Machine Learning Applied to COVID-19: A Review of the Initial Pandemic Period

Diagnostic and decision-making processes in the 2019 Coronavirus treatment have combined new standards using patient chest images, clinical and laboratory data. This work presents a systematic review aimed at studying the Artificial Intelligence (AI) approaches to the patients' diagnosis or evolution with Coronavirus 2019. Five electronic databases were searched, from December 2019 to October 2020, considering the beginning of the pandemic when there was no vaccine influencing the exploration of Artificial Intelligence-based techniques. The first search collected 839 papers. Next, the s were reviewed, and 138 remained after the inclusion/exclusion criteria was performed. After thorough reading and review by a second group of reviewers, 64 met the study objectives. These papers were carefully analyzed to identify the AI techniques used to interpret the images, clinical and laboratory data, considering a distribution regarding two variables: (i) diagnosis or outcome and (ii) the type of data: clinical, laboratory, or imaging (chest computed tomography, chest X-ray, or ultrasound). The data type most used was chest CT scans, followed by chest X-ray. The chest CT scan was the only data type that was used for diagnosis, outcome, or both. A few works combine Clinical and Laboratory data, and the most used laboratory tests were C-reactive protein. AI techniques have been increasingly explored in medical image annotation to overcome the need for specialized manual work. In this context, 25 machine learning (ML) techniques with a highest frequency of usage were identified, ranging from the most classic ones, such as Logistic Regression, to the most current ones, such as those that explore Deep Learning. Most imaging works explored convolutional neural networks (CNN), such as VGG and Resnet. Then transfer learning which stands out among the techniques related to deep learning has the second highest frequency of use. In general, classification tasks adopted two or three datasets. COVID-19 related data is present in all papers, while pneumonia is the most common non-COVID-19 class among them.

Ivermectin as a possible treatment for COVID-19: a review of the 2022 protocols.

Ivermectin is a safe and effective drug in humans and has been approved for use in numerous parasitic infections for over 50 years. In addition, many studies have already shown its antiviral activity. Ivermectin is generally well tolerated, with no indication of central nervous system-associated toxicity at doses up to 10 times the highest FDA-approved dose of 200 µg/kg. The in vitro results of ivermectin for reducing SARS-CoV-2 viral load are promising and show that Ivermectin kills SARS-CoV-2 within 48 hours. A hypothesized mechanism of action for this drug is a likely inhibition of IMPα/ß1-mediated nuclear import of viral proteins as demonstrated for other RNA viruses. However, controlled and randomized studies are needed to prove its effectiveness in COVID-19 in humans. In a single in vivo study with published results, patients confirmed to be infected with SARS-CoV-2 received at least one dose of ivermectin at any time during hospitalization. The use of ivermectin was associated with lower mortality during treatment with COVID-19, especially in patients who required increased inspired oxygen or ventilatory support. Additionally, 81 studies with the clinical use of ivermectin in humans are being carried out worldwide according to ClinicalTrials.gov. However, none of these data has been published so far. However, private and public entities in Brazil have been adopting this drug in their protocols as prophylaxis and in the initial phase of the disease. In addition, ivermectin has been used in mass treatment to prevent onchocerciasis and lymphatic filariasis in sub-Saharan Africa for many years. Surprisingly, this region has the lowest proportional mortality rate among the continents, despite the increasing numbers of infected people released by the World Health Organization.

Analise Dos Registros Da Comunicacao De Sinais E Sintomas De Processos Infecciosos Do Hemonucleo Regional De Araraquara

