Microfluidic platform using focused ultrasound passing through hydrophobic meshes with jump availability

Applications in chemistry, biology, medicine, and engineering require the large-scale manipulation of a wide range of chemicals, samples, and specimens. To achieve maximum efficiency, parallel control of microlitre droplets using automated techniques is essential. Electrowetting-on-dielectric (EWOD), which manipulates droplets using the imbalance of wetting on a substrate, is the most widely employed method. However, EWOD is limited in its capability to make droplets detach from the substrate (jumping), which hinders throughput and device integration. Here, we propose a novel microfluidic system based on focused ultrasound passing through a hydrophobic mesh with droplets resting on top. A phased array dynamically creates foci to manipulate droplets of up to 300 μL. This platform offers a jump height of up to 10 cm, a 27-fold improvement over conventional EWOD systems. In addition, droplets can be merged or split by pushing them against a hydrophobic knife. We demonstrate Suzuki-Miyaura cross-coupling using our platform, showing its potential for a wide range of chemical experiments. Biofouling in our system was lower than in conventional EWOD, demonstrating its high suitability for biological experiments. Focused ultrasound allows the manipulation of both solid and liquid targets. Our platform provides a foundation for the advancement of micro-robotics, additive manufacturing, and laboratory automation.

Change in Geographic Accessibility to Dental Clinics Affects Access to Care

Access to dental clinics is a feature of the neighborhood service environment that may influence oral health care utilization. However, residential selection poses a challenge to causal inference. By studying the involuntary relocation of survivors of the 2011 Great East Japan Earthquake and Tsunami (GEJE), we examined the association between changes in geographic distance to dental clinics and dental visits. Longitudinal data from a cohort of older residents in Iwanuma City directly impacted by the GEJE were analyzed in this study. The baseline survey was conducted in 2010, 7 mo before the occurrence of GEJE, and a follow-up was conducted in 2016. Using Poisson regression models, we estimated the incidence rate ratios (IRR) and 95% confidence intervals (CIs) for the uptake of denture use (as a proxy for dental visits) according to changes in distance from the nearest dental clinic to their house. Age at baseline, housing damage by the disaster, deteriorating economic conditions, and worsened physical activity were used as confounders. Among the 1,098 participants who had not worn dentures before the GEJE, 495 were men (45.1%), with a mean ± SD age at baseline of 74.0 ± 6.9 y. During the 6-year follow-up, 372 (33.9%) participants initiated denture use. Compared to those who experienced a large increase in distance to dental clinics (>370.0-6,299.1 m), a large decrease in distance to dental clinics (>429.0-5,382.6 m) was associated with a marginally significantly higher initiation of denture use among disaster survivors (IRR = 1.28; 95% CI, 0.99-1.66). The experience of major housing damage was independently associated with higher initiation of denture use (IRR = 1.77; 95% CI, 1.47-2.14). Improved geographic access to dental clinics may increase dental visits of disaster survivors. Further studies in non-disaster-affected areas are needed to generalize these findings.

Adverse Drug Event Profile Associated with Anti-dementia Drugs: Analysis of a Spontaneous Reporting Database

Adverse drug events (ADEs) rates associated with anti-dementia acetylcholinesterase inhibitors are estimated to be 5%-20% and show a wide range of symptoms. No report has examined whether there is a difference in the anti-dementia drugs' ADEs profile. This study aimed to establish whether anti-dementia drugs' ADEs profile differed. Data was based on the Japanese Adverse Drug Event Report (JADER) database. The reporting odds ratios (RORs) was used to analyze data for ADEs from April 2004-October 2021. The target drugs were donepezil, rivastigmine, galantamine, and memantine. The top ten most frequently occurring adverse events were selected. The association between the RORs and antidementia drug ADEs was evaluated, and compared the distribution rate of expression age related to ADEs and each ADEs' timing of onset due to anti-dementia drugs. The primary outcome was RORs. Secondary outcome were expression age and time-to-onset of ADE associated with anti-dementia drugs. A total of 705,294 reports were analyzed. The adverse events incidence differed. Bradycardia, loss of consciousness, falls, and syncope incidence were significantly diverse. The Kaplan-Meier curve results for the cumulative ADEs incidence showed that donepezil had the slowest onset, while galantamine, rivastigmine, and memantine had approximately the same timing of onset.

