Orientador: Regina Célia Popim / Resumo: Introdução: As feridas complexas apresentam difícil resolução e associam-se a perda cutânea extensa, infecções importantes, comprometimento da viabilidade dos tecidos e/ou associação com doenças sistêmicas que prejudicam os processos normais de cicatrização, cursam com elevada morbimortalidade e têm sido apontadas como grave problema de saúde pública. Na prática clínica, é importante avaliar as feridas e documentar a avaliação. O registro incompleto sobre o paciente e o tratamento em uso é apontado como um desafio no acompanhamento das feridas e também prejudica ações de gestão, pesquisa e educação. A incorporação de fotografias de feridas à pratica profissional, mostra-se como uma estratégia para auxiliar profissionais na observação, evolução e registro claro e preciso. O Optimum-Path Forest (OPF) é um framework para o desenvolvimento de técnicas de reconhecimento de padrões baseado em partições de caminhos ótimos e particularmente eficiente para a classificação de imagens. O classificador OPF gera resultados a partir do cruzamento das classes e características selecionadas. Objetivo: Descrever as etapas do desenvolvimento de um aplicativo para dispositivos móveis capaz de segmentar e classificar tecidos de feridas complexas baseado no Optimum-Path Forest (OPF) supervisionado. Método: Foi aplicada uma nova metodologia inteligente para análise e classificação de imagens de feridas complexas por meio de técnicas de processamento digital de imagens e aprendizado de máquina com ... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Introduction: Complex wounds are difficult to resolve and are associated with extensive cutaneous loss, major infections, compromised tissue viability and / or are related to systemic diseases that impair normal healing processes, have high morbidity and mortality and have been identified as severe public health problem. In clinical practice, it is important to evaluate the wounds and document the evaluation. The incomplete record on the patient and the treatment in use is pointed out as a challenge in the follow up of the wounds and also impairs management, research and education actions. The incorporation of wounds’ photos in the professional practice, stands out as a strategy to assist professionals in the observation, evolution and clear and precise recording. Optimum-Path Forest (OPF) is a framework for the development of pattern recognition techniques based on optimal path partitions and is particularly efficient for image classification. The OPF classifier generates results from the intersection of the selected classes and characteristics. Objective: Describe the steps in developing a mobile application capable of segmenting and sorting complex wound tissue based on the supervised Optimum-Path Forest (OPF). Method: A new intelligent methodology was applied for the analysis and classification of complex wound images using digital image processing and machine learning techniques with the supervised Optimum-Path Forest (OPF) standards classifier. The image bank of 27 comp... (Complete abstract click electronic access below) / Mestre

Submitted by Talita De Azevedo Coelho Furquim Pereira (talitapereira@bauru.sp.gov.br) on 2018-04-24T00:47:22Z No. of bitstreams: 1 Dissertação - final repositório.pdf: 1807192 bytes, checksum: 3197e207676b603a47d2146acf83b045 (MD5) / Approved for entry into archive by ROSANGELA APARECIDA LOBO null (rosangelalobo@btu.unesp.br) on 2018-04-26T16:58:24Z (GMT) No. of bitstreams: 1 pereira_tacf_me_bot.pdf: 1807192 bytes, checksum: 3197e207676b603a47d2146acf83b045 (MD5) / Made available in DSpace on 2018-04-26T16:58:24Z (GMT). No. of bitstreams: 1 pereira_tacf_me_bot.pdf: 1807192 bytes, checksum: 3197e207676b603a47d2146acf83b045 (MD5) Previous issue date: 2018-02-23 / Introdução: As feridas complexas apresentam difícil resolução e associam-se a perda cutânea extensa, infecções importantes, comprometimento da viabilidade dos tecidos e/ou associação com doenças sistêmicas que prejudicam os processos normais de cicatrização, cursam com elevada morbimortalidade e têm sido apontadas como grave problema de saúde pública. Na prática clínica, é importante avaliar as feridas e documentar a avaliação. O registro incompleto sobre o paciente e o tratamento em uso é apontado como um desafio no acompanhamento das feridas e também prejudica ações de gestão, pesquisa e educação. A incorporação de fotografias de feridas à pratica profissional, mostra-se como uma estratégia para auxiliar profissionais na observação, evolução e registro claro e preciso. O Optimum-Path Forest (OPF) é um framework para o desenvolvimento de técnicas de reconhecimento de padrões baseado em partições de caminhos ótimos e