Image Processing for Skin Ulcers in Tropical Areas (IMPULSO)

Tema de investigación: Imagenología hyperespectral, Leishmaniasis, Ultrasonido cuantitativo

Año de inicio: 2016

Investigadores: Benjamin Castañeda, Ph.D. (PUCP) - International Project Coordinator, Roberto Lavarello, Ph.D. (PUCP) - Co-investigator, Jorge Arévalo, Ph.D. (UPCH) - Co-investigator, Francisco Bravo, M.D. (UPCH) - Co-investigator, Alejandro Llanos-Cuentas, M.D. (UPCH) - Co-investigator, Vanessa Adaui, Ph.D. (UPCH) - Co-investigator, Franck Marzani, Ph.D. (UB) - Co-investigator, Yannick Marzani, Ph.D. (UB) - Co-investigator, Sylvie Treuillet, Ph.D. (UO) - Co-investigator, Yves Lucas, Ph.D. (UO) - Co-investigator, July Galeano, Ph.D. (ITM) - Co-investigator, María Moncada, Ph.D. (ITM) - Co-investigator, Carlos Vargas, M.Sc. (ITM) - Co-investigator, Sandra Milena Pérez, M.Sc. (UPB) - Ph.D. Student, Lina Hoyos, Ph.D. (UPB) - Co-investigator

Image Processing for Skin Ulcers in Tropical Areas (IMPULSO)

This project is oriented to basic research on image processing techniques applied to optical, hyperspectral and ultrasonic images. The goal is to extract features from these images which may have the potential to improve diagnosis and monitoring of treatment at primary health-care units in tropical areas.
A research network which integrates knowledge of dermatology, 3D imaging, hyperspectral imaging, ultrasonic imaging, tissue engineering and photodynamic therapy will be formed. The synergetic collaboration will aim to identify the best imaging features which can then be readily translated into point of care devices which can benefit the healthcare decision process in tropical areas with respect to skin ulcers.
In addition, this network will work within a conceptual framework which is not part of current proposal but it will guide towards computer-aided-diagnosis to a wide number of primary health settings via telemedicine tools. Therefore, the ultimate outcomes of this network is to have an impact on public health of under-served areas settings through improvement of diagnosis and treatment follow up of skin wounds.

Images of different types of ulcers and formats have been obtaneid: 3D ultrasound, ultrasound, multi-spectral,color images, 3D color reconstruction and laser (See detail).
IMPULSO project offers free access to these images, which must be cited if you want to use them. If you are interested in these image you can contact with us to castaneda.b@pucp.edu.pe.

Institutions

  • Pontificia Universidad Católica del Perú (PUCP)

  • Universidad Peruana Cayetano Heredia (UPCH)

  • Instituo Tecnológico Metropolitano (ITM)

  • Universidad Pontificia Bolivariana (UPB)

  • Université d’Orleans, PRISME Lab (UO)

  • Université de Bourgogne (UB), LE2I Lab (UB)

 

Goals, motivation, methodology and contribution of each participating institution

Even though skin ulcers (SU) are one of the most frequent causes of consultation in primary health-care units (PHU) in tropical areas, the lack of dermatological expertise and modern diagnostic tools can lead to their incorrect diagnosis and treatment. Furthermore, the regional prevalence of the causing agents influences the diagnosis. In Latin America, physicians tend to over-diagnose cutaneous leishmaniasis since it is highly endemic, while in fact, the SU may be caused by (i) other infectious agents like mycobacteria, bacteria or fungi or (ii) non-infectious conditions like venous diseases or carcinomas. The sub-optimal management of SU produces increased morbidity and has a negative effect on the estimation of disease burden and control.
Current technological advancements in imaging equipment and telecommunication infrastructure create new possibilities to improve these shortcomings at PHUs in tropical and under-served areas promoting the development of point-of-care healthcare tools. Miniaturization and hardware cost-reduction are enabling the development of cost-effective equipment with enough computational power to run embedded computer aided diagnosis tools. On the other hand, network connectivity is becoming increasingly available in tropical settings. In the case of the Peruvian and Colombian governments, they both are investing in the deployment of telecommunications infrastructure in tropical areas. This will enable the application of tele-dermatological tools (i.e. an operator can acquire the information on-site whereas the diagnosis is performed remotely by a specialized physician).

