Georgia Southern University
On Fβ -Score for Medical Diagnostics Tests of Binary and Multi-Class Diseases: Proposing New Measures of Accuracy
Abstract
dc:description.abstractAccurate differentiation between health states—diseased or non-diseased—is essential in clinical diagnostics. Beyond simple binary classifications, this accuracy becomes even more critical in complex multi-class scenarios where diseases are categorized by stages or subtypes. Optimal cut-off points, or thresholds used to classify test results, are crucial for precise diagnoses. This dissertation introduces the Harmonic Mean of F-score and inverse F-score (HF), a novel metric for a balanced assessment of diagnostic accuracy. HF integrates Specificity (Sp) and Negative Predictive Value (NPV) into the Negative F-score (NFγ), ensuring a comprehensive evaluation of true negatives and negative test reliability. Prioritizing both true positives and true negatives, HF was used in optimal cut-off point estimation under binary and tree/umbrella ordering settings. Simulation results revealed that the HF measure performed well, often surpassing established methods in specific settings. The HF measure and cut-off point selection criterion were applied to real-life data in both binary and multi-class settings under tree ordering, showcasing its ability to provide a balanced evaluation of diagnostic accuracy. The HF measure frequently outperformed traditional metrics particularly in binary settings and performed comparably to the extended Youden index under tree ordering. The HF metric's flexibility, allowing parameter adjustments to accommodate diverse scenarios, enables researchers and clinicians to tailor its emphasis on specific aspects of diagnostic performance depending on the context.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Public Health in Biostatistics (Dr.P.H.)
- Level thesis:degree_level
- Dissertation (restricted to Georgia Southern)
- Discipline thesis:degree_discipline
- Department of Biostatistics, Epidemiology, and Environmental Health Sciences
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Alsharman, Marwan
- Contributors dc:contributor
-
- Divine Wanduku
- Jing Kersey
Subjects
dc:subject × 12- Diagnostic accuracy
- Cut-off point Selection
- Tree ordering
- Machine learning
- F-score
- cancer subtypes
- Analytical, Diagnostic and Therapeutic Techniques and Equipment
- Biostatistics
- Diagnosis
- Other Analytical, Diagnostic and Therapeutic Techniques and Equipment
- Statistical Methodology
- Vital and Health Statistics
Rights
- Licence dc:rights.license
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.georgiasouthern.edu/etd/2839