Performance Evaluation of a Supervised Machine Learning Pain Classification Model Developed by Neonatal Nurses.

BACKGROUND: Early-life pain is associated with adverse neurodevelopmental consequences; and current pain assessment practices are discontinuous, inconsistent, and highly dependent on nurses' availability. Furthermore, facial expressions in commonly used pain assessment tools are not associated with brain-based evidence of pain. PURPOSE: To develop and validate a machine learning (ML) model to classify pain. METHODS: In this retrospective validation study, using a human-centered design for Embedded Machine Learning Solutions approach and the Neonatal Facial Coding System (NFCS), 6 experienced neonatal intensive care unit (NICU) nurses labeled data from randomly assigned iCOPEvid (infant Classification Of Pain Expression video) sequences of 49 neonates undergoing heel lance. NFCS is the only observational pain assessment tool associated with brain-based evidence of pain. A standard 70% training and 30% testing split of the data was used to train and test several ML models. NICU nurses' interrater reliability was evaluated, and NICU nurses' area under the receiver operating characteristic curve (AUC) was compared with the ML models' AUC. RESULTS: Nurses weighted mean interrater reliability was 68% (63%-79%) for NFCS tasks, 77.7% (74%-83%) for pain intensity, and 48.6% (15%-59%) for frame and 78.4% (64%-100%) for video pain classification, with AUC of 0.68. The best performing ML model had 97.7% precision, 98% accuracy, 98.5% recall, and AUC of 0.98. IMPLICATIONS FOR PRACTICE AND RESEARCH: The pain classification ML model AUC far exceeded that of NICU nurses for identifying neonatal pain. These findings will inform the development of a continuous, unbiased, brain-based, nurse-in-the-loop Pain Recognition Automated Monitoring System (PRAMS) for neonates and infants.

Same-day 90Y radioembolization: implementing a new treatment paradigm.

PURPOSE: To assess the feasibility of conducting pretreatment mesenteric angiography, coil embolization, 99mTc macroaggregated albumin (99mTc-MAA) scintigraphy, and 90Y radioembolization treatment in a single, same-day, combined outpatient encounter. METHODS: This was a retrospective study of 78 patients treated during the period 2008 - 2015 who were managed in a single outpatient encounter under the guidance of the Interventional Radiology Department and The Nuclear Medicine Department. Pretreatment planning was performed by reviewing baseline imaging and estimated perfused liver volume bearing the tumor. The region of interest was estimated using 3-D software; this value was used for dosimetry planning. Maximum lung shunting fractions of 10 % for hepatocellular carcinoma and 5 % for liver metastases were assumed. Subsequently, hepatic angiography and 99mTc-MAA scintigraphy were performed followed by 90Y treatment in one outpatient encounter. Total in-room procedure time was recorded. RESULTS: All patients underwent same-day angiography, 99mTc-MAA scintigraphy and 90Y radioembolization. Of the 78 patients, 16 received multiple segmental treatments to both lobes, 44 received treatment to the right lobe, and 18 received treatment to the left lobe. The median dose was 106 Gy. The median number of 90Y vials needed was two (range one to six). The median in-room time was 160 min (75 - 250 min). The residential status of the patients was as follows, 18 % (14/78) were local residents, 55 % (43/78) traveled from outside the city limits, 18 % (14/78) were from out-of-state, and 9 % (7/78) were resident abroad. Of the 78 patients, 61 (77 %) had hepatocellular carcinoma, and 17 (22 %) had liver metastases. The median lung dose was 3.5 Gy. CONCLUSION: This study demonstrated the feasibility of same-day 90Y evaluation and treatment while maintaining the principles of safe and effective 90Y infusion including tumoricidal dosimetry (lobar, segmentectomy), minimization of nontarget flow, and minimization of lung dose. This paradigm translates into expeditious cancer care and significant cost savings.

Types of Research Bias Encountered in IR.

Bias is a systemic error in studies that leads to inaccurate deductions. Relevant biases in the field of IR and interventional oncology were identified after reviewing articles published in the Journal of Vascular and Interventional Radiology and CardioVascular and Interventional Radiology. Biases cited in these articles were divided into three categories: preinterventional (health care access, participation, referral, and sample biases), periinterventional (contamination, investigator, and operator biases), and postinterventional (guarantee-time, lead time, loss to follow-up, recall, and reporting biases).