In particular, we propose to work towards the development of novel computer-aided-diagnosis and telemedicine tools based on image processing technologies which can be easily transferred and applied to under-served areas settings to improve the diagnosis and follow up of treatment of skin wounds. Specifically, this proposal will focus on developing new imaging features which can provide useful information to 1) identify the main types of SU from tropical areas and 2) assess the effectiveness of treatment for different SU types. Moreover, these features will also help to evaluate the effectiveness of therapeutical methodologies for SU. These imaging features will be based on a) 3D imaging (using digital cameras or laser scanners) to acquire surface anatomical information as well as color cues, b) hyper-spectral technologies to provide functional features and c) ultrasound imaging to deliver characteristics from within the tissue. This proposal extends from the current work of the participating research groups and seeks for the identification of the best inter-modality features for the health care of SU in tropical areas. This goal could only be accomplished by the synergistic collaboration of the participants. Specifically, our project has the following six components.

1) Dermatological knowledge (performed by UPCH)

The optimal current SU management requires the concourse of physicians with dermatological knowledge to provide clinical diagnosis, based on lesion observation, histopathology or infectious disease laboratory confirmation, and to assess therapy evolution. Furthermore, the latter fact requires patients return for periodic follow up. This fact implies a considerable drop-out in rural areas where SU are prevalent because of large distances and workable days loss. Since there is a scarce number of dermatologists and primarily concentrated in major urban areas, a large proportion of SUs are diagnosed and treated by non-specialist physicians. They must discriminate between different types of SUs and prescribe a therapy accordingly.
The most prevalent SU on lower limbs are caused by venous stasis extremities but it should be differentiated from SUs caused by trauma, inflammatory vascular problems, neoplasia or tropical infectious diseases (cutaneous leishmaniasis, sporotrichosis, skin tuberculosis or other pathogens with very low prevalence like Buruli ulcer). The last two causes are particularly frequent among patients who attend hospitals or national reference centers. On the other hand, SUs present in other parts of the body are seldom caused by venous stasis and, therefore other aethiologies must be considered for a differential diagnosis.
SU therapies require a periodic follow up to monitor the healing process. This is particularly difficult in situations where either there is high health personnel mobility or when the patients live far away from a highly specialized health center. In these cases the compliance of treatment follow up is severely affected.
Features extracted from three-dimensional, hyperspectral and ultrasound images, stored and processed with specialized software will certainly improve therapy management. Images obtained at the PHU with a simple and affordable device will be transmitted via a smartphone to a diagnosis center. The specific patterns thus obtained will be matched with a particular disease and this result will be transmitted back to PHU. In this way data transfer from non-specialists to dermatological physicians will provide rapid answers to the end user. Moreover, this approach will open the possibility to onset a country data base of SU to establish epidemiology surveys to assess the impact of public control measures, i.e. diagnosis and corresponding therapy. Furthermore, the image technology with telemedicine data transfer will permit to establish follow up research studies to assess the impact of new therapies.
The first step will be to define the appropriate technologies for each type of SU, to validate its Positive Predictive Value and Negative Predictive Value after comparison with the gold standard diagnostic procedures. Moreover, it is necessary to become acquainted with the challenges they will face, first in well-equipped settings and, finally, at the PHU. All these experiences will be translated into standardized written protocols using a plain language to be translated to the human resources working at PHU. This approach will be applied to both diagnosis and therapy monitoring.
This component of the project will be in charge of recruiting patients with different types of skin ulcer. Image records and samples will be obtained after informed consent according the ethical IRB. Laboratory diagnosis, histopathology and molecular biomarkers will support differential diagnosis and it will be contrasted with image analysis diagnosis. The telemedicine developments will result from exchanging experiences between the clinical expertise, both in urban and rural settings, with those specialists on image processing and software development.

