Assessment of frailty in elderly patients attending a multidisciplinary wound care centre: a cohort study.

BACKGROUND: The incidence of frailty and non-healing wounds increases with patients' age. Knowledge of the relationship between frailty and wound healing progress is greatly lacking. METHODS: The aim of this study is to characterize the degree of frailty in elderly patients attending a multidisciplinary wound care centres (MWCC). Additionally, we seek to assess the impact of frailty on the wound healing rate and wound healing time. An open cohort study was conducted on 51 consecutive patients aged > 70 years treated for wounds at an MWCC of an intermediate care hospital. The frailty score was determined according to the Frail-VIG index. Data were collected through patient questionnaires at the beginning of the study, and at 6 months or upon wound healing. Wounds were followed up every 2 weeks. To analyse the relationship between two variables was used the Chi-square test and Student's or the ANOVA model. The t-test for paired data was used to analyse the evolution of the frailty index during follow-up. RESULTS: A total of 51 consecutive patients were included (aged 81.1 ± 6.1 years). Frailty prevalence was 74.5% according to the Frail-VIG index (47.1% mildly frail, 19.6% moderately frail, and 7.8% severely frail). Wounds healed in 69.6% of cases at 6 months. The frailty index (FI) was higher in patients with non-healing wounds in comparison with patients with healing wounds (IF 0.31 ± 0.15 vs IF 0.24 ± 0.11, p = 0.043). A strong correlation between FI and wound healing results was observed in patients with non-venous ulcers (FI 0.37 ± 0.13 vs FI 0.27 ± 0.10, p = 0.015). However, no correlation was observed in patients with venous ulcers (FI 0.17 ± 0.09 vs FI 0.19 ± 0.09, p = 0.637). Wound healing rate is statically significantly higher in non-frail patients (8.9% wound reduction/day, P25-P75 3.34-18.3%/day;AQ6 p = 0.044) in comparison with frail patients (3.26% wound reduction/day, P25-P75 0.8-8.8%/day). CONCLUSION: Frailty is prevalent in elderly patients treated at an MWCC. Frailty degree is correlated with wound healing results and wound healing time.

Convolutional Neural Network for Skin Lesion Classification: Understanding the Fundamentals Through Hands-On Learning.

Deep learning architectures for the classification of images have shown outstanding results in a variety of disciplines, including dermatology. The expectations generated by deep learning for, e.g., image-based diagnosis have created the need for non-experts to become familiar with the working principles of these algorithms. In our opinion, getting hands-on experience with these tools through a simplified but accurate model can facilitate their understanding in an intuitive way. The visualization of the results of the operations performed by deep learning algorithms on dermatological images can help students to grasp concepts like convolution, even without an advanced mathematical background. In addition, the possibility to tune hyperparameters and even to tweak computer code further empower the reach of an intuitive comprehension of these processes, without requiring advanced computational and theoretical skills. This is nowadays possible thanks to recent advances that have helped to lower technical and technological barriers associated with the use of these tools, making them accessible to a broader community. Therefore, we propose a hands-on pedagogical activity that dissects the procedures to train a convolutional neural network on a dataset containing images of skin lesions associated with different skin cancer categories. The activity is available open-source and its execution does not require the installation of software. We further provide a step-by-step description of the algorithm and of its functions, following the development of the building blocks of the computer code, guiding the reader through the execution of a realistic example, including the visualization and the evaluation of the results.

A toolkit for the quantitative evaluation of chronic wounds evolution for early detection of non-healing wounds.

BACKGROUND: Chronic wounds resulting from a number of conditions do not heal properly and can pose serious health problems. Beyond clinician visual inspection, an objective evaluation of the wound is required to assess wound evolution and the effectiveness of therapies. AIM: Our objective is to provide a methodology for the analysis of wound area vs. time for the early prediction of non-healing wounds evolution. METHODS: We propose a two-step approach consisting of: i) wound area quantification from planimetries and ii) classification of wound healing through the inference of characteristic parameters. For the first step, we describe a user-friendly software (Woundaries) to automatically calculate the wound area and other geometric parameters from hand-traced planimetries. For the second, we use a procedure for the objective classification of wound time evolution and the early assessment of treatment efficacy. The methodology was tested on simulations and retrospectively applied to data from 85 patients to compare the effect of a biological therapy with respect to general basic therapeutics. RESULTS: Woundaries provides measurements of wound surface equivalent to a validated device. The two-step methodology allows to determine if a wound is healing with high sensitivity, even with limited amount of data. Therefore, it allows the early assessment of the efficacy of a therapy. CONCLUSION: The performance of this methodology for the quantification and the objective evaluation of wound area evolution suggest it as a useful toolkit to assist clinicians in the early assessment of the efficacy of treatments, leading to a timely change of therapy.

Research Techniques Made Simple: Deep Learning for the Classification of Dermatological Images.

Deep learning is a branch of artificial intelligence that uses computational networks inspired by the human brain to extract patterns from raw data. Development and application of deep learning methods for image analysis, including classification, segmentation, and restoration, have accelerated in the last decade. These tools have been progressively incorporated into several research fields, opening new avenues in the analysis of biomedical imaging. Recently, the application of deep learning to dermatological images has shown great potential. Deep learning algorithms have shown performance comparable with humans in classifying skin lesion images into different skin cancer categories. The potential relevance of deep learning to the clinical realm created the need for researchers in disciplines other than computer science to understand its fundamentals. In this paper, we introduce the basics of a deep learning architecture for image classification, the convolutional neural network, in a manner accessible to nonexperts. We explain its fundamental operation, the convolution, and describe the metrics for the evaluation of its performance. These concepts are important to interpret and evaluate scientific publications involving these tools. We also present examples of recent applications for dermatology. We further discuss the capabilities and limitations of these artificial intelligence-based methods.

Bayesian analysis of data from segmented super-resolution images for quantifying protein clustering.

Super-resolution imaging techniques have largely improved our capabilities to visualize nanometric structures in biological systems. Their application further permits the quantitation relevant parameters to determine the molecular organization and stoichiometry in cells. However, the inherently stochastic nature of fluorescence emission and labeling strategies imposes the use of dedicated methods to accurately estimate these parameters. Here, we describe a Bayesian approach to precisely quantitate the relative abundance of molecular aggregates of different stoichiometry from segmented images. The distribution of proxies for the number of molecules in a cluster, such as the number of localizations or the fluorescence intensity, is fitted via a nested sampling algorithm to compare mixture models of increasing complexity and thus determine the optimum number of mixture components and their weights. We test the performance of the algorithm on in silico data as a function of the number of data points, threshold, and distribution shape. We compare these results to those obtained with other statistical methods, showing the improved performance of our approach. Our method provides a robust tool for model selection in fitting data extracted from fluorescence imaging, thus improving the precision of parameter determination. Importantly, the largest benefit of this method occurs for small-statistics or incomplete datasets, enabling an accurate analysis at the single image level. We further present the results of its application to experimental data obtained from the super-resolution imaging of dynein in HeLa cells, confirming the presence of a mixed population of cytoplasmic single motors and higher-order structures.