Harnessing AI and Machine Learning for Accurate Diagnosis of Disorders of Consciousness

Abstract

Sample Disciplinary Discourse Analysis There is no debate that can dismiss the influence technology holds in a person’s daily life. Technology has humanity in a rather pleasant chokehold even while it allows for revolutionary progression in society. From telecommunications capabilities that help connect workers from different countries, to facilitating leisurely pursuits, like enjoying a TV series on a smartphone, technology is clearly the mainstay of many of our activities. It is evident that technology has lifted weights off many shoulders. But how does technology affect the medical field? Technology impacts this industry primarily through the use of artificial intelligence (AI). AI assisted with making medical procedures more proficient, facilitating quicker creation of medicines, enabling more accurate observation of patients’ health, and much more. But what about AI’s predictive capabilities? Many experts are now starting to venture into this question with current technology, more specifically, through machine learning. “Machine learning (ML) is a discipline of artificial intelligence (AI) that provides machines with the ability to automatically learn from data and past experiences while identifying patterns to make predictions with minimal human intervention” (Kanade, 2022). This essay will analyze the writing conventions used in two articles from the Medical Informatics journal titled BMC medical informatics and decision making, vol.23 (1) (2023). These include “Machine learning and network analysis for diagnosis and prediction in disorders of consciousness” (Narayanan et al.) and “MRI-based brain tumor detection using convolutional deep learning methods and chosen machine learning techniques” (Saeedi et al.). First, there will be a summary of the subject matter of each article, followed by an examination of the organization of the article and the evidence the authors use to support their claims. Lastly, there will be an analysis of the writer’s style, tone and language of the respective articles. Narayanan’s article addresses how machine learning and network analysis can promote a more accurate diagnosis for doctors to determine if a patient is suffering from Prolonged Disorders of Consciousness (PDOC). Empirical analysis is heavily used throughout this article to help support her claims. The author and her team members collect relevant statistical data through several experiments. She explains the methods of the experiments and their results through utilizing many figures such as well labelled charts, diagrams, and tables; each accompanied by a brief description. Narayanan organizes the article via the use of headings and sub-headings. She starts with an Abstract which states the aforementioned thesis, methodology of the experiments, results, a conclusion, and the key terms frequently used in the article. The Introduction gives background information on the main objective of the article. The Material and methods section reveals the preparations taken in the conduct of each experiment, and the procedures for the tests. Lastly, under the Results heading there is an in-depth analysis about the outcome of each trial. After analyzing the results of the experiments, the author concludes by stating that further investigations in a clinical setting can be beneficial, and stating the limitations that affected the accuracy of the tests. Regardless, she states that the results revealed that utilizing machine learning and network analysis shows high accuracy with diagnosing patients suffering from PDOC. The author delivered this research in the third person perspective, which is the norm for most academic researchers. She also sticks to a scholarly/formal tone throughout the entire article as she explains the research and formulates necessary hypotheses. Narayanan’s language throughout the article was very advanced. Readers who are not experienced in the field of Medical Informatics would struggle to understand most of the statistical, medical, and technological jargon and figures. However, the use of brief descriptions pertaining to each figure should assist the readers’ understanding. Finally, this article was written in an expository style; moreover, the author utilized complex sentences, and consistently used the passive voice throughout the article. In Sadeei’s article the goal is to determine the accuracy of identifying early stages of brain tumors by utilizing machine learning techniques and other convolutional methods. The structure of this article is identical to the first, showing strong use of empirical analysis to help support any claims. However, this article utilizes a few different sub-headings compared to the first, such as “Contributions to this work” and “Proposed solutions”. The Abstract reveals the thesis, supplementary background information, key terms, methodology for the experiments, results, and a brief conclusion. The Introduction provides insight into the objective of the article. The Materials and methods section reveals what will be used in the tests, the data that will be collected, and the process for the experiments. The Experimental results sections serves to provide a scientific breakdown of each test. Sadeei starts the discussion by defining brain tumors and the annoyance they bring to both patients and doctors. Sadeei explains the necessity of detecting early stages of brain tumors to increase survivability rates of patients. But the traditional means of diagnosing early stages of brain tumor requires surgery to collect a sample of brain tissue (biopsy), which is not only time consuming, but also expensive. Therefore, Sadeei and her team aim to collect enough statistical evidence to prove that with the use of machine learning techniques and other convolutional methods, doctors will be able to detect early stages of brain tumors in suspected patients with high accuracy. Like the first article, Sadeei provides in depth analysis of each test followed by figures such as tables, diagrams, and charts – each accompanied with a brief description. She concludes the article by summarizing her findings and confirms that doctors can be accurate when using these practices to detect early stages of brain tumors. Sadeei’s article is formal like the previous one, ensuring that external references are cited, and other external contributions in the form of referenced sources are noted. Additionally, the writer used the third-person perspective and utilized high-level statistical, medical, and technological terminology, which can in some instances hinder the understanding of the article by average readers. Lastly, this article is shown to be written in an expository style and makes use of complex sentences and strictly uses the passive voice. Medical Informatics has a crucial role for individuals in both medical and technological fields. It serves to provide sufficient statistical evidence to reveal whether new technologies can truly be reliable in medical diagnostics. Both articles addressed this issue via intensive experimentation; also, they both used the same writing style, tone, language, and structure when providing explanations and discussing results/outcomes. Additionally, both writers cited external references and respectfully noted any contributors. Ultimately, the authors were able to determine that machine learning and other technological techniques are highly accurate with medical diagnostics. The information from the articles should increase doctor’s confidence when relying on machine learning techniques during diagnostic tests. References Kanade, V. (2022, August 30). What is machine learning? definition, types, applications, and trends for 2022. Spiceworks. Retrieved March 5, 2023, from https://www.spiceworks.com/tech/artificial-intelligence/articles/what-is-ml/ Narayanan, A., Magee, W. L., & Siegert, R. J. (2023). Machine learning and network analysis for diagnosis and prediction in disorders of consciousness. BMC Medical Informatics and Decision Making, 23(1), 41. https://doi.org/10.1186/s12[phone]8-0 Saeedi, S., Rezayi, S., Keshavarz, H., & R Niakan Kalhori, S. (2023). MRI-based brain tumor detection using convolutional deep learning methods and chosen machine learning techniques. BMC Medical Informatics and Decision Making, 23(1), 16. https://doi.org/10.1186/s12[phone]4-6