2) 3D color imaging (performed by UO)

Even though imaging tools for automated wound diagnosis and monitoring have been investigated for several decades, the most common clinical scenario remains still based on the physician’s periodical manual or visual assessment to monitor shape (area and depth) and appearance (color and type of tissues) of the wound. These features are important for healing appreciation. Indeed, traditional wounds size measurements are performed with a ruler or a Kundin gauge, or by tracing the wound outline on a transparent sheet. Therefore, wound monitoring can be extremely time-consuming for medical staff and lacks reliability as it is based on subjective manual measurements.
Two-dimensional imaging has initially been proposed for wound assessment [1]. It has the advantage of being low cost, non-contact (i.e. less infection risk), and image processing algorithms can be implemented to improve the accuracy and repeatability of the measurements. However, it does not take into account both, the curvature of the body and the irregular shape of the wound and, therefore, it is generally biased depending on the view angle of the acquisition.
Three-dimensional imaging technologies are essential to have volumetric description and precise depth estimation of wounds with complex shape. Active laser technologies [2, 3] and passive photogrammetric methods using only digital cameras [4, 5] have been proposed for wound assessment. The latter are more adapted to 3D modeling of wounds in low-resources settings as well as for tele-dermatology and patient self-monitoring in remote tropical areas. Indeed, photogrammetric methods allow to calculate a 3D model by processing images taken with handled digital cameras so that are included in low-cost systems such as tablet or smartphones by matching similar structures between several views. The 3D reconstruction may be fully automated from images as well as the calculation of most relevant parameters for monitoring wound healing progress: color images enable tissue discrimination (granulation, fibrin, necrosis and epithelium) [6, 7] and the determination of the outline of the wound, then area, depth and volume may be extracted precisely from the 3D numerical model [2].
We propose to evaluate the time evolution of shape (depth, area and volume) and appearance (color and texture) as features for SU type differentiation and therapy monitoring. For this purpose a 3D model of the SU is created by photogrammetric approaches using a standard digital camera [4, 6, 7] in combination with machine learning techniques to provide a robust tissue classification for wound assessment. To be reproducible, multiview tissue classification needs a color correction across camera and illumination changes [8]. Quantitative color information as well as three-dimensional physical measurements of size and volume can be computed from these models with less variability than using manual methods even in the case of highly curved body parts. Although these techniques have been actively used for a variety of applications, there is still a lack of understanding on the clinical value of these new methods with quantitative information besides providing an improved record and qualitative evaluation of the wound. These imaging tools will allow repeatable and accurate non-contact measurements for trackingwound healing over time, even with different users. Existing collaboration between UO and PUCP has focused on applying and improving these imaging techniques on Leishmaniasis cutaneous wounds. A new version has been developed to obtain a 3D modeling of wounds from videos [5]. We will extend this approach to evaluate different types of SU.

3) Hyper-Spectral Imaging (performed by UB)

Multi/hyper-spectral imaging combines the spatial resolution of dermoscopy and the spectral one of spectroscopy. Such systems consist in the acquisition of images of an object of interest, illuminated under different wavelengths. They are based on the principle of light diffuse reflectance so that each pixel of those images corresponds to the reflectance spectrum of the object’s corresponding area. Therefore, a hyper-spectral image is a cube with two spatial dimensions and one spectral one. It can be seen in two different ways: a set of pixels where each one is described by a spectrum or a set of mono-channel bands, one band for each wavelength similar to a grey level image. This modality allows covering both visible and non-visible range (such as UV or NIR).
Hyper-spectral imaging systems have been widely used in the analysis of different types of biological tissues such as skin, stomach and cervix [9]. Indeed, light reflectance can be described as the result of attenuation of the incident light due to scattering and absorption phenomena into the tissue under evaluation [10]. In the case of biological tissue, these phenomena are caused by different substances and particles that reside within the tissue. Thus, by implementing computational techniques of spectral analysis it is possible to determine concentration maps of the different variables that are responsible of light attenuation. Some of the techniques that can be useful in the spectral analysis of hyper-spectral images are blind-source separation methods and model inversion methods for tissue optical parameters fitting. We proposed in 2011 the use of a multi-spectral system for the analysis of skin diseases such as melasma and vitiligo [11, 12] implemented a real-time method for detecting human papillomavirus; Orfanoudaki and its working group [13] used multispectral images for the study of cervical biopsies. We have also applied hyper-spectral imaging in the detection of ulcers and cancers in the digestive system [14].
We propose to extract hyper-spectral features from skin ulcers images by adapting three approaches that have been already developed at UB. The first one is based on an extension of classification methodologies used for color images by combining spatial and spectral features [15]. The second approach is based on source separation. In that case, the spectrum of each pixel is considered as a weighted combination of some specific components. The goal of these methods is to quantify these components for each pixel and so provide a map of these endmembers [14, 16-18]. The last method is based on the light-tissue models used in spectroscopy. They can be applied successively on all the pixels of the image. The result is a map of the quantity of each of the components described inside the model [19, 20].

