An interpretable machine learning system for colorectal cancer diagnosis from pathology slides.

Considering the profound transformation affecting pathology practice, we aimed to develop a scalable artificial intelligence (AI) system to diagnose colorectal cancer from whole-slide images (WSI). For this, we propose a deep learning (DL) system that learns from weak labels, a sampling strategy that reduces the number of training samples by a factor of six without compromising performance, an approach to leverage a small subset of fully annotated samples, and a prototype with explainable predictions, active learning features and parallelisation. Noting some problems in the literature, this study is conducted with one of the largest WSI colorectal samples dataset with approximately 10,500 WSIs. Of these samples, 900 are testing samples. Furthermore, the robustness of the proposed method is assessed with two additional external datasets (TCGA and PAIP) and a dataset of samples collected directly from the proposed prototype. Our proposed method predicts, for the patch-based tiles, a class based on the severity of the dysplasia and uses that information to classify the whole slide. It is trained with an interpretable mixed-supervision scheme to leverage the domain knowledge introduced by pathologists through spatial annotations. The mixed-supervision scheme allowed for an intelligent sampling strategy effectively evaluated in several different scenarios without compromising the performance. On the internal dataset, the method shows an accuracy of 93.44% and a sensitivity between positive (low-grade and high-grade dysplasia) and non-neoplastic samples of 0.996. On the external test samples varied with TCGA being the most challenging dataset with an overall accuracy of 84.91% and a sensitivity of 0.996.

A CAD system for automatic dysplasia grading on H&E cervical whole-slide images.

Cervical cancer is the fourth most common female cancer worldwide and the fourth leading cause of cancer-related death in women. Nonetheless, it is also among the most successfully preventable and treatable types of cancer, provided it is early identified and properly managed. As such, the detection of pre-cancerous lesions is crucial. These lesions are detected in the squamous epithelium of the uterine cervix and are graded as low- or high-grade intraepithelial squamous lesions, known as LSIL and HSIL, respectively. Due to their complex nature, this classification can become very subjective. Therefore, the development of machine learning models, particularly directly on whole-slide images (WSI), can assist pathologists in this task. In this work, we propose a weakly-supervised methodology for grading cervical dysplasia, using different levels of training supervision, in an effort to gather a bigger dataset without the need of having all samples fully annotated. The framework comprises an epithelium segmentation step followed by a dysplasia classifier (non-neoplastic, LSIL, HSIL), making the slide assessment completely automatic, without the need for manual identification of epithelial areas. The proposed classification approach achieved a balanced accuracy of 71.07% and sensitivity of 72.18%, at the slide-level testing on 600 independent samples, which are publicly available upon reasonable request.

Quality Control in Digital Pathology: Automatic Fragment Detection and Counting.

Manual assessment of fragments during the pro-cessing of pathology specimens is critical to ensure that the material available for slide analysis matches that captured during grossing without losing valuable material during this process. However, this step is still performed manually, resulting in lost time and delays in making the complete case available for evaluation by the pathologist. To overcome this limitation, we developed an autonomous system that can detect and count the number of fragments contained on each slide. We applied and compared two different methods: conventional machine learning methods and deep convolutional network methods. For conventional machine learning methods, we tested a two-stage approach with a supervised classifier followed by unsupervised hierarchical clustering. In addition, Fast R-CNN and YOLOv5, two state-of-the-art deep learning models for detection, were used and compared. All experiments were performed on a dataset comprising 1276 images of colorec-tal biopsy and polypectomy specimens manually labeled for fragment/set detection. The best results were obtained with the YOLOv5 architecture with a map@0.5 of 0.977 for fragment/set detection.

Tailoring Negative Thermal Expansion via Tunable Induced Strain in La-Fe-Si-Based Multifunctional Material.

Zero thermal expansion (ZTE) composites are typically designed by combining positive thermal expansion (PTE) with negative thermal expansion (NTE) materials acting as compensators and have many diverse applications, including in high-precision instrumentation and biomedical devices. La(Fe1-x,Six)13-based compounds display several remarkable properties, such as giant magnetocaloric effect and very large NTE at room temperature. Both are linked via strong magnetovolume coupling, which leads to sharp magnetic and volume changes occurring simultaneously across first-order phase transitions; the abrupt nature of these changes makes them unsuitable as thermal expansion compensators. To make these materials more useful practically, the mechanisms controlling the temperature over which this transition occurs and the magnitude of contraction need to be controlled. In this work, ball-milling was used to decrease particles and crystallite sizes and increase the strain in LaFe11.9Mn0.27Si1.29Hx alloys. Such size and strain tuning effectively broadened the temperature over which this transition occurs. The material's NTE operational temperature window was expanded, and its peak was suppressed by up to 85%. This work demonstrates that induced strain is the key mechanism controlling these materials' phase transitions. This allows the optimization of their thermal expansion toward room-temperature ZTE applications.

iMIL4PATH: A Semi-Supervised Interpretable Approach for Colorectal Whole-Slide Images.

