This course offers a series of lectures for thesis students and supervisors alike: taking the student throughout their journey from start of topic conceptualization to finalizing conclusions and recommendations. The series is suitable for any stage: whether you are venturing out on a new topic for supervision or are half-way through the thesis.
This course uses examples in the field of Medical Image Processing, yet other Thesis areas and specializations may apply similar strategies emphasized in this course. Medical Image Processing is selected as an exciting field for research students and lecturers interested in algorithm design and development, for detection and classification of structures and anomalies inside the human body that can be captured in an image data set.
With over 15 years of experience in academic Thesis supervision, including medical image processing, and having supervised a range of theses (topics and levels), I have compiled material that offers support, guidance, advice and teachings, using this research specialization as an example.
Overview of Lecture Series: a snapshot of Approach, Research Considerations, Skill Set Requirements and Supervisory; topics of which are to be covered in the 15 Lecture Series.
Lecture 1 Part 1 covers topics including: How to Start Your Thesis, Supervision, Specialization and Topic Selection, Where to Begin and How to do the Research: Writing (output) and Reading (critiquing the research literature), Medical Image Processing - Explained: Overall Approach and Strategies, Terminologies, Considerations and Areas of Research Contribution. Lecture 1 Part 2 continues this lecture topic.
Lecture 1 Part 2 is a continuation of Lecture 1, continuing to look at Medical Image Processing - Explained: Overall Approach and Strategies, Terminologies, Considerations and Areas of Research Contribution. The Lecture has an Accompanying Activity on Getting Started with the Thesis Topic.
A Quick Quiz on the Basics of Getting Started on Your Thesis and What We Covered in Lecture 1 (Parts A and B).
Lecture 2 examines and discusses how to take the Annotated Bibliography from Paper Summaries to Analyses and Critiques: Refining the AB to Articulate the Utility of the Research Relative to Your Topic. Paper Analyses are then Regrouped into Sub-Topic Areas. The Worksheet Supporting this Lecture Assists in this Exercise. An Example is Discussed in this Lecture, showing an Initial Student AB, prior to Refinement.
You will take your Annotated Bibliography and Re-write it, Section by Section, forming your Literature Review. There is a Worksheet and Example that accompanies this lecture.
Advice is given on how to Refine and Improve the Literature Review. This is relative to the Stages embedded within Medical Image Processing for Decision Support. We consider Example Section and Sub-Section Titles, Sequence and advice on Contents. An Example Overview of Structure is provided as well as a Check-List for you as you update and refine your Literature Review.
This is the last lecture in the series on Getting Started. Techniques to improve Academic Writing, Citation and Referencing, and the Chapters of the Thesis are discussed, concluding this Section, prior to the Technical Aspects of Medical Image Processing discussed in the following Section. Writing examples and a Quiz are provided, to conclude this Section of the work.
Complete this Quiz to complete Section 1 out of 3 for this Lecture Series. This Quiz offers a brief review for all previously covered topics. After successfully completing this Quiz you should be able to start Section 2: which focuses on more technical aspects of the medical image processing work.
The first stages of Image Acquisition, Image Registration in the Development Software, followed by Pre-Processing for removal of undesirable data within the image, are discussed. Approaches and strategies for this technical work are considered, with elaboration on some of the more common methods, as well as items to consider when selecting an appropriate strategy, as related to the Thesis Topic and Medical Specialization.
This lecture considers Region of Interest Identification, including determination of Biomarkers or Anatomical Structures of Interest, or Anomalies, within the Image, that are of interest for the thesis. Detection strategies, including common approaches, are considered, examples are given and success criteria are summarized.
Technical considerations and approaches for ROI Segmentation are discussed in this lecture, with examples of methods and results for various Segmentation strategies for separation of the ROI from background information.
Now that you have Detected and Segmented your Region(s) of Interest, you will extract and quantify specific Features, for input to your Classifier (next stage); the output of which provides some Decision Support in line with your Thesis Aims. This lecture examines what is meant by 'Feature Extraction' and considers a variety of evaluation methods for Feature quantification.
Following Feature Set Extraction, Feature Values are Analyzed and Considered for construction of the Classifier: the Output of which determines the Overall System Decision. There are several strategies that can be used for Classification; success criteria relates to Accuracy of algorithm performance in terms of correctness of decision. The Classifier and system as a whole must be validated in terms of its performance.
This is the final lecture in Section 2 and considers the Decision Support System Output in terms of Accuracy, degree of Automation and level of Innovation. Considerations are given toward improving system and sub-system Accuracy, moving towards more Automated sub-systems by reducing human intervention, Optimization for faster system performance, and degree of Innovation. Innovation is important for identification of Research Contributions. Expectations are discussed, in terms of self-expectations, supervisor expectations and reviewer expectations; setting credible, realistic goals toward Thesis Submission.
Review of Section 2: stages from Image Acquisition to Classification and DSS Output Enhancement.
We consider the role and relationship of the Supervisor and Thesis Candidate, to maximize Thesis success. Thesis is a long, complex journey and as such, a Supervisor is critical to Thesis strategy, candidate guidance, candidate motivation and overall submission success.
We consider the most common Thesis Candidate blunders and simple ways of how to avoid them. Blunders are from not saving or documenting work, communication or not addressing the 'So What?' in their Thesis, to supervisor mismatch or lack of appropriate reporting. We also consider Supervisor Blunders such as associated with Industry experts, publications or role expectations.
After the phases of work are complete, the Thesis Document must be updated and finalized. This Lecture - in two parts - describes considerations for full completion of the Thesis Document, including major Chapters of Work, as well as other Sections of the Report, towards Thesis Submission.
Part II of Thesis Final Touches, with Lecture notes and Worksheet.
An Overview of the Thesis Submission Process is provided with guidance as to the Submission approach, what to expect before, during and after submission. The Acceptance process is outlined. Publishing from the Research work is discussed, with considerations provided for Publication Submission. Options for Life After Thesis are described in brief. A final note on this Lecture series is provided at the end of the Lecture.
Final check as you move toward Submission.
I have over 15 years experience in Research and Lecturing, in the UAE (UOWD), Australia and the USA, in addition to Thesis Candidate Supervision in Engineering, Computer Science, Business and Management at PhD, Masters and Bachelors (Honors) levels. I am head of a Research Center in Simulation and Smart Systems Technology, a journal reviewer for Multiple, High-ranking Journals. Research Specialization is varied, in: Medical Engineering, Assistive Device Development, Simulation, Control Systems, Education and Business. I offer UDEMY courses to help students and supervisors, in these areas.