Identification of lthB, a Gene Encoding a Putative Glycosyltransferase Family 8 Protein Required for Leptothrix Sheath Formation

The bacterium Leptothrix cholodnii generates cell chains encased in sheaths that are composed of woven nanofibrils. The nanofibrils are mainly composed of glycoconjugate repeats, and several glycosyltransferases (GTs) are required for its biosynthesis. However, only one GT (LthA) has been identified to date. In this study, we screened spontaneous variants of L. cholodnii SP6 to find those that form smooth colonies, which is one of the characteristics of sheathless variants. Genomic DNA sequencing of an isolated variant revealed an insertion in the locus Lcho_0972, which encodes a putative GT family 8 protein. We thus designated this protein LthB and characterized it using deletion mutants and antibodies. LthB localized adjacent to the cell envelope. ΔlthB cell chains were nanofibril free and thus sheathless, indicating that LthB is involved in nanofibril biosynthesis. Unlike the ΔlthA mutant and the wild-type strain, which often generate planktonic cells, most ΔlthB organisms presented as long cell chains under static conditions, resulting in deficient pellicle formation, which requires motile planktonic cells. These results imply that sheaths are not required for elongation of cell chains. Finally, calcium depletion, which induces cell chain breakage due to sheath loss, abrogated the expression of LthA, but not LthB, suggesting that these GTs cooperatively participate in glycoconjugate biosynthesis under different signaling controls. IMPORTANCE In recent years, the regulation of cell chain elongation of filamentous bacteria via extracellular signals has attracted attention as a potential strategy to prevent clogging of water distribution systems and filamentous bulking of activated sludge in industrial settings. However, a fundamental understanding of the ecology of filamentous bacteria remains elusive. Since sheath formation is associated with cell chain elongation in most of these bacteria, the molecular mechanisms underlying nanofibril sheath formation, including the intracellular signaling cascade in response to extracellular stimuli, must be elucidated. Here, we isolated a sheathless variant of L. cholodnii SP6 and thus identified a novel glycosyltransferase, LthB. Although mutants with deletions of lthA, encoding another GT, and lthB were both defective for nanofibril formation, they exhibited different phenotypes of cell chain elongation and pellicle formation. Moreover, LthA expression, but not LthB expression, was influenced by extracellular calcium, which is known to affect nanofibril formation, indicating the functional diversities of LthA and LthB. Such molecular insights are critical for a better understanding of ecology of filamentous bacteria, which, in turn, can be used to improve strategies to control filamentous bacteria in industrial facilities.

Development of a Gene Replacement Method for the Filamentous BacteriumLeptothrix cholodniiSP-6

Genetic strategies such as gene disruption and fluorescent protein tagging largely contribute to understanding the molecular mechanisms of biological functions in bacteria. However, the methods for gene replacement remain underdeveloped for the filamentous bacteriaLeptothrix cholodniiSP-6. Their cell chains are encased in sheath composed of entangled nanofibrils, which may prevent the conjugation for gene transfer. Here, we describe a protocol optimized for gene disruption through gene transfer mediated by conjugation withEscherichia coliS17-1 with details on cell ratio, sheath removal, and loci validation. The obtained deletion mutants for specific genes can be used to clarify the biological functions of the proteins encoded by the target genes. Graphical overview.

Intracellular Phage Tail-Like Nanostructures Affect Susceptibility of Streptomyces lividans to Osmotic Stress