Colorectal cancer (CRC) diagnosis is based on samples obtained from biopsies, assessed in pathology laboratories. Due to population growth and ageing, as well as better screening programs, the CRC incidence rate has been increasing, leading to a higher workload for pathologists. In this sense, the application of AI for automatic CRC diagnosis, particularly on whole-slide images (WSI), is of utmost relevance, in order to assist professionals in case triage and case review. In this work, we propose an interpretable semi-supervised approach to detect lesions in colorectal biopsies with high sensitivity, based on multiple-instance learning and feature aggregation methods. The model was developed on an extended version of the recent, publicly available CRC dataset (the CRC+ dataset with 4433 WSI), using 3424 slides for training and 1009 slides for evaluation. The proposed method attained 90.19% classification ACC, 98.8% sensitivity, 85.7% specificity, and a quadratic weighted kappa of 0.888 at slide-based evaluation. Its generalisation capabilities are also studied on two publicly available external datasets.

Digital Pathology Implementation in Private Practice: Specific Challenges and Opportunities.

Digital pathology (DP) is being deployed in many pathology laboratories, but most reported experiences refer to public health facilities. In this paper, we report our experience in DP transition at a high-volume private laboratory, addressing the main challenges in DP implementation in a private practice setting and how to overcome these issues. We started our implementation in 2020 and we are currently scanning 100% of our histology cases. Pre-existing sample tracking infrastructure facilitated this process. We are currently using two high-capacity scanners (Aperio GT450DX) to digitize all histology slides at 40×. Aperio eSlide Manager WebViewer viewing software is bidirectionally linked with the laboratory information system. Scanning error rate, during the test phase, was 2.1% (errors detected by the scanners) and 3.5% (manual quality control). Pre-scanning phase optimizations and vendor feedback and collaboration were crucial to improve WSI quality and are ongoing processes. Regarding pathologists' validation, we followed the Royal College of Pathologists recommendations for DP implementation (adapted to our practice). Although private sector implementation of DP is not without its challenges, it will ultimately benefit from DP safety and quality-associated features. Furthermore, DP deployment lays the foundation for artificial intelligence tools integration, which will ultimately contribute to improving patient care.

The Potential of Visible Spectroscopy as a Tool for the In-Line Monitoring of Lignin Methylolation.

Out of the 50 to 70 million tons of lignin that are produced annually, only 1 to 2% are used for value-added products. Currently, 90% of the total market of this compound corresponds to lignosulphonates (LS). The most successful industrial attempts to use lignin for wood adhesives rely on using it as a partial substitute in phenol-formaldehyde or urea-formaldehyde resins. However, lignin's aromatic ring presents a low number of reactive sites. Several methods have been proposed to improve its reactivity, such as prior methylolation with formaldehyde. Off-line methods are commonly applied to monitor this reaction's progress, but this introduces a significant delay in the analysis. This study proposes a new method for in-line monitoring of the methylolation reaction using visible spectroscopy. In order to monitor the reaction progress, principal component analysis was applied to the spectra, and the obtained scores were analyzed. When these results were plotted against those obtained by the off-line methods, a satisfactory regression was obtained at 50 °C (R2 = 0.97) and 60 °C (R2 = 0.98) for two different LS samples. Therefore, it was concluded that visible spectroscopy is a promising technique for studying lignin methylolation.

Lignosulphonates as an Alternative to Non-Renewable Binders in Wood-Based Materials.

