Quantitative kidney ultrasound from macroscale to microscale
Bibliographic record
Abstract
Chronic kidney disease impacts 1 in 10 adults globally, with exponential increases in hospitalization, adverse cardiovascular events and mortality risk. Those who reach end-stage kidney disease require renal replacement therapy, a significant burden on a patient’s quality of life and on the Canadian healthcare budget. Serum and urine biomarkers of kidney disease are insensitive, potentially producing false negative re- sults, whereas tissue biopsy is an invasive procedure that is not routinely performed and comes with complications such as bleeding and infection. There is a need for non-invasive characterization of the kidney. This thesis investigates several methods of kidney tissue characterization, ranging from the macro whole-organ scale to the microstructural scale, using ultrasound imaging methods and machine learning. It presents an open detailed high quality data set for kidney segmentation, with a demonstration of how automatic morphological measurements can be obtained in clinical ultrasound settings using machine learning. This automated measurement is comparable to human experts. It contributes how physics-based data augmentation techniques can improve the robustness of such algorithms, showing that a time-gain compensation augmentation reduces algorithmic uncertainty. It then investigates the speckle properties of transplanted kidneys, showing such properties are patient- and machine-agnostic. This study also identifies the Nakagami distribution as the best model of speckle in the kidney. Subsequent work demonstrates that ultrasound images alone can be used to predict kidney decline in transplant recipients, with speckle parameters being amongst the most prognostic. Finally, quantitative ultra- sound parameters are measured in a murine study. The results show the versatility and accuracy of quantitative ultrasound to characterize the kidney, and enabling a non-invasive method for quantifying kidney disease.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".