MR imaging findings of the prostate gland following prostate artery embolization: results from a prospective phase 2 study.

PURPOSE: To assess changes in imaging and volume characteristics of the prostate gland by magnetic resonance (MR) following prostatic artery embolization (PAE) for benign prostate hyperplasia. METHODS: With IRB approval, we analyzed prospectively acquired MR data of PAE patients at baseline and 6-month following treatment from 2015 to 2017. We reviewed prostate MRs looking for sequelae of embolization [changes in signal intensity and/or enhancement, infection/inflammation, infarction, edema, and change in intravesical prostatic protrusion (IPP)]. We calculated the total volume (TV) and central gland volumes (CGV) using DynaCAD® and measured change in volumes. Analyses were performed using SPSS with p < 0.05 considered significant. RESULTS: Forty-three patients (n = 43) met our inclusion criteria. 93% (30/43) and 100% (43/43) showed a decrease in TV and CGV at 6-months respectively. At baseline, median TV was 86 cc (range 29.4-232) and median CGV was 54.4 cc (range 12.9-165.5). Median decrease in TV was 18.2% (CI 13.3-27.2) (p = 0.0001) and median decrease in CGV was 26.7% (CI 20.4-35.9) (p = 0.0001). Thirty-seven percent (16/43) of patients had IPP at baseline; 100% showed a decrease in size of median lobe at follow-up. At 6-month follow-up, 33% (14/43) showed imaging features of infarction, 79% (34/43) had decrease in T2-signal intensity, and 51% (22/43) showed a decrease in enhancement. None had edema, peri-prostatic fat changes or infection/inflammation. CONCLUSION: PAE causes a statistically significant reduction in the TV and CGV. There is also a reduction of the degree of IPP. Non-specific findings of infarction, decrease in T2-signal, and enhancement were also seen.

90Y Radioembolization for Locally Advanced Hepatocellular Carcinoma with Portal Vein Thrombosis: Long-Term Outcomes in a 185-Patient Cohort.

We report survival outcomes for patients with advanced-stage hepatocellular carcinoma (HCC) with portal vein thrombosis (PVT) treated with 90Y radioembolization. Methods: With institutional review board approval, we searched our prospectively acquired database for 90Y patients treated between 2003 and 2017. Inclusion criteria were patients who had HCC with tumor PVT. Patients with metastases were excluded. Laboratory data were collected at baseline and 1 mo after 90Y radioembolization. Toxicity grades were reported according to the Common Terminology Criteria for Adverse Events, version 4.0, and long-term survival outcomes were reported and stratified by Child-Pugh class (CP). Overall survival was calculated using the Kaplan-Meier method. Multivariate analysis was performed using Cox proportional hazards regression. A subanalysis for patients with a high level of α-fetoprotein (AFP) (>100 ng/dL) was conducted. Results: In total, 185 patients with HCC PVT underwent 90Y radioembolization. Seventy-four (40%) were CP-A, 51 (28%) were CP-B7, and 60 (32%) were ≥CP-B8. New albumin, bilirubin, and alkaline phosphatase grade 3/4 toxicities were, respectively, 3%, 10%, and 0% for CP-A; 14%, 12%, and 6% for CP-B7; and 23%, 32%, and 3% for ≥CP-B8. Median overall survival for CP-A patients was 13.3 mo (95% confidence interval [CI], 8.7-15.7 mo). CP-B7 and ≥CP-B8 patients exhibited median overall survival of 6.9 mo (95% CI, 5.3-10.1 mo) and 3.9 mo (95% CI, 2.9-5.0 mo), respectively. Significant overall survival prognosticators on univariate analysis were albumin, bilirubin, ascites, tumor size 5 cm or smaller, focality, distribution, infiltration, Eastern Cooperative Oncology Group status, AFP level, and PVT extent. Multivariate analysis showed the prognosticators of overall survival to be bilirubin, no ascites, tumor size 5 cm or smaller, solitary lesion, baseline AFP level lower than 100 ng/dL, and Eastern Cooperative Oncology Group status. Of 123 patients with a high AFP level (>100 ng/dL), 12 patients achieved restored normal AFP levels (<13 ng/dL) and exhibited median overall survival of 23.9 mo (95% CI, 20.1-124.1 mo). AFP responders at 1 mo had better overall survival than nonresponders, at 8.5 mo versus 4.8 mo (P = 0.018); AFP responders at 3 mo had overall survival of 13.3 mo, versus 6.9 mo for nonresponders (P = 0.021). Conclusion:90Y radioembolization can serve as a safe and effective treatment for advanced-stage HCC patients with tumor PVT. Overall survival outcomes are affected by baseline liver function, tumor size, and AFP level.