Introducao: No item 5, do paragrafo 4, do artigo 76, da Secao III, Da Coleta de Sangue do Doador, da Portaria da Consolidacao ndegree5, consta que os doadores serao instruidos para que comuniquem o servico de hemoterapia caso apresentem qualquer sinal e sintoma de processos infecciosos, como febre ou diarreia, ou que tenham tido o diagnostico de alguma doenca infectocontagiosa ate 7 dias apos a doacao. Na Nota Tecnica ndegree5/2020-CGSH/DAET/SAES/MS de 21/02/2020, sobre atualizacao dos criterios tecnicos para triagem de dengue, chikungunya, zica e coronavirus, estendeu-se este prazo de 7 para ate 14 dias apos a doacao. Ate janeiro de 2022, no Hemonucleo Regional de Araraquara (HN), o registro da comunicacao do doador, do resgate dos hemocomponentes em estoque e da busca ativa dos receptores era realizado na Tela do Doador Positivo da Tela da Triagem do Sistema HEMOVIDA. Com a elevacao dos registros da comunicacao no inicio de janeiro 2022 foi necessario criar o Formulario de Rastreabilidade de Sinais e Sintomas de Infeccao apos Doacao. Os comunicados eram recebidos pelas funcionarias da Recepcao via telefone ou WhatsApp, e se necessario, as enfermeiras ou os medicos contatavam os doadores para tirar duvidas sobre o inicio dos sinais e sintomas ou confirmar resultados de exames. Objetivos: Analisar os registros dos comunicados de sinais e sintomas de processos infecciosos, dos resgastes dos hemocomponentes e da busca ativa dos receptores nas Agencias Transfusionais (AT) de 24/01 a 04/07/2022. Material e metodos: Para obter os dados utilizamos a Tela do Servico Social do Sistema Hemovida e os Formularios de Rastreabilidade de Sinais e Sintomas de Infeccao apos Doacao dos periodos de 24/01 a 04 /07/2022. Resultados: No periodo de 24/01 a 04/07/2022, do total de 3130 doadores, 38 (1,2%) entraram em contato com o HN para informar sinais e sintomas de processos infecciosos. Destes, 52,5% eram relacionados a COVID, 39,5% aos sinais e sintomas de infeccao como febre, cefaleia, mialgia e dor de garganta e 8% a dengue. Estes doadores comunicaram o HN, em media, 7 dias apos a data da doacao de sangue. Destes, 2 contatos assintomaticos de casos de COVID estavam negativos para a doenca e os hemocomponentes retornaram para o estoque. Discussao: Neste periodo expurgamos 19 (57,6%) concentrados de hemacias (CH), 10 (34,5%) concentrados de plaquetas (CP) e 32 plasmas frescos congelados (PFC) que estavam no estoque do HN. Dos CH, 14 foram distribuidos para as AT e destes, 7 foram (21,2%) devolvidos para expurgo e 7 (21,2%) foram transfundidos. Dos CP, 19 foram distribuidos e destes, 10 (34,5%) devolvidos para o HN e 9 (31%) transfundidos. Na busca ativa dos 16 receptores das AT, apenas 1 apresentou reacao adversa febril apos transfusao de CP. No total, expurgamos 78 hemocomponentes (83%) e transfundimos 16 (17%). Conclusao: Com as analises do estudo, observamos a importancia do doador nos comunicar estes sinais e sintomas de infeccao o mais rapido possivel para que na busca ativa evitemos a transfusao dos hemocomponentes e possamos acompanhar os receptores envolvidos. Observamos tambem que alguns doadores levaram ate 7 dias apos o aparecimento da febre para comunicar o HN, esperando, talvez, diagnostico definitivo de doenca infectocontagiosa. Para que isto nao ocorra, comecamos a reforcar nas informacoes pos doacao, a importancia de os doadores contatarem o servico logo no inicio dos sinais e sintomas do processo infeccioso, principalmente a ocorrencia de febre. Copyright © 2022

