. METHODS: An aerosol was created using monodisperse polystyrene latex beads in place of infectious respiratory secretions. Detection was by aerodynamic particle spectrometry. Both reusable cloth masks and disposable paper masks were studied. Transmission was simulated indoors during a 3-minute interval to eliminate the effect of variable ventilation rate on aerosol exposure. The study commenced on September 16, 2020, and concluded on December 15, 2020. RESULTS: Compared with a baseline of 1-foot separation with no masks employed, particle count was reduced by 84% at 3 feet of separation and 97% at 6 feet. A modest decrease in particle count was observed when only the receiver was masked. The most substantial exposure reduction occurred when the aerosol source was masked (or both parties were masked). When both the source and target were masked, particle count was reduced by more than 99.5% of baseline, regardless of separation distance or which type of mask was employed. CONCLUSION: These results support the principle of layered protection to mitigate transmission of SARS-CoV-2, the virus causing COVID-19, and other respiratory viruses and emphasize the importance of controlling the spread of aerosol at its source. The combination of masking and distancing reduced the exposure to exhaled particulates more than any individual measure. Combined measures remain the most effective way to combat the spread of respiratory infection.

Preparing for the next pandemic: It is more than just about numbers.

The COVID-19 pandemic has brought enormous hardships to our country and healthcare system. We present our experience navigating through this pandemic with emphasis on reactivating our practice while keeping patients and staff safe. It is hoped that the methods and thought processes provided in this manuscript will help those who are in various stages of managing their practice or provide lessons learned as our country eventually moves beyond this pandemic. Lastly, we aspire to provide a guide for those who are in a position to prepare for the next pandemic.

Noise Suppression for Ultrasound Attenuation Coefficient Estimation Based on Spectrum Normalization.

Ultrasound attenuation coefficient estimation (ACE) has great diagnostic potential for fatty liver detection and assessment. In a previous study, we proposed a reference phantom-free ACE method, called reference frequency method (RFM), which does not require a calibrated phantom for normalization. The power of each frequency component can be normalized by the power of an adjacent frequency component in the spectrum to cancel system-dependent effects such as focusing and time gain compensation (TGC). RFM demonstrated accurate ACE in both phantom and in in-vivo liver studies. However, our study also showed that the robustness and penetration of RFM were affected by noise in the ACE signals. Here we propose a noise suppression (NS) and a signal-to-noise ratio (SNR) quality control method to reduce the influence of noise on ACE-RFM performance. The proposed methods were tested in harmonic ACE because harmonic imaging is a more frequently used mode than fundamental imaging for abdominal applications. After applying the NS and SNR control methods, the noise-induced bias for attenuation estimation in harmonic ACE was effectively reduced, leading to significantly improved effective penetration depth. The proposed methods directly measure the noise spectrum of the ultrasound system, which can also be adapted to other spectrum-based ACE methods, such as the reference phantom method and the spectra shift method.