Deep-learning single-shot detector for automatic detection of brain metastases with the combined use of contrast-enhanced and non-enhanced computed tomography images

PURPOSE

To develop a deep-learning object detection model for automatic detection of brain metastases that simultaneously uses contrast-enhanced and non-enhanced images as inputs, and to compare its performance with that of a model that uses only contrast-enhanced images.

METHOD

A total of 116 computed tomography (CT) scans of 116 patients with brain metastases were included in this study. They showed a total of 659 metastases, 428 of which were used for training and validation (mean size, 11.3 ± 9.9 mm) and 231 were used for testing (mean size, 9.0 ± 7.0 mm). Single-shot detector (SSD) models were constructed with a feature fusion module, and their results were compared per lesion at a confidence threshold of 50%.

RESULTS

The sensitivity was 88.7% for the model that used both contrast-enhanced and non-enhanced CT images (the CE + NECT model) and 87.6% for the model that used only contrast-enhanced CT images (the CECT model). The positive predictive value (PPV) was 44.0% for the CE + NECT model and 37.2% for the CECT model. The number of false positives per patient was 9.9 for the CE + NECT model and 13.6 for the CECT model. The CE + NECT model had a significantly higher PPV (t test, p < 0.001), significantly fewer false positives (t test, p < 0.001), and a tendency to be more sensitive (t test, p = 0.14).

CONCLUSIONS

The results indicate that the information on true contrast enhancement obtained by comparing the contrast-enhanced and non-enhanced images may prevent the detection of pseudolesions, suppress false positives, and improve the performance of deep-learning object detection models.

Denosumab-related osteonecrosis of the jaw after tooth extraction and the effects of a short drug holiday in cancer patients: a multicenter retrospective study

Pre-existing inflammation, corticosteroid therapy, periapical periodontitis, longer duration of denosumab therapy, and female sex were significantly associated with an increased risk of denosumab-related osteonecrosis of the jaw after tooth extraction in patients with cancer on oncologic doses of denosumab. A short drug holiday did not protect against this complication.

INTRODUCTION

This study retrospectively investigated the relationship between various risk factors, including brief discontinuation of denosumab, and development of denosumab-related osteonecrosis of the jaw (DRONJ) after tooth extraction in patients with cancer who were receiving oncologic doses of this agent.

METHODS

Data were collected on demographic characteristics, duration of denosumab therapy, whether or not denosumab was discontinued before tooth extraction (drug holiday), duration of discontinuation, presence of pre-existing inflammation, and whether or not additional surgical procedures were performed. Risk factors for DRONJ after tooth extraction were evaluated by univariate and multivariate analyses.

RESULTS

A total of 136 dental extractions were performed in 72 patients (31 men, 41 women) with cancer who were receiving oncologic doses of denosumab. Post-extraction DRONJ was diagnosed in 39 teeth (28.7%) in 25 patients. Tooth extraction was significantly associated with development of DRONJ only in patients with pre-existing inflammation (odds ratio [OR] 243.77), those on corticosteroid therapy (OR 73.50), those with periapical periodontitis (OR 14.13), those who had been taking oncologic doses of denosumab for a longer period (OR 4.69), and in women (OR 1.04). There was no significant difference in the occurrence of DRONJ between patients who had a drug holiday before tooth extraction and those who did not.

CONCLUSIONS

These findings suggest that inflamed teeth should be extracted immediately in patients with cancer who are receiving oncologic doses of denosumab. Drug holidays have no significant impact on the risk of DRONJ.

Long-term prognostic factors of coronary artery disease patients after out-of-hospital cardiac arrest

Background

The long-term prognosis of survival in patients with out-of-hospital cardiac arrest (OHCA) with coronary artery disease (CAD) remains poor.

Methods

There were 2391 out-of-hospital cardiac arrest (OHCA) patients transferred to 2 hospitals. We included 405 cardiovascular arrest patients, who got return of spontaneous circulation (ROSC) from January 2015 to December 2018. Among them, 204 patients had CAD that caused OHCA (39%: multi-vessel disease, 19%: chronic total occlusion (CTO), 13%: vasospastic angina (VSA)). To predict mortality, we investigated patients' characteristics, pre-hospital information and findings of CAG.

Results

At 1-year later, 104 patients (51%) survived. Younger age (P&lt;0.001), VF survivor (P&lt;0.001), pre-hospital ROSC (P&lt;0.001), bystander CPR (P=0.013), without ECMO (P&lt;0.001), lower lactate level on admission (P&lt;0.001), and higher geriatric nutritional risk index score (P&lt;0.001) were associated with low 1-year mortality, while with ST-segment elevation (P=0.778), BMI level (P=0.344), and sex (0.401) were not. And in the findings of CAG, the past history of CAD (P=0.049), the higher number of coronary vessel disease (P=0.003) such as multi-vessel disease (P=0.022), higher SYNAX score (P=0.016), and larger infarct size (max CK level; P=0.013, max CK-MB level; P&lt;0.001) were associated with high 1-year mortality. On the other hand, acute coronary syndrome (P=0.300), any coronary lesion (RCA (P=0.447), LAD (P=0.089), LCX (P=0.096), or LMT (P=0.842)), and with CTO lesion (P=0.140) were not associated. Zero-vessel disease (VSA, P=0.001) had lower mortality among the CAD patients. In the multivariate Cox proportional hazards model, age (hazards ratio; HR: 1.03, 95%confidence interval (CI) 1.00–1.06, P&lt;0.001) and bystander CPR (HR: 0.36, 95% CI 0.20–0.65, P&lt;0.001) were the independent predictors of mortality.