Asymptomatic SARS-CoV-2 Infection in Children's Tonsils

Introduction: The SARS-CoV-2 pandemic killed over 6 million people worldwide. Children were described to have predominantly mild or asymptomatic infections and to be less exposed to the virus, at least for the initial variants. In the present study, we describe how SARS-CoV-2 can silently infect tonsils and adenoids in children undergoing adenotonsillectomy. Method(s): In this cross-sectional study we assessed children who underwent adenotonsillectomy between October 2020 and September 2021 in a secondary hospital in Brazil. All the caregivers denied any symptom of acute viral upper airway infection in the month prior to surgery. Briefly, nasal cytobrush (NC), nasal wash (NW) and tonsillar tissue fragments posttonsillectomy were tested by RT-PCR, immunohistochemistry (IHC), in situ immunofluorescence (IF), and flow cytometry. Result(s): A total of 48 children (18 females, median age 5.5 years) were enrolled. None of them had been vaccinated against COVID-19 at the time of surgery. Only 2 had a history of previous COVID-19 diagnosis, 3 and 5 months, respectively, before surgery. SARS-CoV-2 RNA was detected in 25% (12) of patients-20% in palatine tonsils, 16.27% in the adenoids, 10.41% in NC, and 6.25% in NW. IHC labeling showed viral nucleoprotein presence in both adenoids and palatine tonsils, in epithelial surface and lymphoid cells from extrafollicular and follicular regions. In 5 out of 7 patients, in situ IF showed the expression of ACE2 and TMPRSS2 and viral spike protein in the tonsillar tissue. Flow cytometry revealed that SARS-CoV-2 is predominantly observed in CD123+ dendritic cells (10.57% of all tested sites), followed by CD14+ monocytes (6.32%). Conclusion(s): According to these results, the prevalence of SARS-CoV-2 infection seems to be higher than expected and underdiagnosed in children at this age group. Palatine tonsils and adenoids are important sites of infection and may be a reservoir for the virus. Nevertheless, it is still unclear the impact of these results on virus transmission.

Proposal of a technological cluster to support eLearning platform

Due to the SARS-COV-2 pandemic, educational institutions are immediately faced with a new challenge to adapt, forcing the transition from face-to-face teaching to distance learning in a short period. Distance education supported by technology is a challenge for educational institutions based on binomial technology/teaching. This paper presents a proposal for an e-learning technology structure, supported by a cluster of servers capable of responding to the requirements of distance learning based on the premises of High Availability, High Performance, Load Balancing. The beginning of this study consisted of a literature review to find the various existing technologies, a way to combine them and create a system capable of providing the necessary functionalities, and whose performance could host all the users of an institution simultaneously. The implemented system results from this combination of technologies and allows its capacity to be scaled at any moment according to momentary needs. In technological terms, the solution was based on a free Linux distribution, the Ubuntu Server installed inside a cluster of servers with VMware ESXi, and a cluster of database nodes based on Gallera technology. The eLearning platform used in this study was Moodle because it is one of the resources most used by institutions. The aspects of teaching, provision of content and execution of evaluation tests, were explored. With the implementation of the presented scenario, it was possible to guarantee the High Availability and load balancing of the platform and guarantee a high performance of the whole solution. © 2022 IEEE.

IMPACTO DA IMPLANTAÇÃO DO AGENDAMENTO DAS DOAÇÕES DE SANGUE EM PERÍODO DE EPIDEMIA DE COVID NOS INDICADORES DE QUALIDADE DO HEMONÚCLEO REGIONAL DE ARARAQUARA

Introdução: O Hemonúcleo Regional de Araraquara está situado na região central do estado de São Paulo e atende à regional de saúde DRS III com uma população de quase 1 milhão de habitantes. Atualmente, fornecemos hemocomponentes para 13 Agências Transfusionais localizados no município de Araraquara e região. Este serviço atende uma média de 11 mil candidatos à doação ao ano e, quando iniciou a pandemia do COVID-19, após reunião de 16/03/2020 entre os representantes da área de saúde com o Grupo de Gestão da Pandemia do município de Araraquara, teve que se adaptar para o atendimento dos candidatos à doação de sangue. A partir do mês de abril começamos a atender apenas por agendamento telefônico ou por Whats Zap, inicialmente, agendando 10 candidatos por hora e a partir da adequação do horário dos funcionários, até 12 candidatos por hora ou 62 agendamentos ao dia. Objetivos: Avaliar os indicadores de Inaptidão da Triagem Clínica e Tempo de Permanência após a implantação do agendamento dos candidatos à doação de sangue. Material e métodos: Para obter os dados utilizamos a Tela do Serviço Social do Sistema Hemovida e os indicadores de Inaptidão da Triagem Clínica e Tempo de Permanência dos períodos de 01/04/2019 a 31/03/2020 e 01/04/2020 a 31/03/2021. Resultados: Comparando os períodos 2019/2020 e 2020/2021 observamos que houve uma redução no número de candidatos à doação de 20,4% (11885 para 9455), mas a inaptidão na triagem passou de 29,2% para 25,0%. O indicador de Tempo de Permanência passou da média de 68,1 para 55,9 minutos. Discussão: Com a implantação do agendamento de doadores percebemos, que apesar da redução do número de candidatos à doação, houve uma melhora nos indicadores de inaptidão na triagem e aumento das bolsas coletadas. A meta deste indicador em 2019/2020 era 30% e em 2020/2021 reduzimos para 26%. Neste primeiro período tivemos meses com perda de 28% e, com o agendamento, tivemos mês com apenas 14,7% de inaptidão na triagem. A meta do indicador de tempo de permanência reduziu-se de 80 para 60 minutos. Os motivos como repouso insuficiente, jejum prolongado e uso de medicamentos, que eram as primeiras causas de inaptidão, reduziram em razão da orientação que cada doador recebia das funcionárias da recepção e captação na hora do agendamento. Com o agendamento por hora também evitamos o acúmulo de atendimento em determinados dias e horários o que reduziu o tempo de espera. Houve períodos em que o candidato à doação esperava até 150 minutos para passar por todo processo de doação e hoje, ele leva menos de 60 minutos para realizar todo este processo. Conclusão: Apesar da redução do número de candidatos à doação, percebemos uma perda menor destes na triagem clínica em razão das orientações individuais no momento do agendamento. O tempo de permanência reduziu-se consideravelmente e percebemos a melhora na qualidade do atendimento ao candidato à doação. Com estes resultados, pensamos em manter o atendimento por agendamento após o fim das restrições adotadas pela pandemia de COVID-19, tendo consciência, que num futuro próximo, deveremos nos adequar ao aumento da demanda de hemocomponentes com a liberação das cirurgias eletivas e aumento das emergências.

