Medical image segmentation automatic quality control: A multi-dimensional approach

In clinical applications, using erroneous segmentations of medical images can have dramatic consequences. Current approaches dedicated to medical image segmentation automatic quality control do not predict segmentation quality at slice-level (2D), resulting in sub-optimal evaluations. Our 2D-based deep learning method simultaneously performs quality control at 2D-level and 3D-level for cardiovascular MR image segmentations. We compared it with 3D approaches by training both on 36,540 (2D) / 3842 (3D) samples to predict Dice Similarity Coefficients (DSC) for 4 different structures from the left ventricle, i.e., trabeculations (LVT), myocardium (LVM), papillary muscles (LVPM) and blood (LVC). The 2D-based method outperformed the 3D method. At the 2D-level, the mean absolute errors (MAEs) of the DSC predictions for 3823 samples, were 0.02, 0.02, 0.05 and 0.02 for LVM, LVC, LVT and LVPM, respectively. At the 3D-level, for 402 samples, the corresponding MAEs were 0.02, 0.01, 0.02 and 0.04. The method was validated in a clinical practice evaluation against semi-qualitative scores provided by expert cardiologists for 1016 subjects of the UK BioBank. Finally, we provided evidence that a multi-level QC could be used to enhance clinical measurements derived from image segmentations.

COVID-19 and Obesity: Role of Ectopic Visceral and Epicardial Adipose Tissues in Myocardial Injury

In March 2020, the WHO declared coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a global pandemic. Obesity was soon identified as a risk factor for poor prognosis, with an increased risk of intensive care admissions and mechanical ventilation, but also of adverse cardiovascular events. Obesity is associated with adipose tissue, chronic low-grade inflammation, and immune dysregulation with hypertrophy and hyperplasia of adipocytes and overexpression of pro-inflammatory cytokines. However, to implement appropriate therapeutic strategies, exact mechanisms must be clarified. The role of white visceral adipose tissue, increased in individuals with obesity, seems important, as a viral reservoir for SARS-CoV-2 via angiotensin-converting enzyme 2 (ACE2) receptors. After infection of host cells, the activation of pro-inflammatory cytokines creates a setting conducive to the "cytokine storm" and macrophage activation syndrome associated with progression to acute respiratory distress syndrome. In obesity, systemic viral spread, entry, and prolonged viral shedding in already inflamed adipose tissue may spur immune responses and subsequent amplification of a cytokine cascade, causing worse outcomes. More precisely, visceral adipose tissue, more than subcutaneous fat, could predict intensive care admission; and lower density of epicardial adipose tissue (EAT) could be associated with worse outcome. EAT, an ectopic adipose tissue that surrounds the myocardium, could fuel COVID-19-induced cardiac injury and myocarditis, and extensive pneumopathy, by strong expression of inflammatory mediators that could diffuse paracrinally through the vascular wall. The purpose of this review is to ascertain what mechanisms may be involved in unfavorable prognosis among COVID-19 patients with obesity, especially cardiovascular events, emphasizing the harmful role of excess ectopic adipose tissue, particularly EAT.

Percutaneous lung and liver CT-guided ablation on swine model using microwave ablation to determine ablation size for clinical practice

PURPOSE

Microwave ablation (MWA) provides an effective treatment of lung and liver tumors but suffers from a lack of reproducibility of ablation size among currently available technologies. In-vitro evaluations are far removed from clinical practices because of uninfused tissue. This study is in-vivo preclinical testing of a new MWA system on swine lungs and liver.

MATERIALS AND METHODS

All ablations were performed under CT guidance and multiple algorithms were tested with a power of 50, 75, and 100 W for durations of 3, 5, 8, 10, and 15 min. A 3 D-evaluation of the ablation zone was carried out using enhanced-CT. The sphericity index, coefficients of variation, and energy efficiency (which corresponds to the volume yield according to the power supplied) were calculated.

RESULTS

Fifty liver and 48 lung ablations were performed in 17 swine. The sphericity index varies from 0.50 to 0.80 for liver ablations and from 0.40 to 0.69 for lung ablations. The coefficient of variation was below 15% for 4/5 and 4/8 protocols for lung and liver ablations, respectively. The energy efficiency seems to decrease with the duration of the ablation from 0.60 × 10^-3 cm³/J (75 W, 3 min) to 0.26 × 10^-3 cm³/J (100 W, 15 min) in the liver and from 0.57 × 10^-3 cm³/J (50 W, 10 min) to 0.42 × 10^-3 cm³/J (100 W, 12 min) in the lungs.