Conclusions

Younger age and pre-hospital support after OHCA with CAD were the predictors of low mortality. Pre-hospital information, systemic condition on arrival, or anatomical coronary complexity were important to predict low mortality.

Funding Acknowledgement

Type of funding sources: None.

Effect of endothelial nitric oxide synthase gene polymorphism on cardiovascular death and nonfatal myocardial infarction in Japanese general population

Introduction

Single nucleotide polymorphisms (SNP) of endothelial nitric oxide synthase (NOS3) have been reported to be associated with diabetes mellitus and myocardial infarction. However, few reports have prospectively investigated the effects of NOS3 SNP on cardiovascular death and nonfatal myocardial infarction.

Purpose

The purpose of this study was to investigate the impact of NOS3 SNP on cardiovascular death and the development of nonfatal myocardial infarction.

Methods

This prospective cohort study included 2,752 subjects (aged ≥40) who participated in a community based health checkup. We genotyped two SNPs within NOS3 (rs1808593, rs1799983). All subjects were prospectively followed during the median follow-up period of 15.4 years with the end point of cardiovascular death and nonfatal myocardial infarction.

Results

The homozygous G-allele (GG), heterozygous (GT), and homozygous T-allele (TT) carriers of rs1808593 were identified in 60 (2%), 706 (26%), and 1,986 (72%) subjects, respectively. Kaplan-Meier analysis demonstrated that homozygous G-allele carriers of rs1808593 had the greater risk than those without. Multivariate Cox proportional hazard regression analysis revealed that the homozygous G allele of rs1808593 was associated with cardiovascular death and the development of nonfatal myocardial infarction after adjusting for confounding risk factors.

Conclusions

NOS3 gene polymorphism could be a genetic risk factor for cardiovascular death and nonfatal myocardial infarction in the Japanese general population.

Funding Acknowledgement

Type of funding sources: None. Figure 1

Neurological outcome at 30-day as an estimator of 1-year functional status after out-of-hospital cardiac arrest with post-encephalopathy

Background

Implantation of implantable cardioverter-defibrillators (ICD) for secondary prevention is fully recommended for those with an estimated survival over 1-year with a good functional status. However, we often face the difficulty to estimate the functional status and hesitate to implant ICD for patients with post-resuscitation encephalopathy.

Methods

There were 2391 out-of-hospital cardiac arrest (OHCA) patients transferred to 2 hospitals. We included 405 cardiovascular arrest patients, who got return of spontaneous circulation (ROSC) from January 2015 to December 2018. Among the patients, 343 patients could be considered to be implanted ICD for secondary prevention according to the initial waveform or the causes of OHCA. At 1-month later, 184 patients (54%) survived. To assess the association of functional status at 30-day and 1-year, we investigated patients' characteristics, pre-hospital information and clinical findings, and evaluated the neurological outcome according to the cerebral performance category (CPC) scale.

Results

At 1-month later, 145 patients (79%) survived with CPC≤2, and 39 patients (21%) survived with CPC&gt;2. Bystander CPR (P=0.009), pre-hospital ROSC (P&lt;0.001), low lactate level on admission (P=0.001), high geriatric nutritional risk index score (P&lt;0.001) and without ECMO (P=0.002) were significantly associated with good neurological outcome at 30-day. The 1-year Kaplan-Meier event rate revealed significantly different survival rate (CPC&gt;2 at 30-day:38.5%, vs CPC≤2 at 30-day:97.2%; P&lt;0.001). In multivariate analysis, CPC scale at 30-day (OR 0.022; 95% CI 0.003–0.140; p&lt;0.001) was the independent predictor of favorable neurological outcome at 1-year. Among the patients with CPC&gt;2 at 30-day, only 3 patients (7.7%) of CPC=3 achieved the improvement of neurological outcome at 1-year (CPC≤2), while no patient of CPC=4 did. And one patient (2.5%) with CPC=3 was implanted ICD during the follow-up period. Twenty-five patients (64%) died of non-cardiovascular death with frailty of post-resuscitation encephalopathy after they were transferred to other hospital with the acceptation and intention of the do-not-attempt-resuscitation.

Conclusions

Neurological prognosis at 30-day after OHCA might be an estimator of 1-year functional status to guide us to implant ICD for secondary prevention.

Funding Acknowledgement

Type of funding sources: None.

