Semiautomated segmentation of hepatocellular carcinoma tumors with MRI using convolutional neural networks.

OBJECTIVE: To assess the performance of convolutional neural networks (CNNs) for semiautomated segmentation of hepatocellular carcinoma (HCC) tumors on MRI. METHODS: This retrospective single-center study included 292 patients (237 M/55F, mean age 61 years) with pathologically confirmed HCC between 08/2015 and 06/2019 and who underwent MRI before surgery. The dataset was randomly divided into training (n = 195), validation (n = 66), and test sets (n = 31). Volumes of interest (VOIs) were manually placed on index lesions by 3 independent radiologists on different sequences (T2-weighted imaging [WI], T1WI pre-and post-contrast on arterial [AP], portal venous [PVP], delayed [DP, 3 min post-contrast] and hepatobiliary phases [HBP, when using gadoxetate], and diffusion-weighted imaging [DWI]). Manual segmentation was used as ground truth to train and validate a CNN-based pipeline. For semiautomated segmentation of tumors, we selected a random pixel inside the VOI, and the CNN provided two outputs: single slice and volumetric outputs. Segmentation performance and inter-observer agreement were analyzed using the 3D Dice similarity coefficient (DSC). RESULTS: A total of 261 HCCs were segmented on the training/validation sets, and 31 on the test set. The median lesion size was 3.0 cm (IQR 2.0-5.2 cm). Mean DSC (test set) varied depending on the MRI sequence with a range between 0.442 (ADC) and 0.778 (high b-value DWI) for single-slice segmentation; and between 0.305 (ADC) and 0.667 (T1WI pre) for volumetric-segmentation. Comparison between the two models showed better performance in single-slice segmentation, with statistical significance on T2WI, T1WI-PVP, DWI, and ADC. Inter-observer reproducibility of segmentation analysis showed a mean DSC of 0.71 in lesions between 1 and 2 cm, 0.85 in lesions between 2 and 5 cm, and 0.82 in lesions > 5 cm. CONCLUSION: CNN models have fair to good performance for semiautomated HCC segmentation, depending on the sequence and tumor size, with better performance for the single-slice approach. Refinement of volumetric approaches is needed in future studies. KEY POINTS: • Semiautomated single-slice and volumetric segmentation using convolutional neural networks (CNNs) models provided fair to good performance for hepatocellular carcinoma segmentation on MRI. • CNN models' performance for HCC segmentation accuracy depends on the MRI sequence and tumor size, with the best results on diffusion-weighted imaging and T1-weighted imaging pre-contrast, and for larger lesions.

Third Ventricular Subependymomas: Clinical Features and Outcomes Over Two Decades.

BACKGROUND: Subependymomas are uncommon, benign slow-growing neoplasms of the central nervous system preferentially arising within the fourth and lateral ventricles. Third ventricle involvement has been described rarely. The aim of this study is to provide the first systematic review of third ventricular subependymomas (TVSE) by analyzing all reported cases over 2 decades and describing a case example. METHODS: MEDLINE and Embase databases were searched for the 20 years ending January 1, 2022, using relevant MeSH and non-MeSH terms, including "subependymoma" and "third ventricle." Methodology followed PRISMA guidelines. RESULTS: Of 804 identified studies, 131 met inclusion eligibility. The literature yielded 17 patients with TVSE plus our example (18 total). Of these patients, 83% (15/18) presented in adulthood (average age, 42 ± 19 years), of whom 73% were women. The pediatric cohort age was 5 ± 1 years, 67% (4/6) of whom were girls. The most common presenting symptom in both cohorts was headache (80%), followed by memory disturbances and vomitus. In adults, symptomatic tumors were approached by open craniotomy in all but 1 case, most using a transcallosal approach. Gross total resection was obtained in 73%. A ventriculoperitoneal shunt was inserted in 2/15 adult and 4/6 pediatric patients. Overall, both cohorts showed symptomatic improvement without disease recurrence. One patient died perioperatively. CONCLUSIONS: Subependymomas should be considered in the differential diagnosis of third ventricular tumors. The clinical presentation of TVSE mainly parallels hydrocephalus symptoms and, hence, awareness is of vital importance for timely treatment. The surgical goal should be gross total resection, which can be curative and offers greatest clinical improvement across the population.

Idiopathic membranous occlusion of the foramen of Monro: an unusual cause of hydrocephalus and headache in adults.

Occlusion of foramen of Monro is an uncommon clinical entity that usually presents in children. Common causes are obstructing mass, infectious etiologies or vascular malformation. Rarely, it may be an idiopathic stricture or membrane. We report a case of idiopathic membranous obstruction of the foramen of Monro in a 45-year-old male with no past medical or surgical history. He presented with new intermittent dull and burning bifrontal severe headache for 2 d, which was alleviated slightly by non-steroidal anti-inflammatory medication. Imaging showed marked dilation of the lateral ventricles with normal third and fourth ventricles. The patient was discharged initially with conservative medical management and close follows up; however, the headache continued to progress and neurosurgical intervention was offered. The patient underwent endoscopic exploration, fenestration of the septum pellucidum, and right ventriculoperitoneal shunt placement. Bilateral membranous obstruction of foramina of Monro and an auto-fenestrated cavum septum pellucidum were identified intraoperatively. The patient reported resolution of headache post-operatively without recurrence on 1-month follow up. This case is unusual in that the patient presented without any known neurologic history or prior intracranial infections. It became apparent at the time of surgery that chronic obstruction of the bilateral foramina with collapse of the third ventricle had developed, and the safest durable treatment for him was septostomy and cerebrospinal fluid (CSF) shunt placement.

