China
Africa
Explore chapters and articles related to this topic
Cooperative Data Fusion for Advanced Monitoring and Assessment in Healthcare Infrastructures
Published in Pietro Salvo, Miguel Hernandez-Silveira, Krzysztof Iniewski, Wireless Medical Systems and Algorithms, 2017
Vasileios Tsoutsouras, Sotirios Xydis, Dimitrios Soudris
Table 7.2 summarizes the sensors directly connected to the SWD. There are sensors related to the pump (speed, current, and pressure), which is the only actuator of the SWD and the most critical element for enforcing NPWT. The rest of the sensors convey information about the status of the wound and are in close contact to it. The table lists the cases for which an alarm or warning should be generated from the data fusion engine of the SWD. It also summarizes the indicative range of values that could raise an alarm in the system. Table 7.2 reports for each sensor with medical relevance allocated to the SWD data fusion engine (i) the inferred wound assessment (i.e., negative pressure, inflammation, and infection), (ii) the normal operation range (column 3) to be used on data fusion level 2, (iii) the minimum and maximum values that trigger an alarm or warning generation (column 4) to be used on data fusion level 3, and (iv) the corresponding receiver per alarm (column 5).
Skin image analysis granulation tissue for healing assessment of chronic ulcers
Published in Ahmad Fadzil Mohamad Hani, Dileep Kumar, Optical Imaging for Biomedical and Clinical Applications, 2017
Ahmad Fadzil Mohamad Hani, Leena Arshad
Colour has been utilised as the key element for analysis in colour image processing in the application of wound assessment for most of the work that has been developed so far. Previously developed methods for wound tissue classification and segmentation performs single colour channel analysis on digital images in traditional colour models such as HIS and RGB (red, green and blue) [12–15]. The ulcer surface typically comprises mixture of different tissues and hence the use of only one colour channel is not adequate to classify each type of tissue fully. Studies demonstrated that for a particular type of ulcer tissue, the clusters of pixels in RGB colour model forms a uniquely shaped three-dimensional (3D) cloud that distinguishes three types of tissues: necrotic, slough and granulation; and hence colour pixels are considered in all colour channels in the image to be able to classify different ulcer tissues [16–18]. Segmentation-based classification of wound tissues utilising colour and texture attribute was also proposed [19,20]. The method utilises unsupervised segmentation to segment colours from wound images into several regions. The method eventually extracts the colour and texture descriptors from the coloured images so that automatic classification and labelling can be done to those regions.
Microfluidic Platforms for Wound Healing Analysis
Published in Raju Khan, Chetna Dhand, S. K. Sanghi, Shabi Thankaraj Salammal, A. B. P. Mishra, Advanced Microfluidics-Based Point-of-Care Diagnostics, 2022
Lynda Velutheril Thomas, Priyadarsini Sreenivasan
Immunohistochemical markers like extracellular and intracellular proteins can also be utilized as cellular biological markers for wound assessment. This application of biosensing the intracellular and extracellular proteins was studied by Becker et al. (1986), Ouhayoun et al. (1990), and Carmichael et al. (1991). Some examples of intracellular markers in wounds are cytokeratins (CKs), vimentin, and vinculin, and extracellular markers are collagen IV, laminin, and fibronectin (Mai et al. 2009). CKs are a part of the cytoskeleton which are intermediate filament proteins found in the cytoplasm of epithelial cells. Moll et al. (1982)identified twenty different CK polypeptides in human tissues with molecular weights ranging from 40 to 68 kDa. There are basically two different types of CK, basically acid and neutral/basic CK, which can be distinguished using electrophoresis. The distribution patterns of CK are different but distinct in various epithelia and within stratified epithelia. Shabana et al. (1989) observed that characteristic distributions of CK are available in the stratum basale, stratum spinosum, stratum granulosum, and stratum corneum. Based on this, Boisnic et al. (1995) concluded that they can be utilized as specific markers for pathways of epithelial differentiation. Barui et al. (2011) investigated the impact of honey-based occlusive dressing on nonhealing (unresponsive to conventional antibiotics) traumatic lower limb wounds through clinicopathological and immunohistochemical (e.g., expression of p63, E-cadherin) and collagen I and III evaluations. Nevertheless, immunohistochemical markers are not yet implemented in microfluidics platforms and their capabilities are still being explored.
Augmented reality: a novel means of measurement in dermatology
Published in Journal of Medical Engineering & Technology, 2021
Austinn C. Miller, Travis W. Blalock
Using AR as a measurement tool in dermatology offers multiple potential advantages for patients, providers, researchers, and the healthcare system as a whole (Table 1). Principally, implementing AR as a measurement tool of cutaneous lesions has the potential to reduce inconsistencies in human measurement. A recent study reported that a majority of dermatologists use hand-held rulers or callipers to measure lesions [3]. However, many dermatologists also reported using visual estimation. Others admitted to forgoing measurements themselves, and instead delegated this task to staff members, who also may perform this task in a non-standardized manner. Mannam et al. further investigated skin lesion measurements with and without a ruler and showed significant differences in accuracy among methods [4]. Even with a ruler, the dermatology participants in the study were able to correctly measure lesions within 1 mm of actual size only 71% of the time. Furthermore, their visual estimations varied by more than 1 mm nearly half of the time. Seat et al. conducted a trial that compared inter-rater and intra-rater reliability of wound measurement using a smartphone application versus the traditional ruler [5]. This study found that a smartphone wound assessment application produced sufficiently consistent results to be clinically useful and appears to be superior to linear measurements with a ruler. While intra-rater results were similar for both the smartphone application and ruler, the reliability between raters was excellent with the application, but poor for the ruler method. The variability illustrated in these studies is not without consequence as these measurement errors may falsely validate or invalidate certain plans of action or treatment regimens. Therefore, introducing an AR application to dermatology measuring, similar to the application described by Seat et al. [11], has the potential to reduce human error, especially that caused by interpersonal variation, and thus improve measurements.