Pictorial essay: imaging findings following Y90 radiation segmentectomy for hepatocellular carcinoma.

Transarterial radioembolization is a novel therapy that has gained rapid clinical acceptance for the treatment of hepatocellular carcinoma (HCC). Segmental radioembolization [also termed radiation segmentectomy (RS)] is a technique that can deliver high doses (> 190 Gy) of radiation selectively to the hepatic segment(s) containing the tumor. The aim of this comprehensive review is to provide an illustrative summary of the most relevant imaging findings encountered after radiation segmentectomy. A 62-patient cohort of Child-Pugh A patients with solitary HCC < 5 cm in size was identified. A comprehensive retrospective imaging review was done by interventional radiology staff at our institution. Important imaging findings were reported and illustrated in a descriptive account. For the purposes of completeness, specific patients outside our initial cohort with unique educational imaging features that also underwent segmentectomy were included in this pictorial essay. This review shows that response assessment after RS requires a learning curve with common drawbacks that can lead to false-positive interpretations and secondary unnecessary treatments. It is important to recognize that treatment responses and pathological changes both are time dependent. Findings such as benign geographical enhancement and initial benign pathological enhancement can easily be misinterpreted. Capsular retraction and segmental atrophy are some other examples of unique post-RS response that are not seen in any other treatment.

Comparison of the Adverse Event Profile of TheraSphere® with SIR-Spheres® for the Treatment of Unresectable Hepatocellular Carcinoma: A Systematic Review.

To compare the safety profiles of TheraSphere® (glass) and SIR-Spheres® (resin) Y90 microspheres for the treatment of hepatocellular carcinoma. A systematic review was conducted using the databases MEDLINE, Embase, and Cochrane Trials Register to identify all relevant studies. Baseline characteristics and adverse events of all grades related to gastrointestinal, hepatobiliary, and respiratory systems were collected along with commonly reported outcomes related to post-embolization syndrome. For all outcomes, data from each study were tabulated for each intervention. Adverse events and patients were summed across studies on TheraSphere® and SIR-Spheres®, respectively, and the resulting proportion of patients experiencing an outcome for both interventions was calculated. Thirty-one observational studies were included in the review. In the adverse events of all grades, more patients treated with resin microspheres reported gastric ulcers, hepatic encephalopathy, cholecystitis, hepatic failure, and pleural effusion. Patients treated with resin microspheres also had more hepatobiliary adverse events of grade 3 or higher. In the events related to post-embolization syndrome, glass microspheres exhibited a similar safety profile compared to resin microspheres. Ascites and nausea grade 3 or higher were recorded more frequently with glass microsphere treatment. Based on this review of the published literature, glass microspheres exhibit a safety profile with fewer gastrointestinal and pulmonary adverse events compared to resin microspheres in the treatment of hepatocellular carcinoma.

Indicators of Lung Shunt Fraction Determined by Technetium-99 m Macroaggregated Albumin in Patients with Hepatocellular Carcinoma.

