Optimization of Full-Inversion Techniques Towards Clinical Ultrasound Elastography
Bibliographic record
Abstract
Breast cancer is one of the most common cancers, representing 25% of all new cancers and 13% of all cancer related deaths in Canadian women. Early detection before treatment of breast cancer is paramount as survival rates decrease significantly over time. Some of the most common diagnostic and screening procedures include breast manual examination, X-ray mammography, ultrasonography and Magnetic Resonance Imaging (MRI). These methods are either unreliable, associated with dangerous ionization or too costly while they all have difficulty differentiating malignant tumors from benign ones without a follow-up biopsy. One technique that has shown a potential to minimize the number of biopsy cases is ultrasound elastography (USE), which images the breast tissue stiffness that is known to be substantially different for normal and pathological tissue. One of the issues plaguing USE is the lack of data quality due to input tissue displacement data quality and quantity. This data is obtained through processing radiofrequency data acquired at two compression states of the tissue that need to be acquired under the same ultrasound probe orientation to ensure high quality tissue stiffness image. Moreover, there exists no objective and automatic way to assess the quality and consistency of radio-frequency acquired throughout USE. Furthermore, methods capable of producing high quality lateral displacements are limited. As such, part of this research was dedicated to address these issues.. These issues compromise the practical utility of USE in clinical settings. This thesis introduces methodologies to tackle these issues, with the aim of optimizing USE for real time clinical settings, hence allowing reliable breast cancer assessment. It also introduces a series of metrics which can be used to objectively measure data quality. Finally, an open-source software solution was developed to guarantee data quality by generating it in-silico to facilitate the development and assessment of new displacement estimators.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".