Use of a convolutional neural network and quantitative ultrasound for diagnosis of fatty liver

Quantitative ultrasound (QUS) was used to classify rabbits that were induced to have liver disease by placing them on a fatty diet for a defined duration and/or periodically injecting them with CCl₄. The ground truth of the liver state was based on lipid liver percents estimated via the Folch assay and hydroxyproline concentration to quantify fibrosis. Rabbits were scanned ultrasonically in vivo using a SonixOne scanner and an L9-4/38 linear array. Liver fat percentage was classified based on the ultrasonic backscattered radiofrequency (RF) signals from the livers using either QUS or a 1-D convolutional neural network (CNN). Use of QUS parameters with linear regression and canonical correlation analysis demonstrated that the QUS parameters could differentiate between livers with lipid levels above or below 5%. However, the QUS parameters were not sensitive to fibrosis. The CNN was implemented by analyzing raw RF ultrasound signals without using separate reference data. The CNN outputs the classification of liver as either above or below a threshold of 5% fat level in the liver. The CNN outperformed the classification utilizing the QUS parameters combined with a support vector machine in differentiating between low and high lipid liver levels (i.e., accuracies of 74% versus 59% on the testing data). Therefore, although the CNN did not provide a physical interpretation of the tissue properties (e.g., attenuation of the medium or scatterer properties) the CNN had much higher accuracy in predicting fatty liver state and did not require an external reference scan.     

Biomarkers for traumatic brain injury: a short review

Neurosurgical Review - Cellular response to TBI is a mixture of excitotoxicity, neuroinflammation, and cell death. Biomarkers that can track these lesions and inflammatory processes are being...     

Hardening of Additive Manufactured 316L Stainless Steel by Using Bimodal Powder Containing Nanoscale Fraction

The particle size distribution significantly affects the material properties of the additively manufactured parts. In this work, the influence of bimodal powder containing nano- and micro-scale particles on microstructure and materials properties is studied. Moreover, to study the effect of the protective atmosphere, the test samples were additively manufactured from 316L stainless steel powder in argon and nitrogen. The samples fabricated from the bimodal powder demonstrate a finer subgrain structure, regardless of protective atmospheres and an increase in the Vickers microhardness, which is in accordance with the Hall-Petch relation. The porosity analysis revealed the deterioration in the quality of as-built parts due to the poor powder flowability. The surface roughness of fabricated samples was the same regardless of the powder feedstock materials used and protective atmospheres. The results suggest that the improvement of mechanical properties is achieved by adding a nano-dispersed fraction, which dramatically increases the total surface area, thereby contributing to the nitrogen absorption by the material.     

An update on epidemiology,diagnosis & management of NMO-SD in the tertiary neurology department of Marrakech (Morocco)

IntroductionNeuromyelitis optica (NMO) and NMO spectrum disorders (NMO-SD) are inflammatory demyelinating diseases of the central nervous system. There are few epidemiological studies devoted to NMO, especially in Africa and the Middle East, but individual cases and series have been reported from many countries across the African continent.ObjectivesTo describe the epidemiology, diagnosis, and management of NMO patients followed at the Mohammed VI University Hospital of Marrakech.Patients and methodsThis was a hospital-based retrospective study of 52 patients with NMO diagnosed and followed at the Neurology department of the University Hospital of Marrakech from 2004 to July 2019. The 2006 diagnostic criteria of NMOSD were used for patients admitted before 2015 for inflammatory disease of the central nervous system and the 2015 diagnostic criteria of NMO-SD for all patients thereafter. Collected data were analysed using SPSS software.ResultsThe study concerned 52 patients, 18 males and 34 females. Median age at disease onset was 32.5 years (range 7–55). Mean time between symptom onset and diagnosis of NMO was nine months 18 days (range 7 days to 4 years). In most patients, manifestations included visual acuity, tetraparesis, and sensorial disorders. Refractory vomiting and hiccup were noted in the first attack in 19% of patients. Two patients had hypersomnia and polyphagia, and one had been treated for depression ten months before the development of severe tetraplegia. Magnetic resonance imaging did not show any brain lesions in 29% of patients. Cervical myelitis extending to more than three vertebrae was found in 60% of patients. AQP4-antibody assay was performed only in 57.7% of patients, and was positive in 38.4%; anti-MOG was positive in four anti-AQP4 seronegative patients. Management strategies for NMO-SD included methylprednisolone pulses (70% of patients), plasmapheresis (25%), and rituximab (since 2017) for 46%. Outcome was favourable in 40% of patients and has remained stable in 50% of them.ConclusionAnti-NMO assays, made available during the last five years with the help of The Guthy-Jackson Charitable Foundation, have led to a clear jump in the number of cases diagnosed. Major advances in the field of epidemiology, imaging, and pathophysiology of NMO-SD have led to improved patient care and outcome.     

The proteome microenvironment determines the protective effect of preconditioning in cisplatin-induced acute kidney injury

1. Download high-res image (341KB)
2. Download full-size image