Gender differences in the impact of plasma xanthine oxidoreductase activity on coronary artery spasm

Background

It has been reported that decreased nitric oxide bioavailability due to increased reactive oxygen species (ROS) is one of the most important causes of coronary artery spasm (CAS). Xanthine oxidoreductase (XOR) is the rate-limiting enzyme for uric acid (UA) production and plays a pivotal role in generating ROS. It was reported that the gender differences exist in the impact of serum UA levels on cardiovascular risks. We previously demonstrated that increased plasma XOR activity is significantly associated with the incidence of CAS. However, the gender differences in the impact of plasma XOR activity on CAS remain unclear.

Purpose

The aim of this study was to examine the gender differences in the clinical impact of plasma XOR activity on CAS.

Methods

We investigated plasma XOR activity in 132 patients suspected for CAS (male, n=78; female, n=54), and underwent intracoronary acetylcholine provocation test. XOR activity assay was performed using stable isotope-labeled substrate and liquid chromatography-triple quadrupole mass spectrometry. Provoked CAS was defined as total or subtotal occlusion (≥90%) with accompanying symptoms of chest pain and/or ischemic ST-segment changes on the electrocardiogram. We excluded the patients who had significant coronary artery stenosis (≥50%) and/or were taking XOR inhibitors.

Results

Plasma XOR activity was significantly lower in female compared with male patients (30.3 pmol/h/mL, interquartile range (IQR) 22.8–42.7 vs. 51.7 pmol/h/mL, IQR 34.7–101.8; P&lt;0.001). CAS was provoked in 36 male patients and 17 female patients, and they each had significantly higher plasma XOR activity compared with those without, respectively. Multivariate logistic regression analysis showed that plasma XOR activity was independently associated with the incidence of CAS in both genders after adjustment for confounding factors. The optimal cut-off values for predicting CAS were lower in female than those in male patients (52.3 vs. 91.6 pmol/h/mL). Multivariate analysis demonstrated that female patients with high XOR activity (≥52.3 pmol/h/mL; odds ratio [OR] 22.6, P&lt;0.001) exhibited a higher incidence of CAS compared with that in male patients (≥91.6 pmol/h/mL; OR 8.2, P&lt;0.001).

Conclusions

Plasma XOR activity was an independent predictor for the incidence of CAS in both genders. The impact of plasma XOR activity on CAS was stronger in female patients than in male patients.

Funding Acknowledgement

Type of funding sources: None. Figure 1Figure 2

Feature-fusion improves MRI single-shot deep learning detection of small brain metastases

BACKGROUND AND PURPOSE

To examine whether feature-fusion (FF) method improves single-shot detector's (SSD's) detection of small brain metastases on contrast-enhanced (CE) T1-weighted MRI.

METHODS

The study included 234 MRI scans from 234 patients (64.3 years±12.0; 126 men). The ground-truth annotation was performed semiautomatically. SSDs with and without an FF module were developed and trained using 178 scans. The detection performance was evaluated at the SSDs' 50% confidence threshold using sensitivity, positive-predictive value (PPV), and the false-positive (FP) per scan with the remaining 56 scans.

RESULTS

FF-SSD achieved an overall sensitivity of 86.0% (95% confidence interval [CI]: [83.0%, 85.6%]; 196/228) and 46.8% PPV (95% CI: [42.0%, 46.3%]; 196/434), with 4.3 FP (95% CI: [4.3, 4.9]). Lesions smaller than 3 mm had 45.8% sensitivity (95% CI: [36.1%, 45.5%]; 22/48) with 2.0 FP (95% CI: [1.9, 2.1]). Lesions measuring 3-6 mm had 92.3% sensitivity (95% CI: [86.5%, 92.0%]; 48/52) with 1.8 FP (95% CI: [1.7, 2.2]). Lesions larger than 6 mm had 98.4% sensitivity (95% CI: [97.8%, 99.4%]; 126/128) 0.5 FP (95% CI: [0.5, 0.8]) per scan. FF-SSD had a significantly higher sensitivity for lesions < 3 mm (p = 0.008, t = 3.53) than the baseline SSD, while the overall PPV was similar (p = 0.06, t = -2.16). A similar trend was observed even when the detector's confidence threshold was varied as low as 0.2, for which the FF-SSD's sensitivity was 91.2% and the FP was 9.5.

CONCLUSIONS

The FF-SSD algorithm identified brain metastases on CE T1-weighted MRI with high accuracy.

