Effectiveness of Semi-Supervised Active Learning in Automated Wound Image Segmentation

Appropriate wound management shortens the healing times and reduces the management costs, benefiting the patient in physical terms and potentially reducing the healthcare system’s economic burden. Among the instrumental measurement methods, the image analysis of a wound area is becoming one of the cornerstones of chronic ulcer management. Our study aim is to develop a solid AI method based on a convolutional neural network to segment the wounds efficiently to make the work of the physician more efficient, and subsequently, to lay the foundations for the further development of more in-depth analyses of ulcer characteristics. In this work, we introduce a fully automated model for identifying and segmenting wound areas which can completely automatize the clinical wound severity assessment starting from images acquired from smartphones. This method is based on an active semi-supervised learning training of a convolutional neural network model. In our work, we tested the robustness of our method against a wide range of natural images acquired in different light conditions and image expositions. We collected the images using an ad hoc developed app and saved them in a database which we then used for AI training. We then tested different CNN architectures to develop a balanced model, which we finally validated with a public dataset. We used a dataset of images acquired during clinical practice and built an annotated wound image dataset consisting of 1564 ulcer images from 474 patients. Only a small part of this large amount of data was manually annotated by experts (ground truth). A multi-step, active, semi-supervised training procedure was applied to improve the segmentation performances of the model. The developed training strategy mimics a continuous learning approach and provides a viable alternative for further medical applications. We tested the efficiency of our model against other public datasets, proving its robustness. The efficiency of the transfer learning showed that after less than 50 epochs, the model achieved a stable DSC that was greater than 0.95. The proposed active semi-supervised learning strategy could allow us to obtain an efficient segmentation method, thereby facilitating the work of the clinician by reducing their working times to achieve the measurements. Finally, the robustness of our pipeline confirms its possible usage in clinical practice as a reliable decision support system for clinicians.     

Impact of Freeze-Drying on Cord Blood (CB), Serum (S), and Platelet-Rich Plasma (CB-PRP) Preparations on Growth Factor Content and In Vitro Cell Wound Healing

Blood-based preparations are used in clinical practice for the treatment of several eye disorders. The aim of this study is to analyze the effect of freeze-drying blood-based preparations on the levels of growth factors and wound healing behaviors in an in vitro model. Platelet-rich plasma (PRP) and serum (S) preparations from the same Cord Blood (CB) sample, prepared in both fresh frozen (FF) and freeze-dried (FD) forms (and then reconstituted), were analyzed for EGF and BDNF content (ELISA Quantikine kit). The human MIO-M1 glial cell line (Moorfield/Institute of Ophthalmology, London, UK) was incubated with FF and FD products and evaluated for cell migration with scratch-induced wounding (IncuCyte S3 Essen BioScience), proliferation with cyclin A2 and D1 gene expression, and activation with vimentin and GFAP gene expression. The FF and FD forms showed similar concentrations of EGF and BDNF in both the S and PRP preparations. The wound healing assay showed no significant difference between the FF and FD forms for both S and PRP. Additionally, cell migration, proliferation, and activation did not appear to change in the FD forms compared to the FF ones. Our study showed that reconstituted FD products maintained the growth factor concentrations and biological properties of FF products and could be used as a functional treatment option.     

Visual simulation of shockwaves

We present an efficient method for visual simulations of shock phenomena in compressible, inviscid fluids. Our algorithm is derived from one class of the finite volume method especially designed for capturing shock propagation, but offers improved efficiency through physically-based simplification and adaptation for graphical rendering. Our technique is capable of handling complex, bidirectional object–shock interactions stably and robustly. We describe its applications to various visual effects, including explosion, sonic booms and turbulent flows. Furthermore, we explore parallelization schemes and demonstrate the scalability of our method on shared-memory, multi-core architectures.     