4) Quantitative ultrasound imaging (performed by PUCP)

Ultrasound (US) imaging is a medical modality which is capable of visualizing and characterizing the internal structure of the tissue based on its interaction with mechanical waves. In particular, US is well-suited for point-of-care applications due to its relative low cost, portability, safety, and real-time capabilities. B-mode ultrasound is one of the most widely medical imaging modalities worldwide; however due to its qualitative nature, diagnosis based on this type of images is subjective and operator-dependent. In contrast, quantitative ultrasound (QUS) imaging seeks to provide quantitative estimations of parameters which can be extracted from the US radiofrequency (RF) signals.
Several types of QUS imaging have been proposed in the literature. Texture analysis from B-mode images examines the statistical properties from the envelope of the RF data including mean, standard deviation, entropy [21], parametric representations of the envelope data [22] and wavelet coefficients [23]. Good results have been found for liver imaging [24], breast cancer diagnosis [25], nodule characterization in the thyroid [26], and tumor response to chemotherapy [27] or thermal therapy. Other type of QUS imaging seeks to infer mechanical properties of the tissue such as attenuation (attenuation coefficient slopes) and elasticity (Young’s modulus) or estimate its micro-structure (effective scattering diameter, effective acoustic concentration). They have been applied to prostate and breast cancer detection [28, 29], liver fibrosis staging [30], and to image the kidney [31] and eyes [32]. There are also reports of their usefulness to monitor photodynamic therapy [33]. In particular, researchers at PUCP have developed QUS imaging techniques to estimate tissue elasticity applied to prostate cancer detection and breast cancer diagnosis [34, 35]. In addition, they have also implemented techniques to estimate tissue attenuation for diagnosis of nodules in the thyroid and breast [36, 37].
Application of QUS imaging for the evaluation of SUs has been scarce. The main reason for this is that a high frequency ultrasound scanner is required to provide an adequate resolution. Typical scanners in clinical settings have a frequency range from 2 to 15 MHz whereas skin imaging may require frequencies up to 50MHz.
The research group at PUCP has acquired a research ultrasound scanner with a frequency range from 20 to 50 MHz which provides access to RF data. We propose to extract QUS imaging features from high frequency ultrasound images. In particular, we will measure size, shape and texture from B-mode images of different SUs, and utilize RF data to estimate their elasticity and attenuation.

5) Tissue Engineering (performed by ITM)

Tissue engineering seeks to create, repair or replace tissues and organs by using combinations of cells, biomaterials and/or biologically active molecules. To this end, scaffolds are manufactured with the aim to support cell growth and proliferation [38]. The characteristics required for the scaffolds change according to the type of cell/tissue and the material used for its construction [39]. One of the techniques used in scaffold manufacturing is electrospinnig [40]. By careful selection of the material to be processed in this technique, matrices can be obtained with a unique combination of high specific surface area, flexibility, toughness and tensile strength. This properties are key to numerous applications in the biomedical field such as matrices for regeneration and/or repair of different tissues and organs such as vascular grafts, nerves, cartilage, bones, heart muscle, corneas and, in particular, skin [41].
In this project, features extracted from hyper-spectral imaging will be used to characterize and evaluate the potential of these scaffolds for SU healing. By using hyper-spectral imaging, it will be possible to quantify the optical properties of scaffolds and correlate those properties with the results from mechanical and the morphological tests to determine the best choice of material required for SU healing. ITM has the equipment for both, scaffold production and mechanical/morphological characterization: electrospinning, universal testing machine and SEM microscope.