Fungal Planet description sheets: 1284-1382

Novel species of fungi described in this study include those from various countries as follows: Antartica, Cladosporium austrolitorale from coastal sea sand. Australia, Austroboletus yourkae on soil, Crepidotus innuopurpureus on dead wood, Curvularia stenotaphri from roots and leaves of Stenotaphrum secundatum and Thecaphora stajsicii from capsules of Oxalis radicosa. Belgium, Paraxerochrysium coryli (incl. Paraxerochrysium gen. nov.) from Corylus avellana. Brazil, Calvatia nordestina on soil, Didymella tabebuiicola from leaf spots on Tabebuia aurea, Fusarium subflagellisporum from hypertrophied floral and vegetative branches of Mangifera indica and Microdochium maculosum from living leaves of Digitaria insularis. Canada, Cuphophyllus bondii from a grassland. Croatia, Mollisia inferiseptata from a rotten Laurus nobilis trunk. Cyprus, Amanita exilis on calcareous soil. Czech Republic, Cytospora hippophaicola from wood of symptomatic Vaccinium corymbosum. Denmark, Lasiosphaeria deviata on pieces of wood and herbaceous debris. Dominican Republic, Calocybella goethei among grass on a lawn. France (Corsica), Inocybe corsica on wet ground. France (French Guiana), Trechispora patawaensis on decayed branch of unknown angiosperm tree and Trechispora subregularis on decayed log of unknown angiosperm tree. Germany, Paramicrothecium sambuci (incl. Paramicrothecium gen. nov.) on dead stems of Sambucus nigra. India, Aureobasidium microtermitis from the gut of a Microtermes sp. termite, Laccaria diospyricola on soil and Phylloporia tamilnadensis on branches of Catunaregam spinosa. Iran, Pythium serotinoosporum from soil under Prunus dulcis. Italy, Pluteus brunneovenosus on twigs of broadleaved trees on the ground. Japan, Heterophoma rehmanniae on leaves of Rehmannia glutinosa f. hueichingensis. Kazakhstan, Murispora kazachstanica from healthy roots of Triticum aestivum. Namibia, Caespitomonium euphorbiae (incl. Caespitomonium gen. nov.) from stems of an Euphorbia sp. Netherlands, Alfaria junci, Myrmecridium junci, Myrmecridium juncicola, Myrmecridium juncigenum, Ophioceras junci, Paradinemasporium junci (incl. Paradinemasporium gen. nov.), Phialoseptomonium junci, Sporidesmiella juncicola, Xenopyricularia junci and Zaanenomyces quadripartis (incl. Zaanenomyces gen. nov.), from dead culms of Juncus effusus, Cylindromonium everniae and Rhodoveronaea everniae from Evernia prunastri, Cyphellophora sambuci and Myrmecridium sambuci from Sambucus nigra, Kiflimonium junci, Sarocladium junci, Zaanenomyces moderatricis-academiae and Zaanenomyces versatilis from dead culms of Juncus inflexus, Microcera physciae from Physcia tenella, Myrmecridium dactylidis from dead culms of Dactylis glomerata, Neochalara spiraeae and Sporidesmium spiraeae from leaves of Spiraea japonica, Neofabraea salicina from Salix sp., Paradissoconium narthecii (incl. Paradissoconium gen. nov.) from dead leaves of Narthecium ossifragum, Polyscytalum vaccinii from Vaccinium myrtillus, Pseudosoloacrosporiella cryptomeriae (incl. Pseudosoloacrosporiella gen. nov.) from leaves of Cryptomeria japonica, Ramularia pararhabdospora from Plantago lanceolata, Sporidesmiella pini from needles of Pinus sylvestris and Xenoacrodontium juglandis (incl. Xenoacrodontium gen. nov. and Xenoacrodontiaceae fam. nov.) from Juglans regia. New Zealand, Cryptometrion metrosideri from twigs of Metrosideros sp., Coccomyces pycnophyllocladi from dead leaves of Phyllocladus alpinus, Hypoderma aliforme from fallen leaves Fuscopora solandri and Hypoderma subiculatum from dead leaves Phormium tenax. Norway, Neodevriesia kalakoutskii from permafrost and Variabilispora viridis from driftwood of Picea abies. Portugal, Entomortierella hereditatis from a biofilm covering a deteriorated limestone wall. Russia, Colpoma junipericola from needles of Juniperus sabina, Entoloma cinnamomeum on soil in grasslands, Entoloma verae on soil in grasslands, Hyphodermella pallidostraminea on a dry dead branch of Actinidia sp., Lepiota sayanensis on litter in a mixed forest, Papiliotrema horticola from Malus communis, Paramacroventuria ribis (incl. Paramacroventuria gen. nov.) from leaves of Ribes aureum and Paramyrothecium lathyri from leaves of Lathyrus tuberosus. South Africa, Harzia combreti from leaf litter of Combretum collinum ssp. sulvense, Penicillium xyleborini from Xyleborinus saxesenii, Phaeoisaria dalbergiae from bark of Dalbergia armata, Protocreopsis euphorbiae from leaf litter of Euphorbia ingens and Roigiella syzygii from twigs of Syzygium chordatum. Spain, Genea zamorana on sandy soil, Gymnopus nigrescens on Scleropodium touretii, Hesperomyces parexochomi on Parexochomus quadriplagiatus, Paraphoma variabilis from dung, Phaeococcomyces kinklidomatophilus from a blackened metal railing of an industrial warehouse and Tuber suaveolens in soil under Quercus faginea. Svalbard and Jan Mayen, Inocybe nivea associated with Salix polaris. Thailand, Biscogniauxia whalleyi on corticated wood. UK, Parasitella quercicola from Quercus robur. USA, Aspergillus arizonicus from indoor air in a hospital, Caeliomyces tampanus (incl. Caeliomyces gen. nov.) from office dust, Cippumomyces mortalis (incl. Cippumomyces gen. nov.) from a tombstone, Cylindrium desperesense from air in a store, Tetracoccosporium pseudoaerium from air sample in house, Toxicocladosporium glendoranum from air in a brick room, Toxicocladosporium losalamitosense from air in a classroom, Valsonectria portsmouthensis from air in men's locker room and Varicosporellopsis americana from sludge in a water reservoir. Vietnam, Entoloma kovalenkoi on rotten wood, Fusarium chuoi inside seed of Musa itinerans, Micropsalliota albofelina on soil in tropical evergreen mixed forests and Phytophthora docyniae from soil and roots of Docynia indica. Morphological and culture characteristics are supported by DNA barcodes. Citation: Crous PW, Osieck ER, Jurjević Ž, et al. 2021. Fungal Planet description sheets: 1284-1382. Persoonia 47: 178-374. https://doi.org/10.3767/persoonia.2021.47.06.