Contractile injection systems (CISs) are a large group of phage tail-like nanostructures conserved among bacteria. Despite their wide distribution, the biological significance of CISs in bacteria remains largely unclear except for a few unicellular bacteria. Here, we show that Streptomyces lividans-a model organism of filamentous Gram-positive bacteria with highly conserved CIS-related gene clusters-produces intracellular CIS-like nanostructures (Streptomyces phage tail-like particles [SLPs]) that affect phenotypes of this bacterium under hyperosmotic conditions. In contrast to typical CISs released from the cells, SLPs are localized in the cytoplasm of S. lividans. In addition, loss of SLPs leads to (i) delayed erection of aerial mycelia on hyperosmotic solid medium and (ii) decreased growth during the transition from exponential growth phase to stationary phase in hyperosmotic liquid medium. Localization of fluorescent protein-tagged SLPs showed partial correlation with cell wall synthesis-related proteins, including MreB, an actin-like cytoskeleton protein. Our pulldown assay and subsequent quantitative proteome analysis also suggest that 30S ribosomal proteins and cell wall-related proteins, including MreB, are coeluted with SLPs. Furthermore, an interaction assay using the recombinant proteins revealed a direct interaction between a sheath protein of SLP and ribosomal protein S16. Results of cross-linking experiments show indirect interactions between SLPs and translation elongation factors. These findings collectively suggest that SLPs are directly or indirectly associated with a protein interaction network within the cytoplasm of S. lividans and that SLP loss ultimately affects the susceptibility of the bacterium to certain stress conditions. IMPORTANCE Recent bioinformatic analyses have revealed that CIS-related gene clusters are highly conserved in Gram-positive actinomycetes, especially members of the genus Streptomyces known for their ability to produce therapeutic antibiotics. While typical CISs are released from the cells and can act as protein translocation systems that inject effector proteins into the target cells, our results indicate the unique intracellular localization of SLPs, CIS-related nanostructures produced by S. lividans. In addition, the direct and indirect interactions of SLPs with cytoplasmic proteins and SLP localization within specific regions of mycelia suggest that the biological significance of SLPs is related to intracellular processes. Further, SLP loss leads to increased susceptibility of S. lividans to osmotic stress, suggesting that production of these phage tail-like nanostructures ultimately affects the fitness of the bacterium under certain stress conditions. This work will provide new insight into the phage tail-like nanostructures highly conserved in Streptomyces species.

Development of low-temperature and ultrahigh-vacuum photoinduced force microscopy

In this paper, we develop optical and electronic systems for photoinduced force microscopy (PiFM) that can measure photoinduced forces under low temperature and ultrahigh vacuum (LT-UHV) without artifacts. For our LT-UHV PiFM, light is irradiated from the side on the tip-sample junction, which can be adjusted through the combination of an objective lens inside the vacuum chamber and a 90° mirror outside the vacuum chamber. We measured photoinduced forces due to the electric field enhancement between the tip and the Ag surface, and confirmed that photoinduced force mapping and measurement of photoinduced force curves were possible using the PiFM that we developed. The Ag surface was used to measure the photoinduced force with high sensitivity, and it is effective in enhancing the electric field using the plasmon gap mode between the metal tip and the metal surface. Additionally, we confirmed the necessity of Kelvin feedback during the measurement of photoinduced forces, to avoid artifacts due to electrostatic forces, by measuring photoinduced forces on organic thin films. The PiFM, operating under low temperature and ultrahigh vacuum developed here, is a promising tool to investigate the optical properties of various materials with very high spatial resolution.

Search for Gamma-Ray Spectral Lines from Dark Matter Annihilation up to 100 TeV toward the Galactic Center with MAGIC

Linelike features in TeV γ rays constitute a "smoking gun" for TeV-scale particle dark matter and new physics. Probing the Galactic Center region with ground-based Cherenkov telescopes enables the search for TeV spectral features in immediate association with a dense dark matter reservoir at a sensitivity out of reach for satellite γ-ray detectors, and direct detection and collider experiments. We report on 223 hours of observations of the Galactic Center region with the MAGIC stereoscopic telescope system reaching γ-ray energies up to 100 TeV. We improved the sensitivity to spectral lines at high energies using large-zenith-angle observations and a novel background modeling method within a maximum-likelihood analysis in the energy domain. No linelike spectral feature is found in our analysis. Therefore, we constrain the cross section for dark matter annihilation into two photons to ⟨σv⟩≲5×10^{-28}  cm^{3} s^{-1} at 1 TeV and ⟨σv⟩≲1×10^{-25}  cm^{3} s^{-1} at 100 TeV, achieving the best limits to date for a dark matter mass above 20 TeV and a cuspy dark matter profile at the Galactic Center. Finally, we use the derived limits for both cuspy and cored dark matter profiles to constrain supersymmetric wino models.

How brain diseases affect the lower urinary tract function?