The impact of COVID-19 on cancer care and oncology clinical research: an experts' perspective.

The coronavirus disease-19 (COVID-19) pandemic promises to have lasting impacts on cancer clinical trials that could lead to faster patient access to new treatments. In this article, an international panel of oncology experts discusses the lasting impacts of the pandemic on oncology clinical trials and proposes solutions for clinical trial stakeholders, with the support of recent data on worldwide clinical trials collected by IQVIA. These lasting impacts and proposed solutions encompass three topic areas. Firstly, acceleration and implementation of new operational approaches to oncology trials with patient-centric, fully decentralized virtual approaches that include remote assessments via telemedicine and remote devices. Geographical differences in the uptake of remote technology, including telemedicine, are discussed in the article, focusing on the impact of the local adoption of new operational approaches. Secondly, innovative clinical trials. The pandemic has highlighted the need for new trial designs that accelerate research and limit risks and burden for patients while driving optimization of clinical trial objectives and endpoints, while testing is being minimized. Areas of considerations for clinical trial stakeholders are discussed in detail. In addition, the COVID-19 pandemic has exposed the underrepresentation of minority groups in clinical trials; the approach for oncology clinical trials to improve generalizability of efficacy and outcomes data is discussed. Thirdly, a new problem-focused collaborative framework between oncology trial stakeholders, including decision makers, to leverage and further accelerate the innovative approaches in clinical research developed during the COVID-19 pandemic. This could shorten timelines for patient access to new treatments by addressing the cultural and technological barriers to adopting new operational approaches and innovative clinical trials. The role of the different stakeholders is described, with the aim of making COVID-19 a catalyst for positive change in oncology clinical research and eventually in cancer care.

PO39 GLOBAL SURVEY OF 104 CANCER PATIENT ORGANISATIONS REVEALS DEVASTATING IMPACT OF COVID-19 Roberta Ventura1

Background: The Global Cancer Coalitions Network (GCCN), established in May 2020, collectively represents over 750 cancer patient organisations representing over 14 million patients around the world. Cancer services have faced challenges as a result of COVID-19, including suspension of screening and diagnostic services;delays in diagnosis leading to higher mortality rates;cancellation/deferral of life-saving treatments;changes in treatment regimens and suspension of vital research. Substantial increases in the number of avoidable cancer deaths are to be expected as a result of diagnostic delays due to the COVID-19 pandemic. Methods: 6 global cancer coalitions surveyed their member organisations in December 2020. Results: Among 104 organisations from 46 countries representing advanced breast, bladder, colorectal, lymphoma, ovarian, and pancreatic cancer patient groups: • Demand for services has increased • 2/3 organisations experienced a fall in income from December 2020, averaging -48% • Over 1 in 10 organisations have closed temporarily, and some permanently • Only 1 in 10 organisations believe their 2021 income will return to levels seen before the pandemic • Almost half report that their ability to operate is under threat • Half do not have access to any national funding schemes to ensure operation during the pandemic • Staff shrunk -20%;volunteer numbers -70% • 20% organisations report normalised cancer services in December 2020;more respondents report services are “worse than ever” • Patient distress, isolation and financial hardship have increased markedly Conclusions: For organisations providing support to cancer patients, declining income, the need to reduce staff and move to virtual working practices has added strain while demand for support due to the pandemic has increased. Emergency support, including funding, must be made available to these organisations to ensure that the needs of cancer patients worldwide continue to be met.