Gamma-Ray Observation of the Cygnus Region in the 100-TeV Energy Region

We report observations of gamma-ray emissions with energies in the 100-TeV energy region from the Cygnus region in our Galaxy. Two sources are significantly detected in the directions of the Cygnus OB1 and OB2 associations. Based on their positional coincidences, we associate one with a pulsar PSR J2032+4127 and the other mainly with a pulsar wind nebula PWN G75.2+0.1, with the pulsar moving away from its original birthplace situated around the centroid of the observed gamma-ray emission. This work would stimulate further studies of particle acceleration mechanisms at these gamma-ray sources.

Structural connectivity changes in the cerebral pain matrix in burning mouth syndrome: a multi-shell, multi-tissue-constrained spherical deconvolution model analysis

PURPOSE

Burning mouth syndrome (BMS) is a chronic intraoral pain syndrome. Previous studies have attempted to determine the brain connectivity features in BMS using functional and structural magnetic resonance imaging. However, no study has investigated the structural connectivity using multi-shell, multi-tissue-constrained spherical deconvolution (MSMT-CSD), anatomically constrained tractography (ACT), and spherical deconvolution informed filtering of tractograms (SIFT). Therefore, this study aimed to assess the differences in brain structural connectivity of patients with BMS and healthy controls using probabilistic tractography with these methods, and graph analysis.

METHODS

Fourteen patients with BMS and 11 age- and sex-matched healthy volunteers underwent 3-T magnetic resonance imaging. MSMT-CSD-based probabilistic structural connectivity was computed using the second-order integration over fiber orientation distributions algorithm based on nodes set in 84 anatomical cortical regions with ACT and SIFT. A t-test was performed for comparisons between the BMS and healthy control brain networks.

RESULTS

The betweenness centrality was significantly higher in the left insula, right amygdala, and right lateral orbitofrontal cortex and significantly lower in the right inferotemporal cortex in the BMS group than that in healthy controls. However, no significant difference was found in the clustering coefficient, node degree, and small-worldness between the two groups.

CONCLUSION

Graph analysis of brain probabilistic structural connectivity, based on diffusion imaging using an MSMT-CSD model with ACT and SIFT, revealed alterations in the regions comprising the pain matrix and medial pain ascending pathway. These results highlight the emotional-affective profile of BMS, which is a type of chronic pain syndrome.

Automated detection of brain metastases on non-enhanced CT using single-shot detectors

PURPOSE

To develop and investigate deep learning-based detectors for brain metastases detection on non-enhanced (NE) CT.

METHODS

The study included 116 NECTs from 116 patients (81 men, age 66.5 ± 10.6 years) to train and test single-shot detector (SSD) models using 89 and 27 cases, respectively. The annotation was performed by three radiologists using bounding-boxes defined on contrast-enhanced CT (CECT) images. NECTs were coregistered and resliced to CECTs. The detection performance was evaluated at the SSD's 50% confidence threshold using sensitivity, positive-predictive value (PPV), and the false-positive rate per scan (FPR). For false negatives and true positives, binary logistic regression was used to examine the possible contributing factors.

RESULTS

For lesions 6 mm or larger, the SSD achieved a sensitivity of 35.4% (95% confidence interval (CI): [32.3%, 33.5%]); 51/144) with an FPR of 14.9 (95% CI [12.4, 13.9]). The overall sensitivity was 23.8% (95% CI: [21.3%, 22.8%]; 55/231) and PPV was 19.1% (95% CI: [18.5%, 20.4%]; 98/ of 513), with an FPR of 15.4 (95% CI [12.9, 14.5]). Ninety-five percent of the lesions that SSD failed to detect were also undetectable to radiologists (168/176). Twenty-four percent of the lesions (13/50) detected by the SSD were undetectable to radiologists. Logistic regression analysis indicated that density, necrosis, and size contributed to the lesions' visibility for radiologists, while for the SSD, the surrounding edema also enhanced the detection performance.