CONCLUSION

A shorter treatment time provides the best energy efficiency, and the best reproducibility is obtained for a 10 min treatment duration. The system tested provides an interesting reproducibility in both lung and liver measurements. Our results may help interventional radiologists in the optimal selection of treatment parameters.

Description of morphological evolution of lung tumors treated by percutaneous radiofrequency ablation: long term follow-up of 100 lesions with chest CT

PURPOSE

Radiofrequency ablation (RFA) is a safe and effective minimally invasive treatment for pulmonary tumors. Patterns on chest computed tomography (CT) after RFA are classified into five types; however, the follow-up has not been fully described. The objectives of this study were to describe (1) the CT pattern 3 years after RFA and (2) its evolution over 7 years.

MATERIALS AND METHODS

Lesions treated with RFA between 2009 and 2017 and with ≥3 years of follow-up CT data were included. Lesions with local recurrences were excluded from the study. The morphology of the ablation zone was classified as nodular, fibrotic, atelectatic, cavitary, and disappeared. Other initial anatomical parameters were recorded. Kruskal-Wallis or Chi-square tests were used to compare the groups.

RESULTS

One hundred lung RFA scars were included, and a retrospective longitudinal study was performed. Three years after RFA, nodular, fibrotic, atelectatic, and cavitary scars, and disappearance were observed in 49%, 36%, 5%, 3%, and 6% of the scars, respectively. Evolution over 7 years showed that the fibrosis, atelectasis, and disappearance remained stable over time, whereas 28% of nodular scars evolved into fibrotic scars. Additionally, 45% of cavitary scars evolved into nodular scars. Pleural contact was associated with disappearance, and the use of a 20-mm needle was associated with atelectasis.

CONCLUSION

Follow-up after RFA showed that fibrosis, disappearance, and atelectasis remained stable over time. Nodular scars could evolve into fibrotic scars, and cavitary scars could evolve into nodular scars.

Deep Learning-based Automated Segmentation of Left Ventricular Trabeculations and Myocardium on Cardiac MR Images: A Feasibility Study

Purpose

To develop and evaluate a complete deep learning pipeline that allows fully automated end-diastolic left ventricle (LV) cardiac MRI segmentation, including trabeculations and automatic quality control of the predicted segmentation.

Materials and Methods

This multicenter retrospective study includes training, validation, and testing datasets of 272, 27, and 150 cardiac MR images, respectively, collected between 2012 and 2018. The reference standard was the manual segmentation of four LV anatomic structures performed on end-diastolic short-axis cine cardiac MRI: LV trabeculations, LV myocardium, LV papillary muscles, and the LV blood cavity. The automatic pipeline was composed of five steps with a DenseNet architecture. Intraobserver agreement, interobserver agreement, and interaction time were recorded. The analysis includes the correlation between the manual and automated segmentation, a reproducibility comparison, and Bland-Altman plots.

Results

The automated method achieved mean Dice coefficients of 0.96 ± 0.01 (standard deviation) for LV blood cavity, 0.89 ± 0.03 for LV myocardium, and 0.62 ± 0.08 for LV trabeculation (mean absolute error, 3.63 g ± 3.4). Automatic quantification of LV end-diastolic volume, LV myocardium mass, LV trabeculation, and trabeculation mass-to-total myocardial mass (TMM) ratio showed a significant correlation with the manual measures (r = 0.99, 0.99, 0.90, and 0.83, respectively; all P < .01). On a subset of 48 patients, the mean Dice value for LV trabeculation was 0.63 ± 0.10 or higher compared with the human interobserver (0.44 ± 0.09; P < .01) and intraobserver measures (0.58 ± 0.09; P < .01). Automatic quantification of the trabeculation mass-to-TMM ratio had a higher correlation (0.92) compared with the intra- and interobserver measures (0.74 and 0.39, respectively; both P < .01).

Conclusion

Automated deep learning framework can achieve reproducible and quality-controlled segmentation of cardiac trabeculations, outperforming inter- and intraobserver analyses.Supplemental material is available for this article.© RSNA, 2020.

T1-Based Synthetic Magnetic Resonance Contrasts Improve Multiple Sclerosis and Focal Epilepsy Imaging at 7 T

OBJECTIVES

Ultra-high field magnetic resonance imaging (MRI) (≥7 T) is a unique opportunity to improve the clinical diagnosis of brain pathologies, such as multiple sclerosis or focal epilepsy. However, several shortcomings of 7 T MRI, such as radiofrequency field inhomogeneities, could degrade image quality and hinder radiological interpretation. To address these challenges, an original synthetic MRI method based on T1 mapping achieved with the magnetization-prepared 2 rapid acquisition gradient echo (MP2RAGE) sequence was developed. The radiological quality of on-demand T1-based contrasts generated by this technique was evaluated in multiple sclerosis and focal epilepsy imaging at 7 T.