Lignin is a widely abundant renewable source of phenolic compounds. Despite the growing interest on using it as a substitute for its petroleum-based counterparts, only 1 to 2% of the global lignin production is used for obtaining value-added products. Lignosulphonates (LS), derived from the sulphite pulping process, account for 90% of the total market of commercial lignin. The most successful industrial attempts to use lignin for wood adhesives are based on using this polymer as a partial substitute in phenol-formaldehyde or urea-formaldehyde resins. Alternatively, formaldehyde-free adhesives with lignin and lignosulphonates have also been developed with promising results. However, the low number of reactive sites available in lignin's aromatic ring and high polydispersity have hindered its application in resin synthesis. Currently, finding suitable crosslinkers for LS and decreasing the long pressing time associated with lignin adhesives remains a challenge. Thus, several methods have been proposed to improve the reactivity of lignin molecules. In this paper, techniques to extract, characterize, as well as improve the reactivity of LS are addressed. The most recent advances in the application of LS in wood adhesives, with and without combination with formaldehyde, are also reviewed.

CAD systems for colorectal cancer from WSI are still not ready for clinical acceptance.

Most oncological cases can be detected by imaging techniques, but diagnosis is based on pathological assessment of tissue samples. In recent years, the pathology field has evolved to a digital era where tissue samples are digitised and evaluated on screen. As a result, digital pathology opened up many research opportunities, allowing the development of more advanced image processing techniques, as well as artificial intelligence (AI) methodologies. Nevertheless, despite colorectal cancer (CRC) being the second deadliest cancer type worldwide, with increasing incidence rates, the application of AI for CRC diagnosis, particularly on whole-slide images (WSI), is still a young field. In this review, we analyse some relevant works published on this particular task and highlight the limitations that hinder the application of these works in clinical practice. We also empirically investigate the feasibility of using weakly annotated datasets to support the development of computer-aided diagnosis systems for CRC from WSI. Our study underscores the need for large datasets in this field and the use of an appropriate learning methodology to gain the most benefit from partially annotated datasets. The CRC WSI dataset used in this study, containing 1,133 colorectal biopsy and polypectomy samples, is available upon reasonable request.

A Criança, a Diabetes e a Escola: Capacitação de Professores e Profissionais não docentes / The Child, Diabetes and the School: Training of Teachers and Non-teaching Professionals

A diabetes tipo 1 é uma das doenças crónicas mais comuns na infância. A permanência na escola compreende um período alargado do dia da criança, pelo que os agentes educativos assumem um importante papel no reconhecimento de situações adversas e consequente atuação em eventos relacionados com a condição de saúde destas crianças. Este relatório de estágio serve para apresentar o projeto "A Criança, a Diabetes e a Escola ­ Capacitação dos professores e profissionais não docentes", que teve como objetivo promover a capacitação destes profissionais na intervenção à criança/jovem com diabetes mellitus tipo 1 no contexto escolar. Foi realizado através da metodologia do planeamento em saúde, tendo sido realizadas sessões de educação para a saúde. Estas capacitaram os participantes com conhecimentos e confiança, tornando-os mais competentes para tomar decisões sobre aspetos relacionados com a gestão da doença, no contexto escolar.

Type 1 diabetes is one of the most common chronic diseases in childhood. The permanence in the school comprehends an extended period of the day of the child, reason why the educational agents assume an important role in the recognition of adverse situations and consequent action in events related to the health condition of these children. This internship report serves to present the project "Children, Diabetes and School - Teacher Training and Non-Teaching Professionals", which aimed to promote the training of these professionals in the intervention of the child / youth with type 1 diabetes mellitus in the context school. It was carried out through the methodology of health planning, and health education sessions were held. These trained the participants with knowledge and confidence, making them more competent to make decisions on aspects related to the management of the disease, in the school context.

Olhar a sexualidade masculina na gravidez: desenvolvimento de competências do EEESMO / Looking at male sexuality during pregnancy: development of EEESMO skills