CONCLUSION

The SSD model we developed could detect brain metastases larger than 6 mm to some extent, a quarter of which were even retrospectively unrecognizable to radiologists.

Accelerated Isotropic Multiparametric Imaging by High Spatial Resolution 3D-QALAS With Compressed Sensing: A Phantom, Volunteer, and Patient Study

OBJECTIVES

The aims of this study were to develop an accelerated multiparametric magnetic resonance imaging method based on 3D-quantification using an interleaved Look-Locker acquisition sequence with a T2 preparation pulse (3D-QALAS) combined with compressed sensing (CS) and to evaluate the effect of CS on the quantitative mapping, tissue segmentation, and quality of synthetic images.

MATERIALS AND METHODS

A magnetic resonance imaging system phantom, containing multiple compartments with standardized T1, T2, and proton density (PD) values; 10 healthy volunteers; and 12 patients with multiple sclerosis were scanned using the 3D-QALAS sequence with and without CS and conventional contrast-weighted imaging. The scan times of 3D-QALAS with and without CS were 5:56 and 11:11, respectively. For healthy volunteers, brain volumetry and myelin estimation were performed based on the measured T1, T2, and PD. For patients with multiple sclerosis, the mean T1, T2, PD, and the amount of myelin in plaques and contralateral normal-appearing white matter (NAWM) were measured. Simple linear regression analysis and Bland-Altman analysis were performed for each metric obtained from the datasets with and without CS. To compare overall image quality and structural delineations on synthetic and conventional contrast-weighted images, case-control randomized reading sessions were performed by 2 neuroradiologists in a blinded manner.

RESULTS

The linearity of both phantom and volunteer measurements in T1, T2, and PD values obtained with and without CS was very strong (R2 = 0.9901-1.000). The tissue segmentation obtained with and without CS also had high linearity (R2 = 0.987-0.999). The quantitative tissue values of the plaques and NAWM obtained with CS showed high linearity with those without CS (R2 = 0.967-1.000). There were no significant differences in overall image quality between synthetic contrast-weighted images obtained with and without CS (P = 0.17-0.99).

CONCLUSIONS

Multiparametric imaging of the whole brain based on 3D-QALAS can be accelerated using CS while preserving tissue quantitative values, tissue segmentation, and quality of synthetic images.

First Detection of sub-PeV Diffuse Gamma Rays from the Galactic Disk: Evidence for Ubiquitous Galactic Cosmic Rays beyond PeV Energies

We report, for the first time, the long-awaited detection of diffuse gamma rays with energies between 100 TeV and 1 PeV in the Galactic disk. Particularly, all gamma rays above 398 TeV are observed apart from known TeV gamma-ray sources and compatible with expectations from the hadronic emission scenario in which gamma rays originate from the decay of π^{0}'s produced through the interaction of protons with the interstellar medium in the Galaxy. This is strong evidence that cosmic rays are accelerated beyond PeV energies in our Galaxy and spread over the Galactic disk.

Differentiation between multiple sclerosis and neuromyelitis optica spectrum disorders by multiparametric quantitative MRI using convolutional neural network

Multiple sclerosis and neuromyelitis optica spectrum disorders are both neuroinflammatory diseases and have overlapping clinical manifestations. We developed a convolutional neural network model that differentiates between the two based on magnetic resonance imaging data. Thirty-five patients with relapsing-remitting multiple sclerosis and eighteen age-, sex-, disease duration-, and Expanded Disease Status Scale-matched patients with anti-aquaporin-4 antibody-positive neuromyelitis optica spectrum disorders were included in this study. All patients were scanned on a 3-T scanner using a multi-dynamic multi-echo sequence that simultaneously measures R1 and R2 relaxation rates and proton density. R1, R2, and proton density maps were analyzed using our convolutional neural network model. To avoid overfitting on a small dataset, we aimed to separate features of images into those specific to an image and those common to the group, based on SqueezeNet. We used only common features for classification. Leave-one-out cross validation was performed to evaluate the performance of the model. The area under the receiver operating characteristic curve of the developed convolutional neural network model for differentiating between the two disorders was 0.859. The sensitivity to multiple sclerosis and neuromyelitis optica spectrum disorders, and accuracy were 80.0%, 83.3%, and 81.1%, respectively. In conclusion, we developed a convolutional neural network model that differentiates between multiple sclerosis and neuromyelitis optica spectrum disorders, and which is designed to avoid overfitting on small training datasets. Our proposed algorithm may facilitate a differential diagnosis of these diseases in clinical practice.

Age-Related Changes in Relaxation Times, Proton Density, Myelin, and Tissue Volumes in Adult Brain Analyzed by 2-Dimensional Quantitative Synthetic Magnetic Resonance Imaging

OBJECTIVES

Quantitative synthetic magnetic resonance imaging (MRI) enables the determination of fundamental tissue properties, namely, T1 and T2 relaxation times and proton density (PD), in a single scan. Myelin estimation and brain segmentation based on these quantitative values can also be performed automatically. This study aimed to reveal the changes in tissue characteristics and volumes of the brain according to age and provide age-specific reference values obtained by quantitative synthetic MRI.