MATERIALS AND METHODS

This retrospective study was carried out from October 2017 to September 2019 and included 21 patients with different phenotypes of multiple sclerosis and 35 patients with focal epilepsy who underwent MRI brain examinations using a whole-body investigative 7 T magnetic resonance system. The quality of 2 proposed synthetic contrast images were assessed and compared with conventional images acquired at 7 T using the MP2RAGE sequence by 4 radiologists, evaluating 3 qualitative criteria: signal homogeneity, contrast intensity, and lesion visualization. Statistical analyses were performed on reported quality scores using Wilcoxon rank tests and further multiple comparisons tests. Intraobserver and interobserver reliabilities were calculated as well.

RESULTS

Radiological quality scores were reported higher for synthetic images when compared with original images, regardless of contrast, pathologies, or raters considered, with significant differences found for all 3 criteria (P < 0.0001, Wilcoxon rank test). None of the 4 radiologists ever rated a synthetic image "markedly worse" than an original image. Synthetic images were rated slightly less satisfying for only 3 epileptic patients, without precluding lesion identification.

CONCLUSION

T1-based synthetic MRI with the MP2RAGE sequence provided on-demand contrasts and high-quality images to the radiologist, facilitating lesion visualization in multiple sclerosis and focal epilepsy, while reducing the magnetic resonance examination total duration by removing an additional sequence.

Augmented Reality Guided Laparoscopic Surgery of the Uterus

A major research area in Computer Assisted Intervention (CAI) is to aid laparoscopic surgery teams with Augmented Reality (AR) guidance. This involves registering data from other modalities such as MR and fusing it with the laparoscopic video in real-time, to reveal the location of hidden critical structures. We present the first system for AR guided laparoscopic surgery of the uterus. This works with pre-operative MR or CT data and monocular laparoscopes, without requiring any additional interventional hardware such as optical trackers. We present novel and robust solutions to two main sub-problems: the initial registration, which is solved using a short exploratory video, and update registration, which is solved with real-time tracking-by-detection. These problems are challenging for the uterus because it is a weakly-textured, highly mobile organ that moves independently of surrounding structures. In the broader context, our system is the first that has successfully performed markerless real-time registration and AR of a mobile human organ with monocular laparoscopes in the OR.

Prevalence and risk factors for lung involvement on low-dose chest CT (LDCT) in a paucisymptomatic population of 247 patients affected by COVID-19

Background

Low-dose chest CT (LDCT) showed high sensitivity and ability to quantify lung involvement of COVID-19 pneumopathy. The aim of this study was to describe the prevalence and risk factors for lung involvement in 247 patients with a visual score and assess the prevalence of incidental findings.

Methods

For 12 days in March 2020, 250 patients with RT-PCR positive tests and who underwent LDCT were prospectively included. Clinical and imaging findings were recorded. The extent of lung involvement was quantified using a score ranging from 0 to 40. A logistic regression model was used to explore factors associated with a score ≥ 10.

Results

A total of 247 patients were analyzed; 138 (54%) showed lung involvement. The mean score was 4.5 ± 6.5, and the mean score for patients with lung involvement was 8.1 ± 6.8 [1–31]. The mean age was 43 ± 15 years, with 121 males (48%) and 17 asymptomatic patients (7%). Multivariate analysis showed that age > 54 years (odds ratio 4.4[2.0–9.6] p < 0.001) and diabetes (4.7[1.0–22.1] p = 0.049) were risk factors for a score ≥ 10. Multivariate analysis including symptoms showed that only age > 54 years (4.1[1.7–10.0] p = 0.002) was a risk factor for a score ≥ 10. Rhinitis (0.3[0.1–0.7] p = 0.005) and anosmia (0.3[0.1–0.9] p = 0.043) were protective against lung involvement. Incidental imaging findings were found in 19% of patients, with a need for follow-up in 0.6%.

Conclusion

The prevalence of lung involvement was 54% in a predominantly paucisymptomatic population. Age ≥ 55 years and diabetes were risk factors for significant parenchymal lung involvement. Rhinitis and anosmia were protective against LDCT abnormalities.

Low-dose chest CT for diagnosing and assessing the extent of lung involvement of SARS-CoV-2 pneumonia using a semi quantitative score

OBJECTIVES

The purpose is to assess the ability of low-dose CT (LDCT) to determine lung involvement in SARS-CoV-2 pneumonia and to describe a COVID19-LDCT severity score.