PO40 PATIENTS HELPING PATIENTS: DEVELOPMENT AND EVALUATION OF A VIRTUAL PEER NAVIGATION PROGRAM FOR METASTATIC/ADVANCED BREAST CANCER

Background: Breast cancer is the most frequently diagnosed cancer and the leading cause of cancer deaths in over 100 countries. Although advances have been slow for metastatic/ advanced breast cancer (MBC/ABC), some patients are living longer with this incurable - albeit treatable - disease and experiencing some survivorship issues. Healthcare professionals report a lack of time and understanding of the diverse challenges facing MBC/ABC patients, especially in low- and middle-income countries. Patient support through peer navigation focuses on addressing nonmedical needs while helping patients and families connect with their healthcare providers. Peer navigators share unique insights in the experience of living with MBC/ABC. However, peer navigators need comprehensive training to increase their knowledge of the disease as well as developing the skills to address barriers to optimal care and resources for these patients. Methods: The Peer Navigator Steering Committee developed a culturally mindful evidence-based peer navigator training program. Following pilot evaluation of the content in two low-/ middle-income countries, the program was designed as an online training platform for international accessibility considering the COVID-19 pandemic. The five-module didactic content will be presented through a variety of training methods, including lectures, videos, interactive discussions, and activities (role play). Upon course completion, participants will receive an endorsed certificate of completion. Results: Approximately 100 patient advocates from 5 countries will participant in the first 18 months of the program. Three types of outcome evaluation measures will be reported: (1) participant self-reports of knowledge and skills before and after training, (2) participant evaluations of presenters' knowledge and skills, and (3) participant evaluations of the materials used and the training platform. Conclusion: A virtual online training program is a feasible option to train peer navigators and to achieve global coverage, thereby fully embracing the digital transformation. This program will soon be hosted at the ABC Global Alliance website. Final results will be presented in the poster.

Global survey of 104 cancer patient organisations reveals devastating impact of COVID-19

Background: The Global Cancer Coalitions Network (GCCN), established in May 2020, collectively represents over 750 cancer patient organisations representing over 14 million patients around the world. Cancer services have faced challenges as a result of COVID-19, including suspension of screening and diagnostic services;delays in diagnosis leading to higher mortality rates;cancellation/deferral of life-saving treatments;changes in treatment regimens and suspension of vital research. Substantial increases in the number of avoidable cancer deaths are to be expected as a result of diagnostic delays due to the COVID-19 pandemic. Methods: 6 global cancer coalitions surveyed their member organisations in December 2020. Results: Among 104 organisations from 46 countries representing advanced breast, bladder, colorectal, lymphoma, ovarian, and pancreatic cancer patient groups: · Demand for services has increased · 2/3 organisations experienced a fall in income from December 2020, averaging -48% · Over 1 in 10 organisations have closed temporarily, and some permanently · Only 1 in 10 organisations believe their 2021 income will return to levels seen before the pandemic · Almost half report that their ability to operate is under threat · Half do not have access to any national funding schemes to ensure operation during the pandemic · Staff shrunk -20%;volunteer numbers -70% · <20% organisations report normalised cancer services in December 2020;more respondents report services are “worse than ever” · Patient distress, isolation and financial hardship have increased markedly. Conclusions: For organisations providing support to cancer patients, declining income, the need to reduce staff and move to virtual working practices has added strain while demand for support due to the pandemic has increased. Emergency support, including funding, must be made available to these organisations to ensure that the needs of cancer patients worldwide continue to be met. Legal entity responsible for the study: The authors. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest.