A comparative study of a fluidised bed reactor and a gas lift loop reactor for the IBE process: Part I. Reactor design and scale down approach

For the continuous production of isopropanol-butanol mixtures by immobilised Clostridium spp. (the IBE process) two reactor types were studied: a fluidised bed reactor with liquid recycling (FBR) and an external loop gas lift reactor (GLR). A large scale design (50–65 m³) was made for both reactors. A regime analysis, by evaluating the time constants for e.g. mixing and conversion, identified the ruling regime. Via the scale down approach two representative model reactors were developed: a 10 dm³ FBR reactor (H/D=25, D=0.08 m) and a 15 dm³ external loop GLR (H/D = 12.5, D_r=D_d=0.08 m). For both reactors the hydrodynamical behaviour and the total reactor performance were studied and are described in parts II and III of this study.     

Development and application of C60-fullerene bound silica for solid-phase extraction of biomolecules

Sample pretreatment is the most important procedure to remove the matrix for interfacing with mass spectrometry (MS). Additionally, for the samples with low concentration, the process of preconcentration is required before MS analysis. We have newly developed a solid-phase extraction stationary phase based on C60-fullerene covalently bound to silica for purification of biomolecules of different characteristics. Silica particles of different porosity are modified with aminopropyl linker and then covalently bound to C60-fullerenoacetic acid or C60-epoxyfullerenes. The developed materials have been successfully applied as an alternative to commercially available reversed-phase materials for solid-phase extraction. C60-fullerene silica is able to retain small and hydrophilic molecules like phosphopeptides, which can be easily lost by reversed-phase sorbents. The novel materials are applied for desalting and preconcentration of proteins and peptides, especially phosphopeptides. In addition, the C60-fullerene silica is applied for the solid-phase extraction of selected flavonoids with recoveries of approximately 99%. The recoveries are compared with the commercially available solid-phase extraction materials.     

The alpha-to-beta conformational transition of Alzheimer's Abeta-(1-42) peptide in aqueous media is reversible: a step by step conformational analysis suggests the location of beta conformation seeding

Current views of the role of beta-amyloid (Abeta) peptide fibrils range from regarding them as the cause of Alzheimer's pathology to having a protective function. In the last few years, it has also been suggested that soluble oligomers might be the most important toxic species. In all cases, the study of the conformational properties of Abeta peptides in soluble form constitutes a basic approach to the design of molecules with "antiamyloid" activity. We have experimentally investigated the conformational path that can lead the Abeta-(1-42) peptide from the native state, which is represented by an alpha helix embedded in the membrane, to the final state in the amyloid fibrils, which is characterized by beta-sheet structures. The conformational steps were monitored by using CD and NMR spectroscopy in media of varying polarities. This was achieved by changing the composition of water and hexafluoroisopropanol (HFIP). In the presence of HFIP, beta conformations can be observed in solutions that have very high water content (up to 99 % water; v/v). These can be turned back to alpha helices simply by adding the appropriate amount of HFIP. The transition of Abeta-(1-42) from alpha to beta conformations occurs when the amount of water is higher than 80 % (v/v). The NMR structure solved in HFIP/H2O with high water content showed that, on going from very apolar to polar environments, the long N-terminal helix is essentially retained, whereas the shorter C-terminal helix is lost. The complete conformational path was investigated in detail with the aid of molecular-dynamics simulations in explicit solvent, which led to the localization of residues that might seed beta conformations. The structures obtained might help to find regions that are more affected by environmental conditions in vivo. This could in turn aid the design of molecules able to inhibit fibril deposition or revert oligomerization processes.     

Heterogeneous oligonucleotide-hybridization assay based on hot electron-induced electrochemiluminescence of a rhodamine label at oxide-coated aluminum and silicon electrodes

This paper describes a heterogeneous oligonucleotide-hybridization assay based on hot electron-induced electrochemiluminescence (HECL) of a rhodamine label. Thin oxide-film coated aluminum and silicon electrodes were modified with an aminosilane layer and derivatized with short, 15-mer oligonucleotides via diisothiocyanate coupling. Target oligonucleotides were conjugated with tetramethylrhodamine (TAMRA) dye at their amino modified 5′ end and hybridization was detected using HECL of TAMRA. Preliminary results indicate sensitivity down to picomolar level and low nonspecific adsorption. The sensitivity was better on oxide-coated silicon compared to oxide-coated aluminum electrodes and two-base pair mismatched hybrids were successfully discriminated. The experimental results presented here might be useful for the design of disposable electrochemiluminescent DNA biosensors.