6) Photodynamic therapy (performed by UPB)

Photodynamic therapy (PDT) uses a photosensitizing agent or drug that is only activated when exposed to a specific wavelength. This technique has the advantage to deliver the cargo at a desired site at a required time. Nanoparticles can be used as a photosensitizing agent. Several types of nanocarriers have been synthesized for drug delivery such as dendrimers, liposomes, solid lipid nanoparticles, polymersomes, polymer-drug conjugates, polymeric nanoparticles, peptide nanoparticles, micelles, nanoemulsions, nanospheres, nanoshells, carbon nanotubes, and gold nanoparticles. Light is an attractive option to be used as an external stimulus, due to its noninvasiveness and excellent remote spatio-temporal control over the release of cargo from nanocarriers.
The use of nanoparticles in PDT makes this technique also suitable not only for drug delivery but also for tissue imaging. Thanks to nanoparticles properties, they can be also used as biomarkers offering a theranostic agent for imaging and therapy [42]. To this end, multispectral systems are used for the imaging of the intensity and locations of the drugs delivered by PDT.
PDT may be an effective alternative of treatment for SUs. In particular, for cutaneous Leishmaniasis (CL), Enk et al. [43]demonstrated that PDT offers relatively fast localized healing without damaging healthy tissues and showing no sign of systemic toxicity [44]. Additional studies have found that PDT flattens, reduces the size, and selectively eliminates inflammatory infiltrate of CL lesions, leading to rapid healing without scarring [45, 46]. Other studies have applied PDT to CL by directly using a photosensitizer [47, 48], or using PEGylated silver doped zinc oxide nanoparticles as photosensitizers [49].
Based on UPB’s experience in nanotechnology and nanomaterials applied to cardiovascular dynamics, we propose to study the use of Hyperspectral imaging in combination with adequately designed nanoparticles to be theranostic agents for imaging and therapy of skin ulcers. In other words, we will study the possibility of implementing active targeting strategies for the intelligent delivery of drug and molecules in different types of skin ulcers. The initial focus of this component will be given to CL since there are already reports in the literature which suggest the advantages of PDT therapy.
To summarize, we are proposing a highly inter-disciplinary project focus on the identification of image processing features, from hyper-spectral, ultrasound, and 3D imaging modalities, which are clinically useful for diagnosis and monitoring the treatment of SUs. The aim of this project can only be achieved by the collaboration of different research groups. In particular, two Peruvian groups are providing expertise in dermatology and ultrasonic imaging; two French groups are adding their knowledge in 3D and hyperspectral imaging; and two Colombian groups are proposing new therapies for SUs.