O presente Relatório de Estágio encontra-se integrado no 7º Curso de Mestrado e Pós-Licenciatura em Saúde Materna e Obstetrícia, da Escola Superior de Enfermagem de Lisboa, o qual reflete o percurso de aprendizagem realizado no Estágio com Relatório, centrando-se no desenvolvimento de competências do Enfermeiro Especialista em Enfermagem de Saúde Materna e Obstetrícia (EEESMO) e de mestre, descrevendo as atividades desenvolvidas. Assim como todo o percurso efetuado para desenvolver a temática por mim proposta, de acordo com o plano de estudo do CMESMO, durante os cuidados de enfermagem especializados abordando a temática da sexualidade masculina na gravidez. Para desenvolvimento das competências, sobre a temática escolhida, foi efetuado uma Scoping Review segundo a metodologia de Institute Joanna Briggs. Os cuidados de Enfermagem foram prestados de acordo como o modelo teórico de Enfermagem de Nancy Roper. Dos dados obtidos pude constatar que existem alterações na sexualidade masculina por medos e dúvidas relacionadas com as relações sexuais na gravidez, desconforto das parceiras; desconhecimento das alterações sexuais femininas durante a gravidez. O companheiro sente necessidade de informação para se adaptar à nova condição de vida, contudo a maioria dos Profissionais de Saúde não se sente muito à vontade em abordar a temática. Os companheiros esperam dos Enfermeiros Especialistas apoio e informação para desenvolver mecanismos de coping, para se readaptarem à sexualidade. Os resultados encontram-se de acordo com a evidência científica encontrada. Esses mesmos resultados permitem-me identificar o conhecimento dos companheiros sobre as mudanças sexuais femininas durante a gravidez, as alterações sexuais sentidas pelos companheiros, e as necessidades sentidas de forma a desenvolver mecanismos de coping para as ultrapassar, permitindo, assim, a criação de programas de intervenção precoce, devidamente contextualizados e avaliados pelos enfermeiros especialistas

The present Internship Report is integrated in the 7th Master and Post- -Graduate Course in Maternal Health and Obstetrics of Lisbon Higher School of Nursing, which reflects the learning path carried out in the Internship with Report, focusing on the development of the skills of Specialist Nurse in Maternal Health and Obstetrics and Masters, describing the activities carried out. As well as all the path carried out to develop the theme proposed by me, in accordance with the CMESMO study plan, during the specialized nursing care, addressing the theme of male sexuality in pregnancy. A Scoping Review was undertaken, according the methodology of The Joanna Briggs Institute, to develop skills on the chosen theme. Nursing care were provided according the theoretical Nancy Roper Model of Nursing. From the data obtained with the male partners, I could determine that there are changes in male sexuality due to fears and doubts related to sexual intercourse in pregnancy, female partners' discomfort and a lack of knowledge about female sexual changes during the pregnancy. Man feel the need of information to adapt to the new condition of life. However, most of Health Professionals do not feel comfortable in addressing the issue. The male partners expect support and information from the Specialist Nurses to develop mechanisms of coping, allowing them to feel readapted to sexuality. The results are in accordance with the scientific evidence found. The results allow us to identify the partners' knowledge about the female sexual changes during pregnancy, the sexual modifications experienced by the male partners, and other needs in order to develop coping mechanisms to overcome those changes. This enables the creation of precocious intervention programs, duly contextualized and evaluated by the specialist nurses.

Dimensão psicológica da qualidade de vida de estudantes de Medicina / Psychological dimension of quality of life among Medical students

Objetivos: Avaliar a Dimensão Psicológica da Qualidade de Vida (DPQV) dos estudantes de Medicina da Universidade Federal da Bahia, caracterizar a população segundo variáveis sociodemográficas e epidemiológicas, e identificar possíveis fatores associados. Método: Estudo transversal baseado em questionário autoaplicável com questões da DPQV propostas pela OMS, aplicado aos estudantes do primeiro ao oitavo semestre. Resultados: Os fatores relevantes de aumento da prevalência de baixa DPQV foram: estar cursando o segundo ano; ser do sexo feminino; ser casado(a); morar com os pais versus morar sozinho ou com outras pessoas; renda familiar menor que 15 salários mínimos ou não saber a renda familiar versus renda de mais de 15 salários mínimos; ter escolhido cursar Medicina por motivos não relacionados à aptidão; não dispor de dinheiro suficiente para as necessidades cotidianas; não ter atividades extra-acadêmicas prazerosas; insatisfação com o custo das refeições, a qualidade dos livros na biblioteca e a segurança no ambiente da faculdade; alimentação não saudável; ingerir álcool versus nenhuma ingestão; e fazer ou ter feito uso de drogas ilícitas. Conclusão: Os resultados sugerem influência do curso de Medicina na DPQV.

Objective: To evaluate the Psychological Dimension of Quality of Life (PDQOL) of medical students at the Federal University of Bahia, examining the sample population in terms of sociodemographic and epidemiological variables and identifying associated factors. Method: A cross-sectional study based on a self-administered questionnaire incorporating questions of PDQOL proposed by the WHO, applied to 1st to 8th semester students. Results: The relevant factors identified for the increased prevalence of low PDQOL were: being a second year medical student, female, married, living with parents versus living alone or with others; having a family income of less than 15 minimum wages or not knowing the family income versus an income of more than 15 minimum wages; having chosen medicine for reasons not related to ability; not having enough money for everyday needs; not practicing any pleasurable extracurricular activities; being dissatisfied with the cost of meals, the quality of the books in the library and safety on campus; unhealthy diet; drinking alcohol versus being teetotal; and using or having used illicit drugs. Conclusion: The results suggest that medical school does have an influence on PDQOL.