MATERIALS AND METHODS

Patients with SARS-CoV-2 infection confirmed by RT-PCR were retrospectively analysed. Clinical data, the National Early Warning Score (NEWS) and imaging features were recorded. Lung features included ground-glass opacities (GGO), areas of consolidation and crazy paving patterns. The COVID19-LDCT score was calculated by summing the score of each segment from 0 (no involvement) to 10 (severe impairment). Univariate analysis was performed to explore predictive factor of high COVID19-LDCT score. The nonparametric Mann-Whitney test was used to compare groups and a Spearman correlation used with p<0.05 for significance.

RESULTS

Eighty patients with positive RT-PCR were analysed. The mean age was 55 years ± 16, with 42 males (53%). The most frequent symptoms were fever (60/80, 75%) and cough (59/80, 74%), the mean NEWS was 1.7±2.3. All LDCT could be analysed and 23/80 (28%) were normal. The major imaging finding was GGOs in 56 cases (67%). The COVID19-LDCT score (mean value = 19±29) was correlated with NEWS (r = 0.48, p<0.0001). No symptoms were risk factor to have pulmonary involvement. Univariate analysis shown that dyspnea, high respiratory rate, hypertension and diabetes are associated to a COVID19-LDCT score superior to 50.

CONCLUSIONS

COVID19-LDCT score did correlate with NEWS. It was significantly different in the clinical low-risk and high-risk groups. Further work is needed to validate the COVID19-LDCT score against patient prognosis.

Joint Position Paper of the Working Group of Pacing and Electrophysiology of the French Society of Cardiology (SFC) and the Société française d'imagerie cardiaque et vasculaire diagnostique et interventionnelle (SFICV) on magnetic resonance imaging in patients with cardiac electronic implantable devices

Magnetic resonance imaging (MRI) has become the reference imaging for the management of a large number of diseases. The number of MR examinations increases every year, simultaneously with the number of patients receiving a cardiac electronic implantable device (CEID). A CEID was considered an absolute contraindication for MRI for years. The progressive replacement of conventional pacemakers and defibrillators by MR-conditional CEIDs and recent data on the safety of MRI in patients with "MR-nonconditional" CEIDs have progressively increased the demand for MRI in patients with a CEID. However, some risks are associated with MRI in CEID carriers, even with "MR-conditional" devices because these devices are not "MR-safe". A specific programing of the device in "MR-mode" and monitoring patients during MRI remain mandatory for all patients with a CEID. A standardized patient workflow based on an institutional protocol should be established in each institution performing such examinations. This joint position paper of the Working Group of Pacing and Electrophysiology of the French Society of Cardiology and the Société française d'imagerie cardiaque et vasculaire diagnostique et interventionnelle (SFICV) describes the effect and risks associated with MRI in CEID carriers. We propose recommendations for patient workflow and monitoring and CEID programming in MR-conditional, "MR-conditional nonguaranteed" and MR-nonconditional devices.

Quantification of diffuse myocardial fibrosis using CMR extracellular volume fraction and serum biomarkers of collagen turnover with histologic quantification as standard of reference

PURPOSE

To compare the assessment of diffuse interstitial myocardial fibrosis in valvular diseases using cardiac magnetic resonance (CMR) extracellular volume fraction (ECV) quantification and serum biomarkers of collagen turnover using results of myocardial biopsy as standard of reference.

MATERIALS AND METHODS

This prospective monocentric study included consecutive patients before aortic valvular replacement. All patients underwent: i), 1.5T CMR with pre and post contrast T₁ mapping sequence and ECV computation; ii), serum quantification of matrix metalloproteinases (MMPs) and tissue inhibitor of metalloproteinases (TIMPs) and iii), myocardial biopsies were collected during surgery to assess collagen volume fraction (CVF). Patients with coronary artery disease were excluded. Correlation between native T1, ECV, CVF and serum biomarkers were assessed using Pearson correlation test. Agreement between basal anteroseptal ECV with global ECV was assessed using Bland-Altman test.

RESULTS

Twenty-one patients, 16 with aortic stenosis and 5 with aortic regurgitation were included. There were 12 men and 9 women with a mean age of 74.1±6.8 (SD) years (range: 32-84 years). Mean global ECV value was 26.7±2.7 (SD) % (range: 23.4-32.5%) and mean CVF value was 12.4±9.7% (range: 3.2-25.7%). ECV assessed at the basal anteroseptal segment correlated moderately with CVF (r=0.6; P=0.0026). There was a strong correlation and agreement between basal anteroseptal ECV and global ECV, (r=0.8; P<0.0001; bias 5.4±6.1%) but no correlation between global ECV and CVF (r=0.5; P=0.10). Global ECV poorly correlated with serum TIMP-1 (r=0.4; P=0.037) and MMP-2 (r=0.4; P=0.047). No correlation was found between serum biomarkers and basal anteroseptal- ECV or native T1.