This article reviewed brain mechanism of the lower urinary tract (LUT). Among autonomic nervous systems, LUT is unique in terms of afferent pathophysiology; bladder sensation is perceived soon after the storage phase and throughout the voiding phase. Within the brain, this is measured in experimental animals by the firing of single neurons and in humans by evoked potentials/functional neuroimaging. The evidence indicates that sphincter information goes up to the precentral motor cortex and other brain areas, and bladder information goes up to the insular cortex (IC)/anterior cingulate (ACG) and further to the prefrontal cortex (PFC). Another LUT-specific phenomenon is efferent pathophysiology: detrusor overactivity (exaggerated micturition reflex) occurs in brain diseases such as stroke (focal disease) and dementia with Lewy bodies (diffuse diseases, may overlap with each other). With the turning off and on of the brain-switch of micturition (at the periaqueductal gray [PAG]), there is a bladder-inhibitory PFC-IC/ACG-hypothalamus-PAG pathway, with interconnections via the PFC with a PFC-nigrostriatal D1 dopaminergic pathway and a PFC-cerebellar pathway. Brain diseases that affect these areas may cause a loss of the brain's inhibition of the micturition reflex, leading to detrusor overactivity. This has a significant clinical impact on patients and requires appropriate management.

Utility of movement disorder society-unified Parkinson's disease rating scale for evaluating effect of subthalamic nucleus deep brain stimulation

Background

The Movement Disorders Society (MDS)-Unified Parkinson's Disease Rating Scale (UPDRS) is increasingly used to assess motor dysfunction before and after subthalamic nucleus deep brain stimulation (STN-DBS).

Objectives

We, therefore, investigated whether the MDS-UPDRS can detect longitudinal changes in motor function after STN-DBS in the same way as UPDRS.

Methods

We examined 21 patients with Parkinson's disease (PD) (mean age 59.2 ± 10.6 years, mean disease duration 12.0 ± 3.0 years) who underwent STN-DBS and whose motor functions were assessed by the UPDRS and MDS-UPDRS before, 3 months after, and 1 year after STN-DBS. We then evaluated the consistency between the scores of Parts II and III of the UPDRS and MDS-UPDRS during the off phase using Lin's concordance coefficient (LCC) and a Bland-Altman plot.

Results

The scores of Parts II and III of both the UPDRS and MDS-UPDRS were significantly decreased 3 months and 1 year after STN-DBS during the off phase. Scores of the UPDRS and MDS-UPDRS showed significant positive correlations before and after STN-DBS. We calculated estimated MDS-UPDRS scores from the UPDRS scores using a regression line and calculated the LCC between the MDS-UPDRS and the estimated MDS-UPDRS scores. The LCC value was 0.59-0.91, which suggests a relatively high consistency between the UPDRS and MDS-UPDRS. The Bland-Altman plot showed that differences between both scores were basically within ±1.96 standard deviations of the difference.

Conclusion

The present preliminary study indicated that the utility of the MDS-UPDRS in evaluating motor function before and after STN-DBS demonstrates its potential equivalency to the UPDRS.

How brain diseases affect the lower urinary tract function?

This article reviewed brain mechanism of the lower urinary tract (LUT). Among autonomic nervous systems, LUT is unique in terms of afferent pathophysiology; bladder sensation is perceived soon after the storage phase and throughout the voiding phase. Within the brain, this is measured in experimental animals by the firing of single neurons and in humans by evoked potentials/functional neuroimaging. The evidence indicates that sphincter information goes up to the precentral motor cortex and other brain areas, and bladder information goes up to the insular cortex (IC)/anterior cingulate (ACG) and further to the prefrontal cortex (PFC). Another LUT-specific phenomenon is efferent pathophysiology: detrusor overactivity (exaggerated micturition reflex) occurs in brain diseases such as stroke (focal disease) and dementia with Lewy bodies (diffuse diseases, may overlap with each other). With the turning off and on of the brain-switch of micturition (at the periaqueductal gray [PAG]), there is a bladder-inhibitory PFC-IC/ACG-hypothalamus-PAG pathway, with interconnections via the PFC with a PFC-nigrostriatal D1 dopaminergic pathway and a PFC-cerebellar pathway. Brain diseases that affect these areas may cause a loss of the brain's inhibition of the micturition reflex, leading to detrusor overactivity. This has a significant clinical impact on patients and requires appropriate management.

Hybrid-mode atomic force microscopy of phase modulation and frequency modulation

We propose hybrid-phase modulation (PM)/frequency modulation (FM)-atomic force microscopy (AFM) to increase the imaging speed of AFM in high-Q environments. We derive the relationship among the phase shift, the frequency shift, and tip—sample interaction force from the equation of motion for the cantilever in high-Q environments. The tip—sample conservative force is approximately given by the sum of the conservative force with respect to the phase shift in the PM mode and that with respect to the frequency shift in the FM mode. We preliminarily demonstrate that the hybrid-PM/FM-AFM is a new and very promising AFM operation mode that can increase imaging speed.