Bibliography

[1] K. Diaz, B. Castañeda, C. Miranda, R. Lavarello, and A. Llanos, “Development of an acquisition protocol and a segmentation algortihm for wounds of cutaneous Leishmaniasis in digital images,” in SPIE Medical Imaging, 2010, pp. 76234L-76234L-11.
[2] F. Zvietcovich, B. Castaneda, B. Valencia, and A. Llanos-Cuentas, “A 3D assessment tool for accurate volume measurement for monitoring the evolution of cutaneous leishmaniasis wounds,” Conf Proc IEEE Eng Med Biol Soc, vol. 2012, pp. 2025-8, 2012.
[3] L. Casas, B. Castaneda, and S. Treuillet, “Imaging technologies applied to chronic wounds: a survey,” in Proceedings of the 4th International Symposium on Applied Sciences in Biomedical and Communication Technologies, 2011, p. 167.
[4] S. Treuillet, B. Albouy, and Y. Lucas, “Three-dimensional assessment of skin wounds using a standard digital camera,” Medical Imaging, IEEE Transactions on, vol. 28, pp. 752-762, 2009.
[5] E. L. Casas, C. Benjamin, and S. Treuillet, “Low-cost uncalibrated video-based tool for tridimensional reconstruction oriented to assessment of chronic wounds,” in 10th International Symposium on Medical Information Processing and Analysis, 2014, p. 10.1117/12.2070999.
[6] H. Wannous, S. Treuillet, and Y. Lucas, “Robust tissue classification for reproducible wound assessment in telemedicine environments,” Journal of Electronic Imaging, vol. 19, pp. 023002-023002-9, 2010.
[7] H. Wannous, Y. Lucas, and S. Treuillet, “Enhanced assessment of the wound-healing process by accurate multiview tissue classification,” Medical Imaging, IEEE Transactions on, vol. 30, pp. 315-326, 2011.
[8] H. Wannous, Y. Lucas, S. Treuillet, A. Mansouri, and Y. Voisin, “Improving color correction across camera and illumination changes by contextual sample selection,” Journal of Electronic Imaging, vol. 21, pp. 023015-1-023015-14, 2012.
[9] G. Lu and B. Fei, “Medical hyperspectral imaging: a review,” Journal of Biomedical Optics, vol. 19, pp. 010901-010901, 2014.
[10] V. Tuchin, Tissue Optics: Light Scattering Methods and Instruments for Medical Diagnosis, First ed.: SPIE, 2007.
[11] R. Jolivot, P. Vabres, and F. Marzani, “Reconstruction of hyperspectral cutaneous data from an artificial neural network-based multispectral imaging system,” Computerized medical imaging and graphics, vol. 35, pp. 85-88, 2011.
[12] S. Nakappan, S. Y. Park, D. Serachitopol, R. Price, M. Cardeno, S. Au, et al., “Methodology of a real-time quality control for the multispectral digital colposcope (MDC),” Gynecol Oncol, vol. 107, pp. S215-22, Oct 2007.
[13] I. M. Orfanoudaki, G. C. Themelis, S. K. Sifakis, D. H. Fragouli, J. G. Panayiotides, E. M. Vazgiouraki, et al., “A clinical study of optical biopsy of the uterine cervix using a multispectral imaging system,” Gynecol Oncol, vol. 96, pp. 119-31, Jan 2005.
[14] J. Galeano, R. Jolivot, Y. Benezeth, F. Marzani, J. F. Emile, and D. Lamarque, “Analysis of Multispectral Images of Excised Colon Tissue Samples Based on Genetic Algorithms,” in Signal Image Technology and Internet Based Systems (SITIS), 2012 Eighth International Conference on, 2012, pp. 833-838.
[15] M. Rastgoo, R. Garcia, O. Morel, and F. Marzani, “Automatic differentiation of melanoma from dysplastic nevi,” Computerized Medical Imaging and Graphics, vol. 43, pp. 44-52, 2015.
11
[16] J. Mitra, R. Jolivot, F. Marzani, and P. Vabres, “Blind source separation of skin chromophores on a hyperspectral cube,” Skin Research and Technology, vol. 16, p.??, 2010.
[17] J. Galeano, R. Jolivot, F. Marzani, and Y. Benezeth, “Unmixing of human skin optical reflectance maps by Non-negative Matrix Factorization algorithm,” Biomedical Signal Processing and Control, vol. 8, pp. 169-175, 2013.