[Urinary tract infections characterization in a spinal cord injury rehabilitation unit]. / Caracterização das infecções urinárias numa unidade de lesões medulares.

INTRODUCTION: Urinary tract infection (UTI) is a common problem among patients with spinal cord injury (SCI) accounting for 50% of nosocomial infections in some international studies. There are several risk identified like: increased residual urine and bladder pressure, bladder overfilled and catheter use. Protocols for UTI's diagnosis and management, used in our Unit, are based on criteria for bladder dysfunction of the National Institute of Disability and Rehabilitation Research (NIDRR). OBJECTIVES: Identify and characterize the bacterial spectrum and susceptibility to antibiotics; determinate the infection frequency according to voiding method and American Spinal Injury Association (ASIA) classification; check the possibility of cross infections between inpatients. METHODS: 158 urine samples were obtained and microbiologically evaluated from 27 inpatients (10 woman and 17 men) during 19 consecutive weeks. Mean age of the patients was 46 years old; 10 could void spontaneously, 9 used indwelling urethral catheters and 8 intermittent catheterization. RESULTS: The most common bacteria were E. coli (49%) and Klebsiella (22%). Most of the agents were more sensitive to Amoxicilin than Fluoroquinolones. We found 24,6% of UTI's in our patients with spinal cord injury, mainly in the ones using a catheterization technique. Patients in indwelling catheterization had a higher number of UTI's than those with intermittent catheters. No definite cross infections could be found. CONCLUSIONS: E. coli continues to be the most frequent bacteria causing this infection. Fluoroquinolones resistance is rising. Amoxicilin/Clavulanic acid is an alternative for immediate empiric treatment. Intermittent catheterization is the best method of bladder management since most of the UTI's were found in patients with indwelling catheters. ASIA D spontaneously voiding patients were the ones with fewer infections. Asymptomatic bacteriúrias should not be treated since we observed a weekly change in microbiological studies of 47% without any therapeutic intervention The small number of UTI's, compared to other SCI Rehabilitation Units, may be due to the strictly fulfilment of the hospital infection preventive program. Moreover, the care given to prompt identification and surveillance of the neurogenic bladder seems to be crucial to prevention and management of UTI's.

[Traumatic spinal cord injury: neurologic and functional recovery]. / Lesão medular traumática: recuperação neurológica e funcional.

STUDY DESIGN: A retrospective study was conducted. OBJECTIVE: Evaluate the neurologic and functional recovery in patients with Traumatic Spinal Cord Injury (TSCI). METHODS: A retrospective study was conducted, including the 93 patients with a diagnosis of TSCI admitted for rehabilitation in the Physical and Rehabilitation Medicine Department, Santo António General Hospital, between January 1993 and December 2002. Demographic and clinical data were collected from the hospital charts. The patients were evaluated at admission for rehabilitation, before discharging, 1 year and 3 years after injury. Neurologic status was registered according to the standards of the American Spinal Injury Association (ASIA) and functional status by the Functional Independence Measure (FIM). RESULTS: The sample population was 87% men, with a mean age of 40 years. Motor vehicle accidents and falls were the commonest mechanism of injury. The majority sustained dorsal and cervical lesions (45% and 42% respectively). The initial management included: intravenous high dose methylprednisolone, administered within eight hours after injury in 54% of the sample; early surgical stabilization, performed on 77%. Incomplete cord injury carried a better prognosis of motor improvement, especially the patients classified as C on the ASIA Impairment Scale (AIS). A significant statistical association was found between the outcomes obtained during the initial rehabilitation, and the ones achieved on the 3 year followup period. The neurologic recovery was not related with the patient's age, injury level, high dose steroid administration or surgery performed in the first 24 hours after injury (p >0.05). The FIM scores demonstrated the favourable functional outcomes achieved, especially during the inpatient rehabilitation. CONCLUSIONS: The results of this study further demonstrate the considerable potential for neurologic recovery after TSCI, emphasizing the importance of the rehabilitation investment continuity, especially for the ones with incomplete cord injury.