Development and validation of machine learning-based predictive model for clinical outcome of decompression surgery for lumbar spinal canal stenosis

BACKGROUND CONTEXT

Although the results of decompression surgery for lumbar spinal canal stenosis (LSS) are favorable, it is still difficult to predict the postoperative health-related quality of life of patients before surgery.

PURPOSE

The purpose of this study was to develop and validate a machine learning model to predict the postoperative outcome of decompression surgery for patients with LSS.

STUDY DESIGN/SETTING

A multicentered retrospective study.

PATIENT SAMPLE

A total of 848 patients who underwent decompression surgery for LSS at an academic hospital, tertiary center, and private hospital were included (age 71±9 years, 68% male, 91% LSS, level treated 1.8±0.8, operation time 69±37 minutes, blood loss 48±113 mL, and length of hospital stay 12±5 days).

OUTCOME MEASURES

Baseline and 2 years postoperative health-related quality of life.

METHODS

The subjects were randomly assigned in a 7:3 ratio to a model building cohort and a testing cohort to test the models' accuracy. Twelve predictive algorithms using 68 preoperative factors were used to predict each domain of the Japanese Orthopedic Association Back Pain Evaluation Questionnaire and visual analog scale scores at 2 years postoperatively. The final predictive values were generated using an ensemble of the top five algorithms in prediction accuracy.

RESULTS

The correlation coefficients of the top algorithms for each domain established using the preoperative factors were excellent (correlation coefficient: 0.95-0.97 [relative error: 0.06-0.14]). The performance evaluation of each Japanese Orthopedic Association Back Pain Evaluation Questionnaire domain and visual analog scale score by the ensemble of the top five algorithms in the testing cohort was favorable (mean absolute error [MAE] 8.9-17.4, median difference [MD] 8.1-15.6/100 points), with the highest accuracy for mental status (MAE 8.9, MD 8.1) and the lowest for buttock and leg numbness (MAE 1.7, MD 1.6/10 points). A strong linear correlation was observed between the predicted and measured values (linear correlation 0.82-0.89), while 4% to 6% of the subjects had predicted values of greater than±3 standard deviations of the MAE.

CONCLUSIONS

We successfully developed a machine learning model to predict the postoperative outcomes of decompression surgery for patients with LSS using patient data from three different institutions in three different settings. Thorough analyses for the subjects with deviations from the actual measured values may further improve the predictive probability of this model.

Albuminuria predicts worsening renal function at 1 month after transcatheter aortic valve implantation

Background

Worsening renal function (WRF) at short term after transcatheter aortic valve replacement (TAVR) is reportedly associated with subsequent poor clinical outcomes. We previously reported preoperative albuminuria predicted mid-term WRF after TAVR. However, the impact of albuminuria on short-term WRF after TAVR is unknown.

Methods

Overall, 206 patients who underwent TAVR for severe aortic stenosis were investigated retrospectively. Patients were divided into two groups according to the preoperative urinary albumin-to-creatinine ratio (ACR): high (ACR ≥30 mg/g) and low (ACR <30 mg/g). The incidence of 1-month WRF, defined as a decrease in estimated glomerular filtration rate (eGFR) ≥10% from baseline after TAVR, was investigated.

Results

More than half of the examined patients (n=106/206 [51.5%]) had a high ACR. Patients with a high ACR had significantly lower eGFR at baseline than those with a low ACR (49.3 [32.5–57.8] vs. 51.4 [37.7–65.8] mL/min/1.73 m2, p=0.03). Patients with a high ACR more frequently had acute kidney injury (8.5% vs. 1.0%, p=0.01) and 1-month WRF than those with a low ACR (29.2% vs. 12.0%, p=0.002). Multivariate analysis showed a high ACR was independently associated with 1-month WRF (odds ratio, 3.72; 95% confidence interval, 1.72–8.08; p<0.001). Meanwhile, eGFR <60 mL/min/1.73 m2 was not an independent predictor of 1-month WRF (odds ratio, 0.48; 95% confidence interval, 0.22–1.04, p=0.06).

Conclusions

Patients with preoperative albuminuria (ACR ≥30 mg/g) more frequently developed AKI and 1-month WRF after TAVR. Albuminuria was independently associated with 1-month WRF, suggesting albuminuria to be a universal predictor of deterioration of renal function at various time points.

Funding Acknowledgement

Type of funding sources: None.

