Alba Hysaj

Department of English

ENGL 21003: Writing for the sciences

Prof. Crystal Rodwell

11/25/2021

 Scientific Report Analysis

Qiu and his co-authors look at how contemporary methods are effective in integrating neuropsychological testing, MRI, and the history of the patient in identifying likely cases and how such methods lack sensitivity coupled with specificity aspects. As reported by the scholars, there exists an “interpretable deep learning strategy that delineates unique Alzheimer’s disease signatures from multimodal inputs of MRI, age, gender, and Mini-Mental State Examination score” ((p.1920). In this scientific report, the researchers report that they designed deep learning (AI) algorithm by analyzing datasets of clinically diagnosed Alzheimer’s patients.  The model they designed predicts the probability of a patient’s Alzheimer’s disease and generates an accurate diagnosis based on patient’s data such as age, gender, or MRI scans. This essay presents a critical rhetorical analysis of the strategies employed by the authors to attain their objective of writing the report. The analysis gives details on how various sections of the scientific report by Qui and his co-authors make meaning to the report’s overall goal with reference to how the various subsections are organized.

In the introduction part, the authors present a general overview of the issue. They make use of ethos to make their presentation of the general overview of the issue clear. The use of ethos is clear in the authors’ decision to refer to peer-reviewed studies and statistics. They use research studies by several scholars such as Nordbeg (2014) to validate their argument that even though there are numerous attempts to establish reliable and efficacious disease-modifying management which have not been successful, there is evidence of significant progress made towards using CSF biomarkers so as to detect Alzheimer’s disease. This convinces the reader that the authors are not the first to invest their time and resources in studying the condition, and as such, there are enough study findings to back their argument. To authors also use ethos to establish their credentials. Though they do not mention this directly, they cite the argument in the research study by McKhann et al. (2011) that the contemporary standards of diagnosis greatly rely on outstandingly professional neurologists to lead an assessment that incorporates the study of patient history, a target intellectual evaluation like bedside Mini-Mental State Examination, abbreviated as, (MMSE) as well as neuropsychological analysis. In this sense, the authors prove their competency and credentials, which further convince the readers that their report is valid as they are well-skilled and can effectively partake in the study and document reliable reports.

The report’s authors understand that to make their argument clear; they need to make good use of pathos in their report. This is clear in their use of imagery. They explain their arguments using visual aids. The authors do not just use images and figure randomly. They use the figure and images to support a particular line of argument. For instance, they use Table 1 and Supplementary Figure 1 to explain their argument that a very strategic model can underscore the rationality and soundness of the deep learning system. They argue that “Association of model predictions with neuropathological findings along with a head-to-head comparison of the model performance with a team of neurologists underscored the validity of the deep learning framework” (Qiu et., 2020, p.1921). The use of figures and images also reflect their effective use of logos to prove their deductive reasoning. They display that they can rely on models, figures, and tables to deduce reasoning accepted by the data findings presented in the figures and tables.

In the methodology section, the authors rely on the use of logos to make inductive reasoning. Logos help authors in taking a specific case to draw generalizations from particular cases. Apart from logos, pathos and ethos are also effectively used in this section to help authors make solid arguments for their materials and data collection methods. At the beginning of the methodology section, the authors refer to previous reputable studies and statistics to prove their argument. They also rely on statistics from ADNI, AIBL, FHS coupled with NACC units or categories to make their argument that AIBL, FHS, and NACC predictions are valid. For instance, the authors say, “Following training and internal testing on the ADNI data, we validated the predictions on AIBL, FHS, and NACC” (Qiu et., 2020, p.1923). In this case, they prove that their argument is backed by the findings on the training and internal testing on the ADNI data.

The authors also use support their arguments with scientific facts from previous studies to make their argument appealing and robust. They understand that their study findings and report depend on reputable and reliable statistics and data findings to appeal to the readers. In line with his thought, the authors present earlier claims as a basis for their argument. They contend that “Since 1976, the FHS expanded to evaluate factors contributing to cognitive decline, dementia, and Alzheimer’s disease” (Qiu et., 2020, p.1922). Through this quote, the authors persuade readers to look at their report based on previous research findings. This shows that their decision to investigate Alzheimer’s disease was based on previous research studies.

Another strategy that has been employed in the report is the use of logical fallacies such as considering the possible contradictory arguments and making clear the reason why such arguments do not hold value based on their research to make explicit their arguments when sources contradict one another. For instance, the authors want to prove their argument that the more extensive idea of illness measure planning with profound learning can be applied in many fields of medication. They also want to prove the fact that “The simple presentation of disease risk as a coherent color map overlaid on traditional imaging modalities aids interpretability” (Qiu et., 2020, p.1930). However, they understand that there are various sources with varying information concerning the same. As such, they acknowledge the argument presented by various sources before making a general conclusion. They acknowledge the past research by arguing that ongoing work has additionally exhibited viable separation of Alzheimer’s sickness together with ordinary comprehension circumstances utilizing a fix grounded inspecting calculation despite being restricted by simultaneous dependence on fluorodeoxyglucose PET, MRI coupled with a model whose information sources believed to be scalar midpoints of powers generated from multi-voxel brain loci. They also contrast their acknowledgment of other sources’ argument by contrary sources presenting data conforming to their argument, such as the saliency planning systems that feature specific pixels dependent on their utility to the inward working of an organization, just as strategies that feature penultimate-layer actuation esteem (Qui et al., 2020). As a result of the comparison of the sources, they conclude that useful physical data is not only preoccupied but also lost. They thus clamor and appeal that their work expands upon such advances by requiring only a solitary imaging methodology on the way to planning a variety of crude pixel esteems to an infection likelihood map that isomorphically safeguards neuroanatomical data.

In conclusion, the authors have made good use of rhetorical strategies to present their argument easier to understand. They have used most of the scientific rhetorical strategies. Their use of rhetorical strategies is effective and help them address the intention of their report.

Reference

Qiu, S., Joshi, P. S., Miller, M. I., Xue, C., Zhou, X., Karjadi, C., … & Kolachalama, V. B. (2020). Development and validation of an interpretable deep learning framework for Alzheimer’s disease classification. Brain, 143(6), 1920-1933.

This entry is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license.