[18] J. Galeano, R. Jolivot, and F. Marzani, “Quantification of melanin and hemoglobin in humain skin from multispectral image acquisition: use of a neuronal network combined to a non-negative matrix factorization,” in Applied and Computational Mathematics, special issue on Applied Artificial Intelligence and Soft Computing. vol. 11, ed, 2012, pp. 257-270.
[19] R. Jolivot, Y. Benezeth, and F. Marzani, “Research Article Skin Parameter Map Retrieval from a Dedicated Multispectral Imaging System Applied to Dermatology/Cosmetology,” 2013.
[20] C. Li, V. Brost, Y. Benezeth, F. Marzani, and F. Yang, “Design and evaluation of a parallel and optimized light–tissue interaction-based method for fast skin lesion assessment,” Journal of Real-Time Image Processing, pp. 1-14, 2015.
[21] V. G. Murmis, J. J. Gisvold, T. M. Kinter, and J. F. Greenleaf, “Texture analysis of ultrasound B-scans to aid diagnosis of cancerous lesions in the breast,” in Ultrasonics Symposium, 1988. Proceedings., IEEE 1988, 1988, pp. 839-842 vol.2.
[22] V. Dutt and J. F. Greenleaf, “Ultrasound Echo Envelope Analysis Using a Homodyned K Distribution Signal Model,” Ultrasonic Imaging, vol. 16, pp. 265-287, 10// 1994.
[23] H. Yoshida, D. D. Casalino, B. Keserci, A. Coskun, O. Ozturk, and A. Savranlar, “Wavelet-packet-based texture analysis for differentiation between benign and malignant liver tumours in ultrasound images,” Phys Med Biol, vol. 48, pp. 3735-53, Nov 2003.
[24] D. Nicholas, D. K. Nassiri, P. Garbutt, and C. R. Hill, “Tissue characterization from ultrasound B-scan data,” Ultrasound Med Biol, vol. 12, pp. 135-43, Feb 1986.
[25] P. M. Shankar, V. A. Dumane, T. George, C. W. Piccoli, J. M. Reid, F. Forsberg, et al., “Classification of breast masses in ultrasonic B scans using Nakagami and K distributions,” Phys Med Biol, vol. 48, pp. 2229-40, Jul 2003.
[26] S. Tsantis, D. Cavouras, I. Kalatzis, N. Piliouras, N. Dimitropoulos, and G. Nikiforidis, “Development of a support vector machine-based image analysis system for assessing the thyroid nodule malignancy risk on ultrasound,” Ultrasound Med Biol, vol. 31, pp. 1451-9, Nov 2005.
[27] S. Huber, M. Medl, M. Vesely, H. Czembirek, I. Zuna, and S. Delorme, “Ultrasonographic tissue characterization in monitoring tumor response to neoadjuvant chemotherapy in locally advanced breast cancer (work in progress),” J Ultrasound Med, vol. 19, pp. 677-86, Oct 2000.
[28] K. C. Balaji, W. R. Fair, E. J. Feleppa, C. R. Porter, H. Tsai, T. Liu, et al., “Role of advanced 2 and 3-dimensional ultrasound for detecting prostate cancer,” J Urol, vol. 168, pp. 2422-5, Dec 2002.
[29] R. M. Golub, R. E. Parsons, B. Sigel, E. J. Feleppa, J. Justin, H. A. Zaren, et al., “Differentiation of breast tumors by ultrasonic tissue characterization,” J Ultrasound Med, vol. 12, pp. 601-8, Oct 1993.
[30] T. Poynard, M. Munteanu, E. Luckina, H. Perazzo, Y. Ngo, L. Royer, et al., “Liver fibrosis evaluation using real-time shear wave elastography: applicability and diagnostic performance using methods without a gold standard,” J Hepatol, vol. 58, pp. 928-35, May 2013.
[31] M. F. Insana, J. G. Wood, T. J. Hall, G. G. Cox, and L. A. Harrison, “Effects of endothelin-1 on renal microvasculature measured using quantitative ultrasound,” Ultrasound Med Biol, vol. 21, pp. 1143-51, 1995.
[32] R. H. Silverman, F. L. Lizzi, B. G. Ursea, M. J. Rondeau, N. B. Eldeen, A. Kaliscz, et al., “High-resolution ultrasonic imaging and characterization of the ciliary body,” Invest Ophthalmol Vis Sci, vol. 42, pp. 885-94, Apr 2001.
12
[33] B. Banihashemi, R. Vlad, B. Debeljevic, A. Giles, M. C. Kolios, and G. J. Czarnota, “Ultrasound imaging of apoptosis in tumor response: novel preclinical monitoring of photodynamic therapy effects,” Cancer Res, vol. 68, pp. 8590-6, Oct 2008.