MATERIALS AND METHODS

This was a prospective study of healthy subjects with no history of brain diseases scanned with a multidynamic multiecho sequence for simultaneous measurement of relaxometry of T1, T2, and PD. We performed myelin estimation and brain volumetry based on these values. We performed volume-of-interest analysis on both gray matter (GM) and white matter (WM) regions for T1, T2, PD, and myelin volume fraction maps. Tissue volumes were calculated in the whole brain, producing brain parenchymal volume, GM volume, WM volume, and myelin volume. These volumes were normalized by intracranial volume to a brain parenchymal fraction, GM fraction, WM fraction, and myelin fraction (MyF). We examined the changes in the mean regional quantitative values and segmented tissue volumes according to age.

RESULTS

We analyzed data of 114 adults (53 men and 61 women; median age, 66.5 years; range, 21-86 years). T1, T2, and PD values showed quadratic changes according to age and stayed stable or decreased until around 60 years of age and increased thereafter. Myelin volume fraction showed a reversed trend. Brain parenchymal fraction and GM fraction decreased throughout all ages. The approximation curves showed that WM fraction and MyF gradually increased until around the 40s to 50s and decreased thereafter. A significant decline in MyF was first noted in the 60s age group (Tukey test, P < 0.001).

CONCLUSIONS

Our study showed changes according to age in tissue characteristic values and brain volumes using quantitative synthetic MRI. The reference values for age demonstrated in this study may be useful to discriminate brain disorders from healthy brains.

Automatic detection of brain metastases on contrast-enhanced CT with deep-learning feature-fused single-shot detectors

PURPOSE

Despite the potential usefulness, no automatic detector is available for brain metastases on contrast-enhanced CT (CECT). The study aims to develop and investigate deep learning-based detectors for brain metastases detection on CECT.

METHOD

The study included 127 CECTs from 127 patients (65.5 years±11.1; 87 men). The ground-truth annotation was performed semi-automatically by applying connected-component analysis to the binarized dataset by three radiologists. Single-shot detector (SSD) algorithms, with and without a feature-fusion module, were developed and trained using 97 scans. The performance was evaluated at the detector's 50 % confidence threshold with the remaining 30 scans using sensitivity, positive-predictive value (PPV), and the false-positive rate per scan (FPR).

RESULTS

Feature-fused SSD achieved an overall sensitivity of 88.1 % (95 % confidence interval [CI]: [85.2 %,88.6 %]; 214/243) and PPV of 36.0 % (95 % CI: [33.7 %,37.1 %]; 233/648), with 13.8 FPR (95 % CI: [12.7,15.0]). Lesions < 3 mm had a sensitivity of 23.1 % (95 % CI: [21.2 %,40.0 %]; 3/13), with 0.2 FPR (95 % CI: [0.23,0.65]). Lesions measuring 3-6 mm had a sensitivity of 80.0 % (95 % CI: [76.0 %,79.8 %]); 60/75) with 5.8 FPR (95 % CI: [5.0,6.2]). Lesions > 6 mm had a sensitivity of 97.4 % (95 % CI: [94.1 %,97.4 %]); 151/155) with 7.9 FPR (95 % CI: [7.2,8.5]). Feature-fused SSD had a significantly higher overall sensitivity (p = 0.03, t = 2.75) or sensitivity for lesions < 3 mm (p = 0.002, t = 4.49) than baseline SSD, while the overall PPV was similar (p = 0.96, t = -0.02).

CONCLUSIONS

The SSD algorithm identified brain metastases on CECT with reasonable accuracy for lesions > 3 mm without pre/post-processing.

Effect of hybrid of compressed sensing and parallel imaging on the quantitative values measured by 3D quantitative synthetic MRI: A phantom study

INTRODUCTION

Recently, three-dimensional (3D) quantitative synthetic magnetic resonance imaging (MRI), which quantifies tissue properties and creates multiple contrast-weighted images, has been enabled by 3D-quantification using an interleaved Look-Locker acquisition sequence with a T2 preparation pulse (3D-QALAS). However, the relatively long scan time has hindered its introduction into clinical practice. A hybrid of compressed sensing and parallel imaging (Compressed sensing-sensitivity encoding: CS-SENSE) can accelerate 3D-QALAS; however, whether CS-SENSE affects the quantitative values acquired by 3D-QALAS remains unexplored. Therefore, this study aimed to examine the effects of reduction factors of CS-SENSE (R_CSS) on the quantitative values derived from 3D-QALAS, by assessing the signal-to-noise ratio (SNR) of the quantitative maps, as well as accuracy (linearity and bias) and repeatability of measured quantitative values.

METHODS

In this study, the ISMRM/NIST standardized phantom was scanned on a 1.5-T MRI scanner with 3D-QALAS using R_CSS in the range between 1 and 3, with intervals of 0.2, and between 3 and 10 with intervals of 0.5. The T1, T2, and proton density (PD) values were calculated from the imaging data. For each quantitative value, the SNR, the coefficient of determination (R²) of a linear regression model, the error rate, and the within-subject coefficient of variation (wCV) were calculated for each R_CSS and compared.