Varicose ulcer(C6) wound image tissue classification using multidimensional convolutional neural networks
Published in The Imaging Science Journal, 2019
V. Rajathi, R. R. Bhavani, G. Wiselin Jiji
Physically venous ulcers are irregular and shallow and are usually located over bony prominences and the granulation tissue occurs in the ulcer base [3]. During the existence of the wound, several tissue types can be identified. To properly diagnose the ulcer and give the best treatment option, the identification of tissue types is essential. The most commonly identified tissue types in any type of ulcers or other open wounds are epithelium, slough, necrotic and granulation [4]. The wound healing depends on the types of tissues like granulation, slough, necrotic, epithelial in the wound bed [5]. The healing stage of varicose ulcer is defined by various phases i.e. haemostasis, inflammation, granulation and maturation. The phases of healing are to identify the damaged tissue and it helps the doctors to assess the wounds periodically. Wound image analysis is used to monitor the wound healing and decide the treatment of varicose ulcer. The fibrinogen and proteins in the blood release the wastes into the tissue, which forms the skin irritation and allergetic reactions to the patients leading to venous eczema. In this stage the colour of the skin changes due to the venous hypertension and it increases the pressure in the cells of the patients leads to venous oedema [6]. Venous ulcers are identified by the physical factors, examination and representation of clinical factors for the correct analysis of treatment and management [7]. Wound image segmentation is the most important step for the clinicians to identify the damaged tissues. It is done by the method of detection of edges in the wound for extracting the region of interest area. In [8] canny edge detection is used for the removal of reflections in the image. Another method is colour-based segmentation by using filters in the HSV value of the image. There are two types of measurements of wound: (i) Contact (direct measurements from the wound) and (ii) Non-Contact (a camera is used to capture the image for tissue classification). Wound assessment is done by the attributes like wounded area and its perimeter [9].
Volumetric monitoring of cutaneous leishmaniasis ulcers: can camera be as accurate as laser scanner?
Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2019
Omar Zenteno, Eduardo González, Sylvie Treuillet, Braulio M. Valencia, Benjamin Castaneda, Alejandro Llanos-Cuentas, Yves Lucas
In this context, tele-dermatology and image processing technologies provide a real opportunity to develop efficient computer-aided-diagnosis tools based on low cost portable imaging devices such as digital cameras or smartphones. Historically, two approaches have been used for 3D ulcer modelling: passive vision or photogrammetry (Plassmann and Jones 1998; Malian et al. 2002, 2004) and active vision (ie structured light techniques (Jones and Plassmann 1995; Ozturk et al. 1996) or laser scanner (Krouskop et al. 2002; Callieri et al. 2003). These first prototypes were not portable and too expensive for routine care in clinics. Therefore, a second generation of ulcer 3D modelling devices was designed based on commercially available digital cameras and specific stereo optics (Plassmann et al. 2007), time-of-flight cameras (Putzer et al. 2014) and laser scanners (Zvietcovich et al. 2012; Pavlovčič et al. 2015). A commercial solution is presented in Nixon et al. (2012), Silhouette by ARANZ (Christchurch, NZ) is a very easy-to-use commercial device for capturing the image of a wound at a fixed distance. It produces a cavity volume estimate with about 5% precision by projecting 3 intersecting laser lines on the wound. All the aforementioned systems still require complex instrumentation. Moreover, considering the current prevalence of smartphones and tablets with high resolution cameras, the possibility to use them in conjunction with structure from motion methods is an attractive option for wound assessment. Some applications using mobile phones have already been presented in Hettiarachchi et al. (2013), Foltynski et al. (2014) and Wang et al. (2015). Unfortunately, these attempts were limited to bi-dimensional information (ie area and perimeter). Recently, Sirazitdinova and Deserno (2017) proposed a concept for a low-cost wound assessment system based on the work presented in Ondrúška et al. (2015) including additional functions like automatic segmentation of wound regions, quantitative characteristics (size, depth, volume, rate of healing) and qualitative characteristics from colour (corrected using a reference colour card), like necrosis detection and tissue characterisation. Unfortunately, this work only presents a concept without any details on segmentation method or volumetric estimates neither experimental results. In addition, Ondruska’s software might not be open-source and has only been tested on common objects of sizes ranging from 5 to 0.3 m. Since the absolute distances compared to depth reconstruction obtained using KinectFusion are about 2 cm, it is unclear whether this method is accurate enough for wound assessment. Treuillet et al. (2009) presented an ulcer 3D modelling technique using a simple hand-held digital camera and two single views of the wound. The 3D model reconstruction allowed simultaneous colour acquisition and accurate volume computation (precision 3.5%) and was used by Wannous et al. (2011) in a complementary study for multi-view tissue classification based on machine-learning. Zenteno et al. (2017) applied a similar technique, but this time using several images extracted from a video instead of only two photos.