Cancer care during COVID-19: Data from 157 patient organizations

Background: Representatives from 8 global cancer coalitions/alliances, representing 650 cancer patient groups and the interests of over 14 million patients have come together during the pandemic to review and evaluate the patient-perspective impact. Cancer services have faced challenges as a result of COVID-19, including suspension of screening and diagnostic services;delays in diagnosis leading to higher mortality rates;cancellation/deferral of life-saving treatments;changes in treatment regimens and suspension of vital research. For organisations that provide support to cancer patients, declining income, the need to reduce staff and move to virtual working practices has put extra strain while demand for support due to the pandemic has increased. Methods: 5 coalitions surveyed their member organisations. A number of coalitions consulted their members by individual surveys or consultations. Results: A survey of 157 organisations representing advanced breast, bladder, lymphoma, ovarian and pancreatic cancer patient groups from 56 countries found that 57% experienced an average increase of 44% in patient calls and emails. 45% reported that their future viability may be under threat because of the impact of COVID-19 on income. Examples of good practice were reported where healthcare systems have acted to protect patients and cancer services. These include the introduction of COVID-free centres, separation of cancer patients from those who may have COVID-19, and the introduction of virtual and telemedicine services. Organisations have also introduced new ways of working including virtual psychological support services and appbased support groups. These best practices should form part of a global plan of action for future health crisis. Conclusions: Collaboration between patient advocacy organisations, governments and health services is needed to ensure the ground lost to the COVID-19 pandemic is regained. Action is required to restore cancer services safely and effectively without delay. Additional resources for organisations that support cancer patients are required to ensure that they continue to provide vital services. Finally, a global plan of action for cancer is required to meet the challenges of any future health crisis.

Intelligent epidemiological surveillance in the Brazilian semiarid

Right after the Chinese example in conducting COVID-19 epidemic originated in Wuhan, the readiness to detect and respond by health authorities to local (sometimes global) epidemics has become central lately. Within the idea of health 4.0, information about the individual is essential in supporting public community health policies. This paper presents a proposal for an epidemiological surveillance system applied to arboviruses. Data mining techniques and Machine Learning (ML) are used to design mathematical models for detecting epidemics enhanced by Aedes Aegypti (vector for dengue, chikungunaya, yellow fever and zica). Based on data, it is proposed an adaptive manner to reach better stability on results. A Prove of Concept (PoC) is presented for dengue epidemics detection, a common endemic disease in the semiarid region of Brazil. © 2021 IEEE.

Protocol for the Implementation and Assessment of "MoodUP": A Stepped Care Model Assisted by a Digital Platform to Accelerate Access to Mental Health Care for Cancer Patients Amid the COVID-19 Pandemic

The COVID-19 pandemic has important consequences for the mental health of populations. Patients with cancer, already at risk for poor mental health outcomes, are not expected to be spared from these consequences, prompting the need for health services to improve responsiveness. This article presents the research protocol for an implementation study designed to describe the uptake of a well-studied and recognized system for the treatment of depression and anxiety (Stepped-care) during the specific context of a Pandemic in an oncological site. The system set-up will be assisted by a digital platform (MoodUP), where patients undergoing cancer treatment will be screened for anxiety and depressive symptoms, triaged by severity level and algorithm-matched to recommended interventions. Patients undergoing cancer treatment at a cancer clinic in Portugal will be invited to subscribe to the MoodUP platform where they will complete a self-reported questionnaire (Hospital Anxiety and Depression Scale) to screen their anxiety and depressive symptoms. Data will be instantly collected, and an algorithm will activate severity-matched intervention suggestions, through a case manager that will coordinate care. The specific objectives of this study will be to describe the implementation and acceptability of the care system by patients and staff, the barriers to and facilitators of implementation, the proportion of patients accessing the system and their pathways through the various stepped-care interventions, and patient perceptions regarding the feasibility and appropriateness of the eHealth platform. Moreover, exploratory analyses will be conducted to describe patterns of anxiety and depression symptoms variation across all patients, as well as within sociodemographically, clinically and contextually characterized subgroups, to characterize their care needs and access, as well as to explore for whom the MoodUP care system may be more appropriate. This study is expected to improve processes for collaborative mental healthcare in oncology and accelerate the digitalization of services, towards the improvement of mental healthcare access, and management of high-risk patients, during the COVID-19 pandemic.