Gadoxetate-enhanced abbreviated MRI is highly accurate for hepatocellular carcinoma screening.

OBJECTIVES: The primary objective was to compare the performance of 3 different abbreviated MRI (AMRI) sets extracted from a complete gadoxetate-enhanced MRI obtained for hepatocellular carcinoma (HCC) screening. Secondary objective was to perform a preliminary cost-effectiveness analysis, comparing each AMRI set to published ultrasound performance for HCC screening in the USA. METHODS: This retrospective study included 237 consecutive patients (M/F, 146/91; mean age, 58 years) with chronic liver disease who underwent a complete gadoxetate-enhanced MRI for HCC screening in 2017 in a single institution. Two radiologists independently reviewed 3 AMRI sets extracted from the complete exam: non-contrast (NC-AMRI: T2-weighted imaging (T2wi)+diffusion-weighted imaging (DWI)), dynamic-AMRI (Dyn-AMRI: T2wi+DWI+dynamic T1wi), and hepatobiliary phase AMRI (HBP-AMRI: T2wi+DWI+T1wi during the HBP). Each patient was classified as HCC-positive/HCC-negative based on the reference standard, which consisted in all available patient data. Diagnostic performance for HCC detection was compared between sets. Estimated set characteristics, including historical ultrasound data, were incorporated into a microsimulation model for cost-effectiveness analysis. RESULTS: The reference standard identified 13/237 patients with HCC (prevalence, 5.5%; mean size, 33.7 ± 30 mm). Pooled sensitivities were 61.5% for NC-AMRI (95% confidence intervals, 34.4-83%), 84.6% for Dyn-AMRI (60.8-95.1%), and 80.8% for HBP-AMRI (53.6-93.9%), without difference between sets (p range, 0.06-0.16). Pooled specificities were 95.5% (92.4-97.4%), 99.8% (98.4-100%), and 94.9% (91.6-96.9%), respectively, with a significant difference between Dyn-AMRI and the other sets (p < 0.01). All AMRI methods were effective compared with ultrasound, with life-year gain of 3-12 months against incremental costs of US$ < 12,000. CONCLUSIONS: NC-AMRI has limited sensitivity for HCC detection, while HBP-AMRI and Dyn-AMRI showed excellent sensitivity and specificity, the latter being slightly higher for Dyn-AMRI. Cost-effectiveness estimates showed that AMRI is effective compared with ultrasound. KEY POINTS: • Comparison of different abbreviated MRI (AMRI) sets reconstructed from a complete gadoxetate MRI demonstrated that non-contrast AMRI has low sensitivity (61.5%) compared with contrast-enhanced AMRI (80.8% for hepatobiliary phase AMRI and 84.6% for dynamic AMRI), with all sets having high specificity. • Non-contrast and hepatobiliary phase AMRI can be performed in less than 14 min (including set-up time), while dynamic AMRI can be performed in less than 17 min. • All AMRI sets were cost-effective for HCC screening in at-risk population in comparison with ultrasound.

Automated measurement of liver attenuation to identify moderate-to-severe hepatic steatosis from chest CT scans.

PURPOSE: Develop and validate an automated method for measuring liver attenuation in non-contrast low-dose chest CT (LDCT) scans and compare it to the standard manual method for identifying moderate-to-severe hepatic steatosis (HS). METHOD: The automated method identifies a region below the right lung within the liver and uses statistical sampling techniques to exclude non-liver parenchyma. The method was used to assess moderate-to-severe HS on two IRB-approved cohorts: 1) 24 patients with liver disease examined between 1/2013-1/2017 with non-contrast chest CT and abdominal MRI scans obtained within three months of liver biopsy, and 2) 319 lung screening participants with baseline LDCT performed between 8/2011-1/2017. Agreement between the manual and automated CT methods, the manual MRI method, and pathology for determining moderate-to-severe HS was assessed using Cohen's Kappa by applying a 40 HU threshold to the CT method and 17.4% fat fraction to MRI. Agreement between the manual and automated CT methods was assessed using the intraclass correlation coefficient (ICC). Variability was assessed using Bland-Altman limits of agreement (LoA). RESULTS: In the first cohort, the manual and automated CT methods had almost perfect agreement (ICC = 0.97, κ = 1.00) with LoA of -7.6 to 4.7 HU. Both manual and automated CT methods had almost perfect agreement with MRI (κ = 0.90) and substantial agreement with pathology (κ = 0.77). In the second cohort, the manual and automated CT methods had almost perfect agreement (ICC = 0.94, κ = 0.87). LoA were -10.6 to 5.2 HU. CONCLUSION: Automated measurements of liver attenuation from LDCT scans can be used to identify moderate-to-severe HS on LDCT.