RESULTS

Within the clinically-relevant dynamic range of the brain of T1 and T2 (T1: 200-1400 ms; T2; 50-400 ms) and PD value of 15-100% calculated from 3D-QALAS, the effects of R_CSS on quantitative values was small between 1 and 2.8, with SNR ≧ 10, R² ≧ 0.9, error rate ≦ 10%, and wCV ≦ 10%, except for T2 values of 186.1 and 258.4 ms.

CONCLUSIONS

CS-SENSE enabled the reduction of the scan time of 3D-QALAS by 63.5% (R_CSS = 2.8) while maintaining the SNR of quantitative maps and accuracy and repeatability of the quantitative values.

3D Quantitative Synthetic MRI in the Evaluation of Multiple Sclerosis Lesions

BACKGROUND AND PURPOSE

Synthetic MR imaging creates multiple contrast-weighted images based on a single time-efficient quantitative scan, which has been mostly performed for 2D acquisition. We assessed the utility of 3D synthetic MR imaging in patients with MS by comparing its diagnostic image quality and lesion volumetry with conventional MR imaging.

MATERIALS AND METHODS

Twenty-four patients with MS prospectively underwent 3D quantitative synthetic MR imaging and conventional T1-weighted, T2-weighted, FLAIR, and double inversion recovery imaging, with acquisition times of 9 minutes 3 seconds and 18 minutes 27 seconds for the synthetic MR imaging and conventional MR imaging sequences, respectively. Synthetic phase-sensitive inversion recovery images and those corresponding to conventional MR imaging contrasts were created for synthetic MR imaging. Two neuroradiologists independently assessed the image quality on a 5-point Likert scale. The numbers of cortical lesions and lesion volumes were quantified using both synthetic and conventional image sets.

RESULTS

The overall diagnostic image quality of synthetic T1WI and double inversion recovery images was noninferior to that of conventional images (P = .23 and .20, respectively), whereas that of synthetic T2WI and FLAIR was inferior to that of conventional images (both Ps < .001). There were no significant differences in the number of cortical lesions (P = .17 and .53 for each rater) or segmented lesion volumes (P = .61) between the synthetic and conventional image sets.

CONCLUSIONS

Three-dimensional synthetic MR imaging could serve as an alternative to conventional MR imaging in evaluating MS with a reduced scan time.

Relationship between cardiac magnetic resonance derived extracellular volume fraction and myocardial strain in patients with non-ischemic dilated cardiomyopathy

Background

The feature tracking (FT) technique has been proposed as a robust method to evaluate the myocardial strain using conventional cine magnetic resonance imaging (MRI) of the left ventricle. Data is limited regarding the relationship between FT-derived myocardial strain and diffuse myocardial fibrosis evaluated by T1 mapping in patients with non-ischemic dilated cardiomyopathy (NIDCM).

Purpose

The aim of this study was to evaluate the correlation between extracellular volume (ECV) by T1 mapping and myocardial strain by FT in patients with NIDCM.

Methods

A total of sixty-four patients with NIDCM (62±12 years) and 15 controls (62±11 years) were studied. Using a 1.5T MR scanner, pre- and post- T1 mapping images of LV wall at mid-ventricular level was acquired to calculate ECV by modified Look-Locker inversion recovery (MOLLI) sequence. Radial strain (RS), circumferential strain (CS) and longitudinal strain (LS) was assessed by FT technique. ECV and myocardial strain were compared using a 6-segment model at mid-ventricular level.

Results

Compared to the controls, the NIDCM patients had a significantly higher ECV (0.30±0.02 vs. 0.24±0.01, p&lt;0.001) and impaired myocardial strain (RS, 24.2±3.0 vs. 52.2±6.2, p&lt;0.001; CS, −7.5±2.1 vs. −15.3±2.2, p&lt;0.001; LS −10.4±3.5 vs. −20.2±4.7, p&lt;0.001, respectively). Similar results were obtained when comparing all 6 myocardial segments (segment 7–12) (all p values &lt;0.001). In a segment-based analysis, a significant positive correlation was found between the ECV and CS (r=0.26 to 0.41; all p values &lt;0.05), a negative correlation was found between the ECV and RS (r=−0.31 to −0.41; all p values &lt;0.05). In a patient-based analysis, there were significant positive correlations between the ECV and CS (r=0.45, p&lt;0.001), ECV and LS from 2-chamber view (r=0.30, p=0.006), ECV and LS from 4-chamber view (r=0.37, p&lt;0.001). There was a significant negative correlation between the ECV and RS (r=−0.43, p&lt;0.001) (FIGURE)

Conclusions

In NIDCM patients, severity of myocardial fibrosis evaluated by T1 mapping is associated with impaired myocardial strain by FT technique.

Correlation between the ECV and strain

Funding Acknowledgement

Type of funding source: None