particularmente eficiente para a classificação de imagens. O classificador OPF gera resultados a partir do cruzamento das classes e características selecionadas. Objetivo: Descrever as etapas do desenvolvimento de um aplicativo para dispositivos móveis capaz de segmentar e classificar tecidos de feridas complexas baseado no Optimum-Path Forest (OPF) supervisionado. Método: Foi aplicada uma nova metodologia inteligente para análise e classificação de imagens de feridas complexas por meio de técnicas de processamento digital de imagens e aprendizado de máquina com o classificador de padrões Optimum-Path Forest (OPF) supervisionado. Criou-se o banco de imagens de 27 feridas complexas, que foram rotuladas por quatro especialistas conforme a classificação dos tecidos em quatro classes: granulação (vermelho), tecido fibrinóide (amarelo), necrose (preto) e hematoma (roxo), gerando 108 imagens rotuladas. Acrescentou-se duas classes: branco (o que está na foto, exceto o leito da ferida) e dúvida (divergência na classificação pelos profissionais). O classificador OPF foi treinado a partir dessas 108 imagens. Aplicou-se o OPF às imagens de feridas e verificou-se a acurácia. Em seguida, iniciou-se a construção do aplicativo. Resultados e Discussão: O presente estudo desenvolveu um esquema de classificação de tecido de feridas assistido por computador para avaliação e gerenciamento de feridas complexas, a partir de fotos de feridas da câmera digital de um smartphone. A aplicação do OPF a feridas complexas trouxe como resultado uma acurácia de 77,52% ± 6,14. Com esta ferramenta, foi desenvolvido como produto desta pesquisa um aplicativo para segmentação, classificação de tecidos e mensuração de feridas complexas. O aplicativo gera um relatório no formato Portable Document Format (PDF) que pode ser enviado por e-mail, impresso ou anexado a prontuário eletrônico compatível. Conclusão: Foi construído um banco com 27 imagens de feridas complexas, que quatro profissionais rotularam para treinamento do classificador OPF, aplicou-se o OPF às imagens de feridas complexas, avaliou-se a acurácia deste processo e desenvolveu-se um aplicativo para dispositivos móveis com as funções de segmentação da ferida, classificação de tecidos e mensuração da ferida. Os resultados mostraram que o valor da acurácia obtido na análise computacional teve valor significativo, equiparando-se a avaliação de especialistas em feridas. Comparando com estudos similares, a análise computacional de feridas mostrou-se com menor variabilidade em relação a avaliação de profissionais, sugerindo que a incorporação desta tecnologia na prática clínica favoreça o cuidado em saúde do paciente com feridas complexas, além de fornecer dados para a gestão, ensino e pesquisa. / Introduction: Complex wounds are difficult to resolve and are associated with extensive cutaneous loss, major infections, compromised tissue viability and / or are related to systemic diseases that impair normal healing processes, have high morbidity and mortality and have been identified as severe public health problem. In clinical practice, it is important to evaluate the wounds and document the evaluation. The incomplete record on the patient and the treatment in use is pointed out as a challenge in the follow up of the wounds and also impairs management, research and education actions. The incorporation of wounds’ photos in the professional practice, stands out as a strategy to assist professionals in the observation, evolution and clear and precise recording. Optimum-Path Forest (OPF) is a framework for the development of pattern recognition techniques based on optimal path partitions and is particularly efficient for image classification. The OPF classifier generates results from the intersection of the selected classes and characteristics. Objective: Describe the steps in developing a mobile application capable of segmenting and sorting complex wound tissue based on the supervised Optimum-Path Forest (OPF). Method: A new intelligent methodology was applied for the analysis and classification of complex wound images using digital image processing and machine learning techniques with the supervised Optimum-Path Forest (OPF) standards classifier. The image bank of 27 complex wounds was created, which were labeled by four specialists according to the classification of the tissues into four classes: granulation (red), fibrinoid (yellow) tissue, necrosis (black) and hematoma (purple), generating 108 images. Two classes were added: white (what is in the photo, except the wound bed) and doubt (divergence in classification by professionals). The OPF classifier was trained from these 108 images. The OPF was applied to the wound images and the accuracy was verified. Then, the