Easy GraphQL: CLI Approach for Accelerated Development; Ethical Considerations in the Deployment of Artificial Intelligence for Cancer Diagnosis

The technical paper titled “Easy GraphQL:CLI Approach for Accelerated Development” introduces “easy-gql,” a Node.js-based command line interface tool designed to simplify the integration of GraphQL into applications, specifically focusing on Typescript and PostgreSQL setups. GraphQL, although beneficial for efficient data handling, poses a challenge during initial setup due to its complexity. This tool aims to streamline the setup process by automating initial configuration, project scaffolding, and database management, thus accelerating project initiation and promoting standardized code generation. The paper discusses related works, particularly the impact of “create-react-app” on React’s popularity, highlighting how tools that simplify technology integration can significantly boost adoption and developer preference. “Easy-gql” functions through a modular architecture divided into several core components: CLI Interface, Project Scaffolder, Code Generator, Database Manager, and Configuration Manager. Anticipated benefits of using “easy-gql” include rapid project setup, reducing the initialization from hours to minutes, improved migration efficiency, and enhanced developer productivity through automated code generation. These improvements are expected to lead to cost savings and quicker delivery of features. The conclusion reflects a broader goal in software engineering to simplify complex processes to enhance creativity and development capabilities. Future enhancements for the tool include adding support for GraphQL subscriptions which handle real time data updates. Integrating federated GraphQL to allow for even more modular infrastructure, and automatically converting REST API endpoints into GraphQL schemas. These proposed updates aim to extend the tool’s capabilities and align with upcoming trends in software engineering. The STS research paper called “Ethical Considerations in the Deployment of Artificial Intelligence for Cancer Diagnosis” examines the feasibility of implementing artificial intelligence (AI) in cancer diagnosis. The paper acknowledges the significant growth in AI, particularly in healthcare, where AI promises to revolutionize cancer diagnosis by facilitating earlier and more accurate detection through advanced pattern recognition capabilities. This potential has prompted a surge in market value for AI in healthcare. However, there are challenges in adopting AI in medical settings due to the opaque nature of AI in decision-making processes, data privacy concerns and other potential errors that could have severe implications in medical contexts. The paper methodically explores the state of AI in cancer diagnostics by reviewing literature that covers various AI applications, including image-based diagnosis and analysis of peripheral blood biomarkers. It highlights studies that show AI’s capability to surpass traditional diagnostic methods in accuracy and efficiency, as demonstrated by increased Area Under the Curve (AUC) scores and improved sensitivity and specificity in detection. Economically, AI is shown to significantly reduce diagnostic and treatment costs over time, making care more affordable and accessible. However, the paper juxtaposes these benefits with significant ethical concerns. It discusses the lack of transparency in AI decision-making, which complicates patient understanding and informed consent, and highlights data privacy issues exacerbated by the need for large datasets for AI training. Furthermore, the paper raises concerns about the security of AI systems against adversarial attacks, which manipulate AI data inputs to produce incorrect outcomes. The discussion section evaluates these findings through the bioethical principles, asserting that while AI can enhance cancer diagnosis (beneficence), it also poses risks that could cause harm (non-maleficence), challenge patient autonomy due to its inexplicability, and potentially exacerbate healthcare disparities (justice). These ethical considerations suggest a cautious approach to AI implementation in healthcare, emphasizing the need for improved transparency, data protection, and measures to reduce bias and ensure equitable access. In conclusion, the paper recognizes the transformative potential of AI in cancer diagnostics but advocates for a balanced approach that addresses ethical, privacy, and reliability issues to truly benefit patient care and uphold medical ethics. The paper suggests that while it is technically feasible to implement AI in cancer diagnosis, the ethical landscape necessitates careful consideration and enhanced regulatory frameworks to ensure that AI's integration into healthcare is both responsible and effective. Both the technical project on the “easy-gql” tool and the research paper on AI in cancer diagnosis explore the theme of integrating new and advanced technology to improve efficiency and effectiveness in their respective fields. Each project highlights the potential benefits of technological advancements, whether in simplifying the setup of GraphQL in applications or enhancing the accuracy of cancer diagnoses. However, both also address significant challenges of initial setup or ethical concerns relating to AI decision-making processes. Both emphasize the necessity of balancing innovation with consideration of implementation hurdles and broader implications, advocating for solutions that are not only technologically advanced but also ethically and practically sound.

School of Engineering and Applied Science

Bachelor of Science in Computer Science

Technical Advisor: Briana Morrison

STS Advisor: Joshua Earle

Technical Team Members: Aryan Magoon