Canine Cancer Diagnostics by X-ray Diffraction of Claws
A foundational study on how X-ray diffraction can reveal biological structures.
We report the results of X-ray diffraction (XRD) measurements of the dogs’ claws and show the feasibility of using this approach for early, non-invasive cancer detection. The obtained two-dimensional XRD patterns can be described by Fourier coefficients, which were calculated for the radial and circular (angular) directions. We analyzed these coefficients using the supervised learning algorithm, which implies optimization of the random forest classifier by using samples from the training group and following the calculation of mean cancer probability per patient for the blind dataset. The proposed algorithm achieved a balanced accuracy of 85% and ROC-AUC of 0.91 for a blind group of 68 dogs. The transition from samples to patients additionally improved the ROC-AUC by 10%. The best specificity and sensitivity values for 68 patients were 97.4% and 72.4%, respectively. We also found that the structural parameter (biomarker) most important for the diagnostics is the intermolecular distance.
Overview
The study focuses on developing a new method for early cancer detection in dogs using X-ray diffraction (XRD) patterns from their claws. This research is significant because:
Cancer is the leading cause of death in dogs
The approach offers non-invasive early detection possibilities
Dogs develop similar cancers to humans, making them valuable for comparative oncology
Methodology:
Collected samples from 265 dogs (104 with cancer, 161 without)
Total dataset included 945 XRD patterns from thin slices of dogs' claws
Used Fourier series decomposition for data analysis
Applied Random Forest classifier for machine learning
Analyzed both radial and circular features from XRD patterns
Key Findings:
Best results achieved using only radial features:
97.4% specificity
72.4% sensitivity
84.9% balanced accuracy
Performance exceeds current liquid biopsy methods
Molecular structure distances were more indicative of cancer than structural orientation
Averaging multiple samples per patient improved accuracy (ROC-AUC increased from 0.83 to 0.91)
Technical Details:
Used custom-developed compact X-ray diffractometer
Samples were 100-200 microns thick with 3x3mm area
Two-minute exposure time per sample
Identified distinct features corresponding to 0.51nm and 0.98nm periodicities in keratin structure
Applied various preprocessing steps including calibration and hot pixel removal
Areas for Improvement:
Need for standardized measurement protocols
Larger patient sample size required
Better XRD calibration needed
Potential transition to 2D Fourier series analysis
Higher resolution detectors desired
Better controlled measurement conditions
This research demonstrates a promising new approach to non-invasive cancer detection in dogs using claw XRD patterns, with implications for both veterinary medicine and potential future applications in human cancer diagnostics.