application developing process was started. Results and Discussion: The present study developed a computer-aided wound tissue classification scheme for evaluation and management of complex wounds from photos of a smartphone. The OPF application to complex wounds resulted in an accuracy of 77.52 ± 6.14. With this 4 tool, it was developed the product of this research: an application for segmentation, tissue classification and measurement of complex wounds. The application generates a Portable Document Format (PDF) report that can be emailed, printed or attached to a compatible electronic medical record. Conclusion: A bank was made with 27 images of complex wounds, which four professionals labeled for training the OPF classifier, applied the OPF to complex wound images, assessed the accuracy of this process and developed a mobile application with the functions of wound segmentation, tissue classification and wound measurement. The results showed that the value of the accuracy obtained in the computational analysis had a significant value, being equal to the evaluation of specialists in wounds. Comparing to similar studies, the computational analysis of wounds showed less variability than professionals´ evaluation, suggesting that the incorporation of this technology in clinical practice favors the health care of patients with complex wounds, besides providing data for the management, teaching and research.

No / OBJECTIVE: To evaluate the effectiveness of a telehealth system, using digital pen-and-paper technology and a modified smartphone, to remotely monitor and support the effectiveness of wound management in nursing home residents. METHOD: A randomised controlled pilot study was conducted in selected nursing homes in Bradford, which were randomised to either the control or evaluation group. All patients with a wound of any aetiology or severity, resident in the selected nursing homes were considered eligible to participate in the study. Residents in the control homes who had, or developed, a wound during the study period, continued to receive unsupported care directed by the nursing home staff (defined as 'standard care'), while those in the evaluation homes received standard care supported by input from the remote experts. RESULTS: Thirty-nine patients with a wound were identified in the 16 participating Bradford nursing homes. Analysis of individual patient management pathways suggested that the system provided improved patient outcomes and that it may offer cost savings by improving dressing product selection, decreasing inappropriate onward referral and speeding healing. Despite initial anxiety related to the technology most nursing-home staff found the system of value and many were keen to see the trial continue to form part of routine patient management. CONCLUSION: The current study supports the potential value of telemedicine in wound care and indicates the value that such a system may have to nursing-home staff and patients. DECLARATION OF INTEREST: This study was funded by a Regional Innovation Fund grant from the Yorkshire and Humberside Strategic Health Authority. The authors have no conflict of interest to declare with respect to the article or its contents.

The Rho family GTPases Cdc42 and Rac1 are critical regulators of the actin cytoskeleton and are essential for skin and hair function. Wiskott-Aldrich syndrome family proteins act downstream of these GTPases, controlling actin assembly and cytoskeletal reorganization, but their role in epithelial cells has not been characterized in vivo. Here, we used a conditional knockout approach to assess the role of neural Wiskott-Aldrich syndrome protein (N-WASP), the ubiquitously expressed Wiskott-Aldrich syndrome-like (WASL) protein, in mouse skin. We found that N-WASP deficiency in mouse skin led to severe alopecia, epidermal hyperproliferation, and ulceration, without obvious effects on epidermal differentiation and wound healing. Further analysis revealed that the observed alopecia was likely the result of a progressive and ultimately nearly complete block in hair follicle (HF) cycling by 5 months of age. N-WASP deficiency also led to abnormal proliferation of skin progenitor cells, resulting in their depletion over time. Furthermore, N-WASP deficiency in vitro and in vivo correlated with decreased GSK-3beta phosphorylation, decreased nuclear localization of beta-catenin in follicular keratinocytes, and decreased Wnt-dependent transcription. Our results indicate a critical role for N-WASP in skin function and HF cycling and identify a link between N-WASP and Wnt signaling. We therefore propose that N-WASP acts as a positive regulator of beta-catenin-dependent transcription, modulating differentiation of HF progenitor cells.