[34] D. T. Ginat, S. V. Destounis, R. G. Barr, B. Castaneda, J. G. Strang, and D. J. Rubens, “US elastography of breast and prostate lesions,” Radiographics, vol. 29, pp. 2007-16, Nov 2009.
[35] R. Rojas, J. Ormachea, A. Salo, P. Rodríguez, K. J. Parker, and B. Castaneda, “Crawling Waves Speed Estimation Based on the Dominant Component Analysis Paradigm,” Ultrason Imaging, Jan 2015.
[36] M. L. Montero, O. Zenteno, B. Castaneda, M. Oelze, and R. Lavarello, “Evaluation of classification strategies using quantitative ultrasound markers and a thyroid cancer rodent model,” in Ultrasonics Symposium (IUS), 2014 IEEE International, 2014, pp. 1916-1919.
[37] O. Zenteno, A. Luchies, M. Oelze, and R. Lavarello, “Improving the quality of attenuation imaging using full angular spatial compounding,” in Ultrasonics Symposium (IUS), 2014 IEEE International, 2014, pp. 2426-2429.
[38] S. Kidoaki, I. K. Kwon, and T. Matsuda, “Mesoscopic spatial designs of nano- and microfiber meshes for tissue-engineering matrix and scaffold based on newly devised multilayering and mixing electrospinning techniques,” Biomaterials, vol. 26, pp. 37-46, Jan 2005.
[39] W. J. Li, C. T. Laurencin, E. J. Caterson, R. S. Tuan, and F. K. Ko, “Electrospun nanofibrous structure: a novel scaffold for tissue engineering,” J Biomed Mater Res, vol. 60, pp. 613-21, Jun 2002.
[40] S. Agarwal, J. H. Wendorff, and A. Greiner, “Use of electrospinning technique for biomedical applications,” Polymer, vol. 49, pp. 5603-5621, 12/8/ 2008.
[41] N. Bhardwaj and S. C. Kundu, “Electrospinning: a fascinating fiber fabrication technique,” Biotechnol Adv, vol. 28, pp. 325-47, 2010 May-Jun 2010.
[42] L. V. Halig, D. Wang, A. Y. Wang, Z. G. Chen, and B. Fei, “Biodistribution Study of Nanoparticle Encapsulated Photodynamic Therapy Drugs Using Multispectral Imaging,” Proc SPIE Int Soc Opt Eng, vol. 8672, Mar 2013.
[43] C. D. Enk, C. Fritsch, F. Jonas, A. Nasereddin, A. Ingber, C. L. Jaffe, et al., “Treatment of cutaneous leishmaniasis with photodynamic therapy,” Arch Dermatol, vol. 139, pp. 432-4, Apr 2003.
[44] E. Latorre-Esteves, O. E. Akilov, P. Rai, S. M. Beverley, and T. Hasan, “Monitoring the efficacy of antimicrobial photodynamic therapy in a murine model of cutaneous leishmaniasis using L. major expressing GFP,” J Biophotonics, vol. 3, pp. 328-35, Jun 2010.
[45] A. Asilian and M. Davami, “Comparison between the efficacy of photodynamic therapy and topical paromomycin in the treatment of Old World cutaneous leishmaniasis: a placebo-controlled, randomized clinical trial,” Clin Exp Dermatol, vol. 31, pp. 634-7, Sep 2006.
[46] O. E. Akilov, W. Yousaf, S. X. Lukjan, S. Verma, and T. Hasan, “Optimization of topical photodynamic therapy with 3,7-bis(di-n-butylamino)phenothiazin-5-ium bromide for cutaneous leishmaniasis,” Lasers Surg Med, vol. 41, pp. 358-65, Jul 2009.
[47] K. Gardlo, Z. Horska, C. D. Enk, L. Rauch, M. Megahed, T. Ruzicka, et al., “Treatment of cutaneous leishmaniasis by photodynamic therapy,” J Am Acad Dermatol, vol. 48, pp. 893-6, Jun 2003.
[48] L. S. Peloi, C. E. Biondo, E. Kimura, M. J. Politi, M. V. Lonardoni, S. M. Aristides, et al., “Photodynamic therapy for American cutaneous leishmaniasis: the efficacy of methylene blue in hamsters experimentally infected with Leishmania (Leishmania) amazonensis,” Exp Parasitol, vol. 128, pp. 353-6, Aug 2011.
[49] A. Nadhman, S. Nazir, M. I. Khan, S. Arooj, M. Bakhtiar, G. Shahnaz, et al., “PEGylated silver doped zinc oxide nanoparticles as novel photosensitizers for photodynamic therapy against Leishmania,” Free Radic Biol Med, vol. 77, pp. 230-8, Dec 2014.

Compartir vía:
Enviar a: