Biomedizinische Technik. Biomedical engineering
-
Intensive care units (ICUs) are responsible for generating a wealth of useful data in the form of electronic health records. We aimed to build a mortality prediction model on a Medical Information Mart for Intensive Care (MIMIC-III) database and to assess whether the use of deep learning techniques like long short-term memory (LSTM) can effectively utilize the temporal relations among clinical variables. The models were built on clinical variable dynamics of the first 48 h of ICU admission of 12,550 records from the MIMIC-III database. ⋯ For training and validation purpose, we have used International Classification of Diseases, 9th edition (ICD-9) codes for extracting the patients with cardiovascular disease, and infections and parasitic disease, respectively. The effectiveness of the LSTM model is achieved over non-recurrent baseline models like naïve Bayes, logistic regression (LR), support vector machine and multilayer perceptron (MLP) by generating state of the art results (area under the curve [AUC], 0.852). Next, by providing attention at each time stamp, we developed a model, LSTM-AT, which exhibits even better performance (AUC, 0.876).
-
Background and objective Spirometry, which is the most commonly used technique for asthma diagnosis, is often unsuitable for small children as it requires them to follow exact instructions and perform extreme inspiration and expiration maneuvers. In contrast, impulse oscillometry (IOS) is a child-friendly technique that could serve as an alternative pulmonary function test (PFT) for asthma diagnosis and control in children as it offers several advantages over spirometry. However, the complex test results of IOS may be difficult to be understood by practitioners due to its reliance on mechanical and electrical models of the human pulmonary system. ⋯ The most relevant results of the articles reviewed are related to the performance of the different classifiers using static features which are solely based on the first pulmonary function testing measurements (IOS and spirometry). These results included an overall classifiers' accuracy performance ranging from 42.24% to 98.61%. Conclusion There is still a great opportunity to improve the utility of IOS by developing more computer-aided robust classifiers, specifically for the asthmatic children population as the classification studies performed to date (1) are limited in number, (2) include features derived from tests that are not optimally suitable for children, (3) are solely bi-class (mostly asthma and non-asthma) and therefore fail to include different degrees of peripheral obstruction for disease prevention and control and (4) lack of validation in cases that focus on multi-class classification of the different degrees of peripheral airway obstruction.
-
In this paper, common time- and frequency-domain variability indexes obtained by pulse rate variability (PRV) series extracted from video-photoplethysmographic signal (vPPG) were compared with heart rate variability (HRV) parameters calculated from synchronized ECG signals. The dual focus of this study was to analyze the effect of different video acquisition frame-rates starting from 60 frames-per-second (fps) down to 7.5 fps and different video compression techniques using both lossless and lossy codecs on PRV parameters estimation. Video recordings were acquired through an off-the-shelf GigE Sony XCG-C30C camera on 60 young, healthy subjects (age 23±4 years) in the supine position. ⋯ Concerning the video compression, the results showed that compression techniques are suitable for the storage of vPPG recordings, although lossless or intra-frame compression are to be preferred over inter-frame compression methods. FFV1 performances are very close to the uncompressed (UNC) version with less than 45% disk size. H.264 showed a degradation of the PRV estimation directly correlated with the increase of the compression ratio.
-
Worldwide, chronic wounds are still a major and increasing problem area in medicine with protracted suffering of patients and enormous costs. Beside conventional wound treatment, for instance kinds of oxygen therapy and cold plasma technology have been tested, providing an improvement in the perfusion of wounds and their healing potential, but these methods are unfortunately not sufficiently validated and accepted for clinical practice to date. Using hyperspectral imaging technology in the visible (VIS) and near infrared (NIR) region with high spectral and spatial resolution, perfusion parameters of tissue and wounds can be determined. ⋯ From hyperspectral data the hemoglobin oxygenation (StO2), the relative concentration of hemoglobin [tissue hemoglobin index (THI)] and the so-called NIR-perfusion index can be determined. The first two parameters are calculated from the VIS-part of the spectrum and represent the perfusion of superficial tissue layers, whereas the NIR-perfusion index is calculated from the NIR-part representing the perfusion in deeper layers. First clinical measurements of transplanted flaps and chronic ulcer wounds show, that the perfusion level can be determined quantitatively allowing sensitive evaluation and monitoring for an optimization of the wound treatment planning and for validation of new treatment methods.
-
Blood perfusion is the supply of tissue with blood, and oxygen is a key factor in the field of minor and major wound healing. Reduced perfusion of a wound bed or transplant often causes various complications. ⋯ Therefore, methods, software and algorithms for a new HSI system are presented which can be used to observe tissue oxygenation and other parameters that are of importance in supervising healing processes. This could offer an improved insight into wound perfusion allowing timely intervention.