PURPOSE: To determine the correlation of pre-procedural and imaging characteristics with lung shunt fraction (LSF) measured by technetium-99 m macroaggregated albumin (99mTc-MAA) scan in patients with hepatocellular carcinoma. METHODS: A retrospective study was conducted of 428 subjects with hepatocellular carcinoma from 2004 to 2011 assessed for lung shunting by 99mTc-MAA scan. Baseline characteristics included age, gender, ethnicity, tumor burden, maximum dimension, number of lesions, presence of extrahepatic metastases, macrovascular (hepatic and portal vein) invasion, ascites on imaging, laboratory values, and alpha-fetoprotein (AFP). Univariate and multivariate logistic regression analysis was performed. Receiver operating characteristic curves were used to obtain sensitivity (SN), specificity (SP), and positive likelihood ratios (LR+) of characteristics for low LSF (LSF <10%) and high LSF (LSF >20%). RESULTS: Statistically significant (p < 0.05) independent indicators of low LSF included bilirubin <1.45 mg/dL (SN = 49.5%, SP = 69.1%, LR+ = 1.60), maximum tumor size <7.15 cm (SN = 66.0%, SP = 75.9%, LR+ = 2.74), AFP ≤200 ng/mL (SN = 64.6%, SP = 65.0%, LR+ = 1.85), and absent macrovascular invasion (SN = 73.9%, SP = 64.9%, LR+ = 2.11). Independent indicators of high LSF included albumin <2.65 g/dL (SN = 64.3%, SP = 64.1%, LR+ = 1.79) and macrovascular invasion (SN = 74.4%, SP = 67.4%, LR+ = 2.28). A combined risk factor model was constructed. If there is no macrovascular invasion: [Formula: see text]. With macrovascular invasion, [Formula: see text] (R 2 = 0.257). Since these factors all have LR+ between 2 and 5, they only reflect slight increase in LSF predictivity. CONCLUSION: Serum AFP, albumin, bilirubin, and portal/hepatic vein invasion on cross-sectional imaging are statistically significant but weak clinical indicators of LSF, as shown by low SN, SP, and LR+ for clinically relevant cutoff LSF values. Thus, these factors cannot be relied upon in clinical practice.

Transarterial Radioembolization with Yttrium-90 for the Treatment of Hepatocellular Carcinoma.

BACKGROUND: Hepatocellular carcinoma (HCC) is a common cause of worldwide mortality. Transarterial radioembolization (TARE) with yttrium-90 (Y90), a transcatheter intra-arterial procedure performed by interventional radiology, has become widely utilized in managing HCC. METHODS: The following is a focused review of TARE covering its commercially available products, clinical considerations of treatment, salient clinical trial data establishing its utility, and the current and future roles of TARE in the management of HCC. RESULTS: TARE is indicated for patients with unresectable, intermediate stage HCC. The two available products are glass and resin microspheres. All patients undergoing TARE must be assessed with a history, physical examination, clinical laboratory tests, imaging, and arteriography with macroaggregated albumin. TARE is safe and effective in the treatment of unresectable HCC, as it has a safer toxicity profile than chemoembolization, longer time-to-progression, greater ability to downsize and/or bridge patients to liver transplant, and utility in tumor complicated by portal vein thrombosis. TARE can also serve as an alternative to ablation and chemotherapy. CONCLUSION: TARE assumes an integral role in the management of unresectable HCC and has been validated by numerous studies.

Hepatic imaging following intra-arterial embolotherapy.

PURPOSE: To discuss guidelines and salient imaging findings of solid tumors treated with common intra-arterial procedures used in interventional oncology. METHODS: A meticulous literature search of PubMed-indexed articles was conducted. Key words included "imaging + embolization," "imaging + TACE," "imaging + radioembolization," "imaging + Y90," "mRECIST," and "EASL." Representative post-treatment cross-sectional images were obtained from past cases in this institution. RESULTS: Intra-arterial therapy (IAT) in interventional oncology includes bland embolization, chemoembolization, and radioembolization. Solid tumors of the liver are the primary focus of these procedures. Cross-sectional CT and/or MR are the main modalities used to image tumors after treatment. Traditional size-based response criteria (WHO and RECIST) alone are of limited utility in determining response to IAT; tumoral necrosis and enhancement must be considered. Specifically for HCC, the EASL and mRECIST guidelines are becoming widely adopted response criteria to assess these factors. DWI, FDG-PET, and CEUS are modalities that play an adjunctive but controversial role. CONCLUSIONS: Radiologists must be aware that the different forms of intra-arterial therapy yield characteristic findings on cross-sectional imaging. Knowledge of these findings is integral to accurate assessment of tumor response and progression.