CONCLUSION

In patients with severe aortic valvulopathy, diffuse myocardial fibrosis assessed by anterosepto-basal ECV correlates with histological myocardial fibrosis. Anteroseptobasal ECV strongly correlates with global ECV, which poorly correlates with TIMP-1 and MMP-2, serum biomarkers involved in the progression of heart failure.

[Central venous access in interventional radiology]

Some patients require iteratives intravenous administrations on a central catheter, for example in oncology or infectiology, which represents a challenge for ambulatory treatment. Interventional radiology could provide solutions with the implant and monitoring of PICC-lines and ports. These are implanted in sterile environment and under imaging guidance in an interventional radiology room by an operator and with a paramedical team that need to be experienced. This development focus on the interest of one method with respect to the other, as well as the differents ways to do, the complications that could arise and the monitoring of these devices.

Tipificación en <em>Gutierrezia</em> (Asteraceae, Astereae)

Durante el transcurso de la revisión de las especies sudamericanas de Gutierrezia, hemos podido hallar nombres que requieren tipificación. Luego del análisis de los protologos y especímenes de herbario, hemos designado ocho lectótipos y un neótipo para nueve nombres del género Gutierrezia. Además, se asignó la categoría de hólotipo a un especimen. Se provee una discusión detallada para cada nombre.

Adrenal hemorrhage in newborn: how, when and why- from case report to literature review

Background

Neonatal adrenal hemorrhage is a relatively uncommon condition (0.2–0.55%). Various risk factors have been reported in addition to birth asphyxia, such as sepsis, coagulation disorders, traumatic delivery, and perinatal injuries. Adrenal hemorrhage usually affects the right adrenal gland (about 70% of cases) while it involves the bilateral adrenal gland only in 10% of cases. In most cases, the event is asymptomatic but, in others, it may be so devastating to determine death by bleeding or adrenal insufficiency.

Case presentation

A case of bilateral neonatal adrenal hemorrhage, with adrenal insufficiency, but with no important risk factors and favorable evolution in a male infant.

Conclusions

This case emphasizes the importance of keeping a non-interventional attitude, avoiding early surgery but carrying out a serial sonographic follow-up. Serial ultrasound monitoring is the most reliable approach during conservative management.

Electronic supplementary material

The online version of this article (10.1186/s13052-019-0651-9) contains supplementary material, which is available to authorized users.

Exercise stress CMR reveals reduced aortic distensibility and impaired right-ventricular adaptation to exercise in patients with repaired tetralogy of Fallot

BACKGROUND

The aim of our study was to evaluate the feasibility of exercise cardiac magnetic resonance (CMR) in patients with repaired tetralogy of Fallot (RTOF) and to assess right and left ventricular adaptation and aortic wall response to exercise in comparison with volunteers.

METHODS

11 RTOF and 11 volunteers underwent prospective CMR at rest and during exercise. A supine bicycle ergometer was employed to reach twice the resting heart rate during continuous exercise, blood pressure and heart rate were recorded. Bi-ventricular parameters and aortic stiffness were assessed using accelerated cine sequences and flow-encoding CMR. A t-test was used to compare values between groups. A Mann Whitney test was used to compare values within groups.

RESULTS

In RTOF both ventricles showed an impaired contractile reserve (RVEF rest 36.2±8.3%, +1.3±3.9% increase after exercise; LVEF rest 53.8±6.1%, +5.7±6.4% increase after exercise) compared to volunteers (RVEF rest 50.5±5.0%, +10.4±7.1% increase after exercise, p = 0.039; LVEF rest 61.9±3.1%, +12.2±4.7% increase after exercise, p = 0.014). RTOF showed a reduced distensibility of the ascending aorta during exercise compared to volunteers (RTOF: 3.4±1.9 10-3.mmHg-1 vs volunteers: 5.1±1.4 10-3.mmHg-1; p = 0.027). Ascending aorta distensibility was correlated to cardiac work in the volunteers but not in RTOF.

CONCLUSION

RTOF showed an impaired contractile reserve for both ventricles. The exercise unmasked a reduced distensibility of the ascending aorta in RTOF, which may be an early sign of increased aortic rigidity.