Improving surgical emergency care in Africa: development of an open mobile application for early detection of high-risk patients
Funding Area: Capacity building / Construcción de capacidad
Problem Statement: The proposed work aims to address the pressing problem of high mortality and complications associated with surgical emergencies, particularly in Africa, to promote equitable access and participation in research. The research project will leverage the power of machine learning (ML) techniques, to develop and validate an ML model for early detection of high-risk patients in surgical emergencies. This model will be implemented in an open mobile application for easy use by medical practitioners whatever their location. This work assumes even greater significance within the context of Sub-Saharan Africa, where surgical emergencies pose a significant under prioritized challenge. In Africa, access to surgical care is severely limited, with an estimated 93% of the population lacking access to essential surgical interventions. The proposed research project targets healthcare professionals operating in surgical emergency settings, aiming to enhance and optimize their capabilities through the development of an open mobile application by ML. By enabling early detection of high-risk patients, the project seeks to significantly improve clinical outcomes, reduce complications, and decrease mortality rates associated with surgical emergencies.
Proposed Activities: Activities 1: Identify and collect data on the most relevant variables for early detection of high-risk patients in surgical emergencies. Conduct an extensive literature review of existing studies. (Month 1-3) Draft a research protocol for retrospective study. (Month 3-4) Validate the research protocol for retrospective study. (Month 4-6) Evaluate the status of physical archives. (Month 3) Recruit and train personnel for data collection. (Month 4-6) Collect data on patients and their outcomes in the context of surgical emergencies. (Month 7-8) Activities 2: Apply ML techniques to the collected data to develop a prediction model for high-risk patients in surgical emergencies (internal validity) using R. Recruit developers’ personnel. (Month 3-4) Establish a team (Month 3-4) Hold regular meetings for the analysis plan. (Every month) Perform the analysis with monitoring of the steps. (Month 8-9) Activities 3: Create an open mobile app Develop a mobile application using the R Shiny package. (Month 9-10) Provide training to physicians and medical practitionners on using the model. (Month 10-11) Make the app openly available (Month 11) Monitor usage traffic of the created mobile application to track and analyze its usage patterns(Month 12) Publish scientific article (Month 12)
Openness: The project aims to engage a diverse community, including healthcare professionals, researchers, and stakeholders. The open mobile application developed as part of the project will be made freely available online and offline. This ensures that individuals working in rural areas with limited internet coverage can access and benefit from the application. This approach promotes equitable access to the application's benefits, irrespective of geographical or economic constraints. The project aims to publish scientific articles detailing the process of developing the open mobile application in open access journals. This ensures that the knowledge and insights gained from the project are readily available to the wider scientific community. The publication will serve as a resource for individuals interested in replicating and adapting the project's approach for addressing other diseases, such as obstetrics and pediatrics. The project team will actively participate in scientific congresses and conferences to share their findings, experiences, and lessons learned.
Challenges: Obtaining necessary ethical approval for the project may involve navigating through bureaucratic processes and addressing any concerns related to data privacy and patient confidentiality. Gaining access to physical archives with relevant data may pose logistical challenges, requiring coordination with healthcare institutions and ensuring compliance with data protection regulations. Securing expertise in machine learning (ML) may be challenging, particularly if there is a shortage of skilled professionals in the specific field of ML within the project's context. Collaboration with experts and potential partnerships may be necessary to address this challenge. Availability of timely financial resources: Ensuring the availability of timely financial resources to support the project's activities is crucial. Securing funding for data collection, personnel, equipment, and analysis may require navigating complex funding landscapes and competing priorities. Addressing these challenges will require effective project management, collaboration with relevant stakeholders, proactive engagement with institutions and experts, and securing adequate and timely financial resources. Flexibility and adaptability will be crucial in navigating the complexities of institutional and economic factors to ensure the successful execution of the research project.
Neglectedness: To the best of our knowledge, there are limited sources of funding available for the type of work proposed in this project. Funding in the field of surgical emergencies, particularly in Africa, is often scarce compared to other areas of healthcare, such as infectious diseases or child and maternal health. This discrepancy in funding allocation is rooted in the historical prioritization of certain health issues over others. Surgery, often referred to as the "Cinderella" of public health systems, has not received the same level of attention and financial investment as other health sectors. The funding landscape has predominantly focused on preventive and curative interventions, such as vaccinations, disease control, and primary healthcare services. While these areas are undoubtedly important, the urgency and impact of surgical emergencies on individuals and communities have not received proportional recognition. However, this scarcity of funding should not undermine the significance and urgency of addressing the high mortality and complications associated with surgical emergencies, particularly in Africa. In summary, the limited availability of funding for surgical emergencies, compared to other health areas, has resulted in a neglectedness of this critical field.
Success: The success of the proposed work can be measured through several key indicators. Here's how success can be measured for the proposed work: -Development and functional ML prediction model (internal validity): Successful development of a machine learning (ML) prediction model for early detection of high-risk patients in surgical emergencies. The model should demonstrate internal validity by accurately predicting patient outcomes based on relevant variables and retrospective data. -Implementation of an application mobile: Creation and deployment of an application mobile that incorporates the functional ML prediction model. The application should be user-friendly, reliable, and accessible to healthcare professionals working in surgical emergency settings. -Publications and presentations: Publication of research articles in reputable scientific journals or presentation of findings at conferences. The dissemination of research findings contributes to the scientific community's knowledge and allows for peer review and validation of the work. These indicators of success demonstrate the effectiveness and impact of the proposed work. By achieving these milestones, the proposed work demonstrates its success in addressing the problem of high mortality and complications in surgical emergencies. It contributes to improving patient outcomes, optimizing healthcare practices, and advancing the understanding and application of ML techniques in healthcare.
Total Budget: 9769.1 US$
Budget File: pdf
Affiliations: Gaston Berger University, Saint-Louis Regional hospital
LMIE Carveout: The working location of the team, and our organization is in Senegal, which is classified as a Low-Income Country (LIC) by the World Bank. Senegal is classified as one of the 30 poorest countries according to the World Bank's classification of Low-Income Countries (LICs). The project's community, including the users of the mobile application, primarily resides in rural areas of Senegal where access to healthcare, especially surgical services, is limited. By targeting these underserved communities, our project aligns with the goal of addressing the disproportionate barriers faced by individuals and organizations in LMIEs, ultimately improving surgical emergency care in resource-constrained settings.
Team Skills: -Abdourahmane Ndong: Clinical Surgery and Research Highest Degree and Professional Status: MD/MPH- Clinical Research/Surgeon Dr. Ndong brings extensive experience in clinical surgery and research, providing valuable insights into the challenges and realities of surgical emergencies. His medical background and expertise in surgery contribute to the project's clinical perspective and ensure the relevance and applicability of the proposed work. Adja Coumba Diallo: Institution: MD/MPH, Clinical Surgery, Research, Data Science, Machine Learning Highest Degree and Professional Status: Data Scientist/Doctor of Medicine/Master's in Public Health - Clinical Research/Surgeon Dr. Diallo possesses a unique combination of skills in clinical surgery, research, data science, and machine learning. Her expertise in leveraging data for healthcare applications, strengthens the project's ability to develop and implement effective machine learning models. Her knowledge and skills in data science contribute to the project's analytical approach and data-driven decision-making. Ibrahima Konaté: Research Project Management , Professor of Surgery .Prof. Konaté brings significant experience in research project management, ensuring efficient coordination and implementation of the proposed work. His expertise in managing complex research projects, including resource allocation and team coordination, enhances the project's organizational and operational aspects.
Submission Number: 21
Loading