Effects of liquid diet intake on nerve growth in salivary glands of growing rats

BACKGROUND

The growth of parotid glands is inhibited by liquid diet intake during growing period, while that of submandibular glands is not affected. This study examined how liquid diet intake affects nerve growth in the parotid and submandibular glands of growing rats, because nerves are closely involved in the maintenance of salivary gland structure.

MATERIALS AND METHODS

Male Wistar rats were weaned at 21 days of age. Then, rats were fed a pellet diet and a liquid diet in the control group and experimental group, respectively. At 0, 2, 4, or 8 weeks, they were euthanised by isoflurane overdose, and parotid and submandibular glands were removed. The frozen sections were made and immuno-stained with anti-protein gene product 9.5 (PGP 9.5) antibody (general nerve marker), anti-tyrosine hydroxylase (TH) antibody (sympathetic nerve marker), or anti-neuronal nitric oxide synthase (nNOS) antibody (parasympathetic nerve marker).

RESULTS

In control parotid glands, scattered punctate or short linear patterns of PGP 9.5-positive sites were observed at week 0. After 2 weeks, PGP 9.5-positive sites, some of which were arranged in long linear patterns, increased in number. There were some TH-positive sites at week 0. After 2 weeks, there were increasing numbers of TH-positive sites, often in long linear patterns. At week 0, there were very few nNOS-positive sites, and nNOS immunoreactivity increased over time. After week 4, they demonstrated linear patterns. In the experimental parotid glands, there were fewer PGP 9.5- and nNOS-positive sites than in control parotid glands at each time point, although TH immunoreactivity was similar between two groups at each time point. In control submandibular glands, few punctate exhibited PGP 9.5-positive site were observed at week 0. At week 4, PGP 9.5 immunoreaction increased and showed linear patterns. TH-positive sites demonstrated punctate or short linear patterns at week 0, and thereafter TH immunoreactivity increased and were arranged in long linear patterns. There were few nNOS-positive sites at week 0, and they gradually increased after week 4. The immunoreactivities of all antibodies in the experimental submandibular glands were similar to those in the control at each time point.

CONCLUSIONS

Parasympathetic nerve growth in rat parotid glands was inhibited by liquid diet intake during the growth period, while liquid diet intake did not affect parasympathetic nerve growth nor sympathetic nerve growth in rat submandibular glands.

Validation of a Denoising Method Using Deep Learning-Based Reconstruction to Quantify Multiple Sclerosis Lesion Load on Fast FLAIR Imaging

BACKGROUND AND PURPOSE

Accurate quantification of WM lesion load is essential for the care of patients with multiple sclerosis. We tested whether the combination of accelerated 3D-FLAIR and denoising using deep learning-based reconstruction could provide a relevant strategy while shortening the imaging examination.

MATERIALS AND METHODS

Twenty-eight patients with multiple sclerosis were prospectively examined using 4 implementations of 3D-FLAIR with decreasing scan times (4 minutes 54 seconds, 2 minutes 35 seconds, 1 minute 40 seconds, and 1 minute 15 seconds). Each FLAIR sequence was reconstructed without and with denoising using deep learning-based reconstruction, resulting in 8 FLAIR sequences per patient. Image quality was assessed with the Likert scale, apparent SNR, and contrast-to-noise ratio. Manual and automatic lesion segmentations, performed randomly and blindly, were quantitatively evaluated against ground truth using the absolute volume difference, true-positive rate, positive predictive value, Dice similarity coefficient, Hausdorff distance, and F1 score based on the lesion count. The Wilcoxon signed-rank test and 2-way ANOVA were performed.

RESULTS

Both image-quality evaluation and the various metrics showed deterioration when the FLAIR scan time was accelerated. However, denoising using deep learning-based reconstruction significantly improved subjective image quality and quantitative performance metrics, particularly for manual segmentation. Overall, denoising using deep learning-based reconstruction helped to recover contours closer to those from the criterion standard and to capture individual lesions otherwise overlooked. The Dice similarity coefficient was equivalent between the 2-minutes-35-seconds-long FLAIR with denoising using deep learning-based reconstruction and the 4-minutes-54-seconds-long reference FLAIR sequence.

CONCLUSIONS

Denoising using deep learning-based reconstruction helps to recognize multiple sclerosis lesions buried in the noise of accelerated FLAIR acquisitions, a possibly useful strategy to efficiently shorten the scan time in clinical practice.